<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by SAP Design on Medium]]></title>
        <description><![CDATA[Stories by SAP Design on Medium]]></description>
        <link>https://medium.com/@SAP_design?source=rss-faa705307dff------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*kOQBrJTG26Pps4gVBdUbtA.png</url>
            <title>Stories by SAP Design on Medium</title>
            <link>https://medium.com/@SAP_design?source=rss-faa705307dff------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Mon, 16 Nov 2026 00:47:58 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@SAP_design/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Goodbye Medium — Join Us at Our New Home on SAP.com/Design]]></title>
            <link>https://medium.com/sap-design/goodbye-medium-join-us-at-our-new-home-on-sap-com-design-cdaae02981b7?source=rss-faa705307dff------2</link>
            <guid isPermaLink="false">https://medium.com/p/cdaae02981b7</guid>
            <category><![CDATA[enterprise-design]]></category>
            <category><![CDATA[design]]></category>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[user-experience]]></category>
            <category><![CDATA[sap]]></category>
            <dc:creator><![CDATA[SAP Design]]></dc:creator>
            <pubDate>Tue, 14 Oct 2025 08:25:21 GMT</pubDate>
            <atom:updated>2025-10-21T08:15:14.118Z</atom:updated>
            <content:encoded><![CDATA[<h3>Goodbye Medium — <strong>Join Us at Our New Home on SAP.com/Design</strong></h3><figure><img alt="A screenshot of the Home page for SAP.com/Design" src="https://cdn-images-1.medium.com/max/1024/1*YVLASx0XMjZWCU0PO2L1cA.png" /></figure><p>We moved! After a decade of sharing enterprise design adventures on Medium, SAP Design has relocated. Come tour our new home on <a href="http://sap.com/design"><strong>sap.com/design</strong></a><strong>.</strong></p><h3><strong>What can you expect on our website?</strong></h3><ul><li>Fresh stories and career inspiration for UX designers</li><li>Product insights and strategy from the brilliant minds behind SAP Design</li><li>Design tools, tips and systems to boost accessibility, research acumen and consistency</li><li>Sneak-peeks at SAP’s innovations and the community that brings them to life</li></ul><h3><strong>Pack your bags!</strong></h3><p>Thank you to our many loyal readers. Our journey continues with this vibrant new platform that we hope you’ll explore, share and bookmark. On <a href="http://sap.com/design">SAP.com/design</a>, we’ll continue exploring what it takes to design for better human outcomes <em>together</em>.</p><p>See you there!<br>Your SAP Design Community</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=cdaae02981b7" width="1" height="1" alt=""><hr><p><a href="https://medium.com/sap-design/goodbye-medium-join-us-at-our-new-home-on-sap-com-design-cdaae02981b7">Goodbye Medium — Join Us at Our New Home on SAP.com/Design</a> was originally published in <a href="https://medium.com/sap-design">Experience Matters</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Designing the Intelligent ERP Experience of Tomorrow]]></title>
            <link>https://medium.com/sap-design/designing-the-intelligent-erp-experience-of-tomorrow-b2e88e6f74d7?source=rss-faa705307dff------2</link>
            <guid isPermaLink="false">https://medium.com/p/b2e88e6f74d7</guid>
            <category><![CDATA[erp-software]]></category>
            <category><![CDATA[cloud]]></category>
            <category><![CDATA[user-experience]]></category>
            <category><![CDATA[human-centered]]></category>
            <category><![CDATA[transformation]]></category>
            <dc:creator><![CDATA[SAP Design]]></dc:creator>
            <pubDate>Thu, 14 Nov 2024 14:40:24 GMT</pubDate>
            <atom:updated>2024-11-14T14:40:24.324Z</atom:updated>
            <content:encoded><![CDATA[<h4>Transforming from rigid infrastructure to dynamic intelligence, ERP systems are now redefining the workplace — empowering users and enabling businesses to thrive in a data-driven future</h4><p><em>Written by Malini Leveque, VP of Business Insights &amp; Advanced Research at SAP</em></p><figure><img alt="The image shows the progression of ERP systems from traditional business process management in the 1990s to becoming “Intelligent Enterprises” by 2025 and beyond. Future ERP systems are expected to evolve from being technology disruptors to business necessities." src="https://cdn-images-1.medium.com/max/930/1*uqm9xu32mzbSU4jyIjapuQ.png" /></figure><p>As enterprise technology rapidly evolves, ERP software is transforming from traditional back-office tools into intelligent, user-centered platforms. Once seen as rigid infrastructure, enterprise software is now prioritizing adaptability, efficiency, and user empowerment, offering seamless, consumer-grade experiences. Advances in cloud technology, machine learning, and AI have turned ERP systems into agile, intuitive tools that redefine how businesses interact with core processes.<em> </em>This evolution represents a reimagining of enterprise software as a productivity partner and catalyst for a more fulfilling work experience. The next frontier is here, placing user experience at the core of enterprise solutions.</p><h3><strong>From Complex Systems to Empowering Experiences</strong></h3><p>Rewind to the 90s, when enterprise software was created to streamline back-office functions like accounting, HR, and logistics. These systems were complex, requiring specialized training and focused on efficiency over ease of use. Fast forward to 2024, and ERP has transformed. Today’s intelligent ERP (iERP) systems, powered by cloud technology, machine learning, and AI, deliver real-time insights that enable informed decisions with just a few clicks.</p><p>Modern ERP suites have evolved from static data repositories to dynamic, adaptable tools that empower users. This shift makes UX essential, not optional. Employees now expect enterprise software to adapt to their roles and workflows, simplifying tasks and boosting productivity and satisfaction.</p><p>Enterprise software is now seen as strategic partners, integrating critical business functions and helping organizations streamline operations and gain actionable insights. With UX at the forefront, intelligent ERP fosters productivity, enhances morale, and promotes a culture of data-driven decision-making — establishing itself as a core asset in digital transformation.</p><h3><strong>The Evolution of UX: Designing for Intuition and Engagement</strong></h3><p>The evolution of UX in intelligent ERP systems has gone beyond integrating Generative AI and large language models (LLMs). It is about designing an experience that feels natural, engaging, and seamlessly supports users. The new generation of ERP software is moving toward greater personalization, simplicity, and inclusivity.</p><p><strong>Personalization</strong> at scale is a key focus for intelligent enterprise software, offering a highly tailored experience that aligns with each user’s role, preferences, and workflows. Imagine an ERP system that understands your unique work patterns, presenting the most relevant data or tools based on your upcoming tasks, meeting reminders, or project deadlines. This AI-driven personalization results in a workspace that intuitively adapts to the user, making interactions more efficient and meaningful.</p><p><strong>Simplicity meets sophistication</strong> as ERP systems address the long-standing barrier of complexity. By introducing streamlined interfaces, voice-enabled commands, and mobile accessibility, intelligent ERP software simplifies user interactions, removing the cumbersome processes that previously hindered adoption. Advanced tools, like digital assistants, now allow users to interact with enterprise software through natural language, speeding up workflows and making the system more accessible, even for non-technical users.</p><p><strong>Trust and inclusivity</strong> are also becoming foundational aspects of modern ERP systems. Transparent and inclusive design, with explainable AI models, builds trust in automated insights, allowing users to rely on data-driven decisions with confidence. By offering understandable and reliable data interpretations, ERP software empower users to make informed decisions. Inclusivity in design ensures that these systems meet the needs of diverse users, catering to a variety of backgrounds and working styles, ultimately enhancing engagement across the organization.</p><h3><strong>Looking to the Future: What’s Next for Intelligent ERP?</strong></h3><p>In the coming years, ERP systems are set to become more autonomous and adaptive, with capabilities that evolve alongside the needs of users and the business landscape. Imagine an ERP system that not only learns from user interactions but also proactively adjusts to changing requirements. Several forward-looking trends are shaping this next phase of ERP.</p><p>Autonomous decision-making and edge computing are poised to drive the future of ERP. AI-powered ERP systems will be self-adjusting, capable of making decisions on behalf of users. By utilizing edge computing, these systems can process data closer to its source, enabling faster decision-making and reducing reliance on centralized data centers. This setup allows ERP systems to automatically optimize workflows, allocate resources, and handle routine tasks, giving users more time to focus on strategic work.</p><p>Additionally, ERP systems are expected to evolve into true digital co-pilots and agents, collaborating with users in innovative ways. These systems are evolving from merely delivering data to actively assisting users in interpreting, analyzing, and applying insights, transforming into true strategic partners in the workplace. Digital agents, powered by advanced AI, are assisting with routine tasks, responding to complex queries, and providing contextual insights, thus streamlining workflows and enhancing decision-making. For example, SAP’s Joule acts as a digital co-pilot, automating repetitive tasks, providing actionable insights, and assisting with complex decisions, making productivity and decision-making more intuitive across the organization.</p><p>The integration of cognitive computing and large language models (LLMs) is also transforming ERP capabilities. These advancements enable ERP systems to understand context, perform complex language-based tasks, and deliver sophisticated insights. As a result, ERP systems can assist users with predictive analytics, answer intricate queries, and generate automated summaries and recommendations based on vast datasets, further enhancing decision-making and productivity.</p><h3><strong>A Human-Centered Future for ERP</strong></h3><figure><img alt="This image displays different ERP user roles in a progression from “Transactional Operators” to “Tech-Savvy Innovators.” Each role is represented by an icon and includes titles for roles such as “Functional Specialists,” “Strategic Analysts,” “Decision-Making Partners,” and finally, “Tech-Savvy Innovators,” indicating an increasing level of strategic influence and technical expertise." src="https://cdn-images-1.medium.com/max/739/1*_B3MjFF4kUZTxf9jpIByDw.png" /></figure><p>At its core, the transformation of ERP UX is about one thing: creating a user-centered experience that makes work more meaningful, productive, and enjoyable. This shift isn’t merely a technical upgrade; it’s a cultural evolution. It’s about designing intelligent enterprise software that empowers every user, from new hires to seasoned leaders, with intuitive tools that adapt to their needs and elevate their impact.</p><p>The next generation of ERP systems is redefining what’s possible by combining intelligence, adaptability, and personalization to support the modern workplace. As we move forward into this exciting future, it’s clear that ERP is no longer just about managing data. It’s about empowering people. This new era of ERP is more than software — it’s a step toward a workplace where technology in empowers and enables us to reach our highest potential.</p><p><em>Experience matters. Follow our journey as we transform the way we build products for enterprise on </em><a href="http://www.sap.com/design"><em>www.sap.com/design</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b2e88e6f74d7" width="1" height="1" alt=""><hr><p><a href="https://medium.com/sap-design/designing-the-intelligent-erp-experience-of-tomorrow-b2e88e6f74d7">Designing the Intelligent ERP Experience of Tomorrow</a> was originally published in <a href="https://medium.com/sap-design">Experience Matters</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Driving Innovation in Human-Centric AI: SAP and DFKI’s Collaborative Vision]]></title>
            <link>https://medium.com/sap-design/driving-innovation-in-human-centric-ai-sap-and-dfkis-collaborative-vision-753b64bef24e?source=rss-faa705307dff------2</link>
            <guid isPermaLink="false">https://medium.com/p/753b64bef24e</guid>
            <category><![CDATA[human-centric-ai]]></category>
            <category><![CDATA[generative-ai-use-cases]]></category>
            <category><![CDATA[ux]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[research]]></category>
            <dc:creator><![CDATA[SAP Design]]></dc:creator>
            <pubDate>Thu, 14 Nov 2024 14:40:13 GMT</pubDate>
            <atom:updated>2024-11-14T14:40:13.936Z</atom:updated>
            <content:encoded><![CDATA[<h4>SAP and DFKI are redefining human-centric AI, blending empathy, ethics, and innovation to empower individuals and elevate productivity. This groundbreaking partnership unites academia and industry, inspiring a future where AI uplifts humanity’s greatest aspirations</h4><p><em>Written by Malini Leveque and Prof. Dr. Gesche Joost</em></p><figure><img alt="This image shows the logos of DFKI (German Research Center for Artificial Intelligence) and SAP placed side by side, connected by a plus symbol, representing a collaboration. The background features a digital network pattern, symbolizing a focus on AI and technology." src="https://cdn-images-1.medium.com/max/800/1*jPX_2Kb0M7qk9QMP3JwFTQ.jpeg" /></figure><p>As artificial intelligence (AI) continues to shape our lives and workplaces, ensuring that these technologies serve real human needs is critical. AI must be rooted in empathy, usability, and ethics to realize its full potential. This vision fuels SAP’s collaboration with the German Research Center for Artificial Intelligence (DFKI) and their Design Research eXplorations (DRX) team of computational AI designers. Together, we are advancing human-centered AI through interdisciplinary research, focusing on creating solutions that prioritize user needs, elevate organizations, and ultimately, improve lives by bridging best AI practices across disciplines.</p><h3><strong>A Commitment to Ethical and Human-Centered AI</strong></h3><p>Our partnership with DFKI is built on a shared commitment to AI that is both innovative and aligned with human values. SAP, a leader in enterprise software, and DFKI, a pioneer in AI research, bring together complementary expertise. This collaboration draws on diverse perspectives — user experience (UX) design, engineering, and business strategy — to ensure that our AI tools respond effectively to real-world challenges while upholding ethical standards.</p><p>One of our core initiatives, the “LaLaMo” (Large Language Models) project, exemplifies our mission. LaLaMo investigates AI’s role in areas like human resources, goal setting, and performance evaluation. Through these AI-driven tools, we aim to streamline operations while supporting employee growth by fostering reflection, engagement, and alignment with organizational goals.</p><h3><strong>The LaLaMo Project: Designing AI for Real-World Impact</strong></h3><p>The LaLaMo project is central to our work, aiming to enhance human-computer interactions in business environments. In SAP SuccessFactors, we’re leveraging large language models to assist employees with the often complex task of goal-setting. This process integrates both top-down strategic goals from corporate leaders and managers, and bottom-up development goals from employees. LaLaMo’s tools assist users in setting and achieving meaningful goals aligned with both personal aspirations and corporate objectives. With natural language processing (NLP) models developed alongside DFKI, we are making goal-setting more intuitive, impactful, and relevant.</p><p>Our research shows that employees succeed more when they actively participate in defining and refining their goals. By providing an adaptive, user-driven goal-setting experience assisted by LLMs, LaLaMo helps employees see how their work connects to broader objectives, enhancing motivation and productivity. This approach demonstrates how AI can support — not replace — human decision-making, working hand-in-hand to create personalized experiences that foster engaging and meaningful outcomes.</p><h3><strong>Building Trust and Transparency in AI</strong></h3><p>Trust is foundational to effective AI, especially when deployed in workplace settings. SAP and DFKI are dedicated to developing transparent AI systems that empower users with control and understanding in their interactions. A key part of our work is trust calibration, where we design interfaces that clearly communicate the role and limitations of AI. Our goal is to create AI tools and interfaces that offer reliable information and feedback, fostering a safe and predictable environment that helps users recognize when they may be over or under trusting AI tools.</p><p>Our joint research addresses essential questions: How can we prevent both over-reliance on and under-utilization of AI tools? How can we ensure employees understand when AI influences important decisions?</p><p>By building trust through clear design and communication, we aim to create AI that enhances human expertise rather than overshadowing it, avoiding shallow AI integration. AI’s power must be balanced with a human-centered approach to drive adoption of enterprise products and foster long-term engagement by employees and customers. Our ongoing efforts are focused on building user confidence, helping people feel comfortable and informed when working with AI.</p><h3><strong>Bridging Academia and Industry for Meaningful Innovation</strong></h3><p>The partnership between SAP and DFKI underscores the immense value of collaboration between academia and industry. Institutions like DFKI provide a foundation of theoretical research and exploratory rigor, pushing the boundaries of AI’s potential by establishing future standards for human-in-the-loop AI, optimizing the CO₂ footprint of large language models, and creating future UX paradigms that go beyond chatbots for more transparent and impactful interactions with AI. Meanwhile, SAP brings practical insights and real-world applications, translating these advancements into solutions with tangible business impact.</p><p>Our collaboration demonstrates how academia and industry can address complex business-technology challenges by blending rigorous research with practical expertise in customer driven solutions that are safe, reliable, and trustworthy. This model not only drives innovation but also sets new standards for ethical AI development in the industry. By engaging in workshops, pilot testing, and user studies, we continuously refine our AI models and LLMs based on real feedback, ensuring they remain effective and aligned with user needs.</p><p>SAP and DFKI’s collaboration serves as an inspiration for more partnerships across disciplines especially in transformative areas like AI. When academia and industry unite, we foster a culture of shared knowledge that drives ethical, impactful AI and ensures that technology and humanity advance together.</p><h3><strong>Empowering People and Businesses: The Road Ahead</strong></h3><p>As we look to the future, SAP and DFKI remain dedicated to scaling a human-centered approach to AI development. Projects like LaLaMo set a benchmark for responsible, user-focused AI in enterprise software, aiming to empower individuals and organizations by embedding AI to enhance productivity and support growth.</p><p>We envision a future where AI is a trusted partner that drives innovation and prioritizes well-being. This collaboration shows the potential when technology and humanity unite, embedding AI in everyday life as a force for good. Our partnership embodies the transformative power of combining academic research with industry insights to push the boundaries of responsible AI development.</p><h3><strong>Inspiring New Standards for Human-Centric AI</strong></h3><p>Our work with DFKI highlights the importance of interdisciplinary collaboration in shaping the future of AI. This partnership encourages more collaborations between academia and industry in fields where ethical considerations and user-centered design are essential towards creating positive user-driven journeys that provide meaningful and sustained value to end-users. Together, we can drive this endeavor of translating research-to-applied customer-use forward in a way that will unlock new human-AI collaboration potential and ensure that AI innovation remains grounded in values that benefit individuals, organizations, and society as a whole.</p><p>The SAP-DFKI partnership exemplifies how academic rigor and industry pragmatism can harmonize to set a new standard for ethical, impactful AI. By fostering collaborations like this, we can inspire a global approach to technology development — one that harmonizes technological advancement with humanity’s greatest needs and aspirations.</p><p><em>We extend our sincere appreciation to the visionary leaders, with special thanks to Philipp Herzig and Arin Bhowmick for their instrumental roles in bringing this collaboration to life.</em></p><p><em>Experience matters. Follow our journey as we transform the way we build products for enterprise on </em><a href="http://www.sap.com/design"><em>www.sap.com/design</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=753b64bef24e" width="1" height="1" alt=""><hr><p><a href="https://medium.com/sap-design/driving-innovation-in-human-centric-ai-sap-and-dfkis-collaborative-vision-753b64bef24e">Driving Innovation in Human-Centric AI: SAP and DFKI’s Collaborative Vision</a> was originally published in <a href="https://medium.com/sap-design">Experience Matters</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[From Chaos to Clarity: Text Insights with Advanced AI Models]]></title>
            <link>https://medium.com/sap-design/from-chaos-to-clarity-text-insights-with-advanced-ai-models-1a5bdd963c5d?source=rss-faa705307dff------2</link>
            <guid isPermaLink="false">https://medium.com/p/1a5bdd963c5d</guid>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[deep-learning]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[natural-language-process]]></category>
            <dc:creator><![CDATA[SAP Design]]></dc:creator>
            <pubDate>Thu, 17 Oct 2024 15:39:13 GMT</pubDate>
            <atom:updated>2024-10-23T06:43:03.461Z</atom:updated>
            <content:encoded><![CDATA[<h4>Learn how AI text classification models like BERT and GloVe are changing the way SAP analyzes customer feedback, making it easier to understand what customers really think and feel, and improving overall satisfaction</h4><p><em>Written by Prithvi Shetty,</em> <em>Data Scientist at SAP</em></p><figure><img alt="Digital representation of a human face constructed from a network of glowing lines and points, with sound waves emanating towards the face, symbolizing AI, speech recognition, or neural networks." src="https://cdn-images-1.medium.com/max/1024/1*O2r7l6-KhpTqKzjjoILsYA.jpeg" /></figure><p>In today’s digital age, we’re swimming in a sea of text — from tweets and reviews to emails and blogs. But how can businesses turn this chaos into valuable insights? Get to know the superheroes of text classification: <strong>BERT </strong>(Bidirectional Encoder Representations from Transformers<strong>) </strong>and <strong>GloVe </strong>(Global Vectors for Word Representation)!</p><p>These advanced models help companies make sense of vast amounts of text and uncover critical insights. One of the key challenges organizations face is how to effectively analyze and classify text to extract meaningful information. This is where text classification models, powered by advanced deep learning techniques like BERT and GloVe embeddings, come into play.</p><h3><strong>What is a “Text Classification Model”?</strong></h3><p>Think of a text classification model as a super-efficient librarian who quickly sorts through millions of books and places them in the right categories. Whether it’s identifying if a review is positive or negative or detecting spam emails, these models are up to the task. A common application of text classification is sentiment analysis, which helps businesses understand the emotions and opinions expressed in text, allowing them to gauge customer sentiment toward their products, services, and brand.</p><h4>1. <strong>Getting the text data and cleaning it</strong></h4><p>Text data is extracted from survey collection tools such as Qualtrics. Non-alphanumeric characters and any unwanted personal information are removed to clean the dataset.</p><figure><img alt="This image presents a comparison titled “Before cleaning” and “After cleaning.” In the “Before cleaning” section, the Gmail notification is written as: “Hey, you have 10,000 messages in the inbox!” In the “After cleaning” section, the notification is simplified to: “gmail notification hey you have 10000 messages in the inbox” The joke highlights how unnecessary elements (like capitalization, punctuation, and commas) are “cleaned up,” yet the message remains essentially the same." src="https://cdn-images-1.medium.com/max/975/1*fHJlXGmiAyPbs5jpQq-Uig.png" /></figure><h4><strong>2. Tokenization</strong></h4><p>Tokenization involves breaking down the text into individual tokens (words). For example, in the sentence “this is a sample text,” each word (“this,” “sample”) becomes a token.</p><figure><img alt="The image illustrates “Tokenization,” where the sentence “This is a sample text” is split into individual words: “This,” “is,” “a,” “sample,” and “text,” each placed in its own box." src="https://cdn-images-1.medium.com/max/839/1*DGtXOGbCy8FrtW-p6siq3A.png" /></figure><h4>3. Vectorization</h4><p>Since machine learning models only understand numbers, it’s essential to convert these tokens into numerical representations. For example, converting “a cat eats dogs” into a vector of numbers based on word frequency is a form of vectorization.</p><figure><img alt="The image shows a bag-of-words representation for four sentences. Each sentence is mapped to a row in a table with columns representing the words: “the,” “red,” “dog,” “cat,” “eats,” and “food.” A ‘1’ in the table indicates the presence of the word in the sentence, while a ‘0’ indicates its absence. The sentences are: “the red dog” “cat eats dog” “dog eats food” “red cat eats” Each sentence is matched with its corresponding row in the table." src="https://cdn-images-1.medium.com/max/975/1*PfYe-3YHaAa2lMlxuudH4w.png" /></figure><h4><strong>4. Training with a machine learning model and tuning for accuracy/prediction</strong></h4><p>Imagine teaching a 5-year-old child math. The more problems they solve, the more confident they become. Similarly, training a machine learning model involves feeding it data (math problems), making predictions (exams), and evaluating its accuracy (test results). Embedding layers can be thought of as specialized teachers helping the child understand complex concepts and perform better.</p><figure><img alt="A table with the words “Machine learning model”, “data”, “prediction”, “accuracy”, and “embedding layers” on the left side and as juxtaposition “5-year old kid”, “mathematics problems”, “examination”, “test results”, and “specialized teachers” on the right side." src="https://cdn-images-1.medium.com/max/1024/1*ERmEUsvxXh4YT6vWcx1WLA.png" /></figure><h4><strong>Training a model</strong></h4><p>The kid here can be imagined as a machine learning model.</p><h4><strong>Prediction</strong></h4><p>When the kid learns to solve problems on their own, they can successfully solve unseen math problems on their own in an examination/test.</p><h4><strong>Accuracy</strong></h4><p>The scores they get on the examination are the results of the model.</p><h4><strong>Embedding layers</strong></h4><p>To improve the test results or accuracy, we can further add embedding layers to the model training. Embedding layers can be imagined as specialized teachers making things way easier for a student to understand the mathematics concepts and score better in examinations</p><p>We’ll further explore the impact of these different embedding layers on the overall performance.</p><h3><strong>How embedding layers improve model performance</strong></h3><h4><strong>BERT (Bidirectional Encoder Representations from Transformers)</strong></h4><p>BERT is trained on a vast dataset, including Wikipedia and Google’s Books Corpus, providing it with a comprehensive understanding of the English language and global context. BERT’s ability to understand context in both directions (left-to-right and right-to-left) makes it highly effective for text classification tasks.</p><h4><strong>GloVe (Global Vectors for Word Representation)</strong></h4><p>GloVe is an unsupervised learning algorithm that generates vector representations for words. These word vectors capture semantic relationships and can be used in various NLP tasks, including text classification.</p><h4><strong>Model Performance Comparison:</strong> <strong>No Embeddings vs BERT vs. GloVe</strong></h4><p>Think of the same example of a 5-year student trying to learn mathematics, but they might not be super-efficient. The standard model with no embedding layers doesn’t use any advanced techniques to understand the text deeply. It does its job, but there’s room for improvement.</p><h4><strong>Enhanced Model Performance with BERT and GloVe</strong></h4><p>Now, imagine specialized math teachers tutoring the student. This is what happens when we use advanced models like BERT and GloVe. These models are much more sophisticated and can understand text in a more nuanced way and they perform way better in terms of accuracy. Here’s an example accuracy score chart.</p><figure><img alt="A table with two criteria “model” on the left side and “accuracy” on the right side. Under “model” there are other criteria listed such as “no embedding layers”, “GloVe”, and “BERT”, and under “accuracy” the results are listed with “76%”, “86”, and “93%”." src="https://cdn-images-1.medium.com/max/1024/1*VE_s8FtXBoNiORLiydILRQ.png" /></figure><h3><strong>How BERT and GloVe Enhance SAP’s Customer Feedback Analysis</strong></h3><h4>Use Case 1</h4><p><strong>Customer Feedback Analysis:</strong> SAP data science teams use BERT and GloVe to extract nuanced insights from customer reviews and social media comments, enabling the company to understand customer sentiment more accurately.</p><ol><li>Analyzing the sentiment of the users from the feedback in Concur and Cloud ERP to truly understand the voice of the user</li><li>Detecting the top user issues from text in HXM and Cloud ERP</li></ol><figure><img alt="The image lists three feedback themes with user comments: Receipts disappearing — Receipts are missing. User: “All my receipts disappeared from the expense report!” Rental car booking — Can’t reserve outside airports. User: “I can only search for airport rental locations.” Smart matching not working — Receipts don’t match, wasting time. User: “None of my receipts matched up, very time-consuming.” This highlights issues needing attention." src="https://cdn-images-1.medium.com/max/975/1*pWTm_DRt-nzWOR1wfs0cLw.png" /></figure><h4>Use Case 2</h4><p><strong>Customer Support Improvement:</strong> By analyzing customer support tickets and chat transcripts, SAP can identify recurring issues and sentiment trends. For instance, detecting a specific product feature bug from customer complaints helped SAP support teams address the issue promptly, improving customer satisfaction and loyalty.</p><ol><li>Classifying the feedback comments from Concur into multiple classes namely: UX issues, Bugs, Performance and Stability issues and reporting it to the respective team.</li><li>Identifying the top issues from Support tickets in Concur thereby improving faster customer resolution</li></ol><figure><img alt="The image shows a bar chart titled “Percentage of Comments Classified as Likely Bugs, Likely UX Issues, and Likely Performance Issues.” 31% of comments are classified as likely bugs. 12% are categorized as likely UX issues. 5% are classified as likely performance issues. The chart highlights that most user feedback relates to bugs, followed by UX and performance issues." src="https://cdn-images-1.medium.com/max/975/1*xUx98yelWZjn890TUlJvxA.png" /></figure><h4>Use case 3</h4><p><strong>Prioritizing Product Updates:</strong> Insights from sentiment analysis prompted SAP to prioritize compatibility improvements in future updates, addressing key pain points for customers and enhancing overall satisfaction.</p><ol><li>Deduction of top themes from 100k user feedback comments help us understand the key pain points for the user and also save hundreds of manhours.</li></ol><figure><img alt="The image shows a thematic reduction using Scattertext. It highlights terms frequently appearing in bugged and non-bugged sentences: Top left: Terms like “errors,” “issues,” and “bugs” appear frequently in bugged sentences. Bottom right: Terms like “intuitive,” “user-friendly,” and “good” appear frequently in non-bugged sentences. The chart visualizes the difference in language between bug-related and non-bug-related user feedback." src="https://cdn-images-1.medium.com/max/975/1*0JmhHvvTQdrmDEWYzzm9uQ.png" /></figure><h3><strong>Conclusion</strong></h3><p>With BERT and GloVe as strategic allies, the potential to transform vast amounts of unstructured text into clear, actionable insights is not just a possibility — it’s a competitive necessity. These advanced models go beyond mere text processing; they unlock a deeper understanding of customers, enabling businesses to anticipate needs, refine offerings, and stay ahead in an ever-evolving market. By leveraging these powerful tools, organizations can turn the chaos of information into clarity, driving smarter decisions, fostering stronger customer relationships, and ultimately achieving greater success. As these technologies continue to advance, they will enable businesses to reach new heights, transforming challenges into opportunities and insights into impact.</p><p><em>Experience matters. Follow our journey as we transform the way we build products for enterprise on </em><a href="http://www.sap.com/design"><em>www.sap.com/design</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1a5bdd963c5d" width="1" height="1" alt=""><hr><p><a href="https://medium.com/sap-design/from-chaos-to-clarity-text-insights-with-advanced-ai-models-1a5bdd963c5d">From Chaos to Clarity: Text Insights with Advanced AI Models</a> was originally published in <a href="https://medium.com/sap-design">Experience Matters</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The SAP Design System as a Mode for Sustainability]]></title>
            <link>https://medium.com/sap-design/the-sap-design-system-as-a-mode-for-sustainability-3243757c656e?source=rss-faa705307dff------2</link>
            <guid isPermaLink="false">https://medium.com/p/3243757c656e</guid>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[accessibility]]></category>
            <category><![CDATA[design-systems]]></category>
            <category><![CDATA[sustainability]]></category>
            <category><![CDATA[design]]></category>
            <dc:creator><![CDATA[SAP Design]]></dc:creator>
            <pubDate>Thu, 10 Oct 2024 10:35:47 GMT</pubDate>
            <atom:updated>2024-10-16T08:46:06.339Z</atom:updated>
            <content:encoded><![CDATA[<h4>SAP’s commitment to sustainability extends to its design system, proving how smart design can drive efficiency and support global sustainability efforts</h4><p><em>Written by Dr. Marieke Storm, Communications and Content Senior Specialist at SAP Design</em></p><figure><img alt="A person using a tablet in a field of crops, with digital holographic overlays displaying various agricultural data, such as weather conditions and crop health. The scene reflects a fusion of technology and farming, highlighting the use of smart farming solutions in agriculture." src="https://cdn-images-1.medium.com/max/1024/1*WjQKi_jlHAiYEu_rZ_cycg.jpeg" /></figure><p>SAP takes its pledge to build a sustainable world seriously. Some notable examples include <a href="https://www.sap.com/sustainability/climate-action.html">ERP in the eco-conscious organization (sap.com)</a>, <a href="https://www.sap.com/about/customer-stories/coldplay.html">Coldplay’s recent world tour app</a>, <a href="https://news.sap.com/2023/10/nhl-skating-toward-sustainable-future-with-sap/">NHL’s Venue Metrics platform</a>, and, perhaps more surprisingly, the SAP Design System.</p><h3>Sustainable Design as a General Trend</h3><p>Sustainability is a general trend in the industry: After the stage of industrial design and <a href="https://medium.com/p/f5081ed498d8">human-centered design</a>, we have now entered the era of sustainable (UX) design (sometimes called <a href="https://katharinaclasen.com/lifecentereddesign">life-centered design</a>), which also encompasses social and environmental considerations. These considerations include factors such as racial and gender equality, clean energy, and responsible consumption.</p><p>Having micro-level, product-driven, and detailed UX design goals on the one hand, and global sustainable development goals such as the ones mentioned above on the other, might seem eons apart, but in this blog I’m going to show you that this isn’t so. We will see that a good design system is the basis for an effective and long-lasting sustainability policy and implementation. In order to do this, we need to define what we mean by sustainability within the context of UX, as the term now encompasses many different concepts. For us in this blog, sustainability means:</p><ul><li>Saving energy and resources (efficiency)</li><li>Enabling the user to save energy and resources (utility)</li><li>Enabling everyone to use the applications (accessibility and inclusivity).</li></ul><h3>Efficiency</h3><h4>The Pillars of UX Sustainability at SAP</h4><p>For many, sustainable design only starts when entering the actual design phase (for example, <a href="https://joint-research-centre.ec.europa.eu/scientific-activities-z/sustainable-product-policy_en">Principles of Sustainable Design (designorate.com),</a> but at SAP, sustainability starts even earlier: with the right mindset. Although <a href="https://experience.sap.com/fiori-design-web/design-principles/#design-principles">SAP Fiori’s design principles</a> were not defined with sustainability in mind per se, they give a clear direction as to which practices are part of SAP design: SAP Fiori design is to provide a role-based, adaptive, simple, coherent, and delightful user experience. These principles are needed to zoom in towards the users’ needs, and to really deal with a few key points well without getting distracted. They also lead to results such as streamlined user journeys, resulting in direct energy savings.</p><p>These principles also pave the way for the foundation that enables everything else that I’m going to describe during the rest of the article: SAP’s design system.</p><blockquote>“A design system offers a library of visual style, components, and other concerns documented and released by an individual, team or community as code and design tools so that adopting products can be more efficient and cohesive.”</blockquote><blockquote>— Nathan Curtis, in <a href="https://designsystem.university/glossary">Defining Design Systems on Medium</a></blockquote><p>With this quote you can already see that companies using a design system are on the right path as regards sustainability: The general advantage of having a design system is that you can lay down rules and conventions. Whether it is using the same calls to action on buttons or defining a design language, conventions speed up work processes and reduce confusion and errors. This saves a lot of time and energy.</p><h4>Design-Led Development Process</h4><p>Next to the what (the design system), the how (the work process) plays an important part, too. This becomes clear in SAP’s <a href="https://experience.sap.com/fiori-design-web/design-led-development-process-external/">design-led development process</a>, consisting of 5 phases.</p><figure><img alt="An illustration of a process flow featuring three key stages: “Discover,” “Design,” and “Deliver,” represented by connected blue shapes. The flow is cyclical, with dotted lines extending from both ends, leading to “Scale” on the right, indicating growth or expansion after delivery. The image represents an iterative approach to product development or problem-solving." src="https://cdn-images-1.medium.com/max/1024/1*piBLBhcH-cJqbLAuAeh8fg.png" /><figcaption><em>The design-led development process</em></figcaption></figure><p>In the discovery phase, designers find out the end user’s tasks and challenges. On-site visits and extensive user research dominate this phase, and they are continued throughout the whole process and beyond. This way, we ensure that we develop exactly the application that is needed. This process ensures that no costly and laborious re-designs will be necessary, which saves time, energy, and resources.</p><h3>Enablement</h3><p>A design system is only as good as it is implemented. It is not very efficient if nobody uses it. The SAP Design System is an open source system to expand access and enable contribution from diverse users. Its adoption is enabled in three different ways:</p><ul><li>Via design specifications for the people who develop the applications.</li><li>Via reusable elements (codes, illustrations, design tokens, or components) that are made available via UI kits, UI5 Web Components, or libraries.</li><li>Via guidelines, for example the SAP Fiori for Web Design Guidelines, and training material.</li></ul><p>That way, we make sure that everybody has access to it and that thousands of people use it world-wide.</p><h3>Utility</h3><h4>Design Tokens and the UI Theme Designer</h4><p>How can we help our customers save energy and resources when branding their websites? We can do this for instance with design tokens and the UI Theme Designer. Design tokens are design-related label value pairs that document repeating decisions between designers and developers, to ensure a joint language. In this way, over 800 tokens define all aspects of our design system, including colors, typography, spacings, and much more.</p><p>To provide different flavors and accessibility offerings, the SAP Horizon theme family consists of a light and dark theme, called SAP Horizon Morning and SAP Horizon Evening, as well as two High Contrast themes, to be inclusive of all user requirements. This results in each design token offering four values each, for our most recent theme family, on top of eleven additional values each for our former themes.</p><p>While we strive for a headless design system, in which all elements can be adjusted based on design tokens, we do want to ensure scoping in the complexity presented to our users and customers. For this, all our tokens exist in a cascading structure. On the highest level one finds only seven primary colors, our so-called primitive tokens. These cascade down to the next layer of semantic tokens, that have names much closer related to their usage, making them easily readable. Lastly, for the full extent of our design tokens impact, we have the layer of component-specific design tokens.</p><p>Through this, our customers are easily able to make sweeping, cascading changes in just a few clicks (such as changing their brand color), or highly specific component changes (such as changing the buttons border on hover). All of this is made available through our UI Theme Designer, keeping in mind a diverse set of expertise.</p><figure><img alt="A visual comparison of two different design themes for a user interface. The top design uses blue and light gray shades, featuring a blue circle for the company logo and labeled colors: “sapCompany Logo,” “sapBrandColor,” “sapShellColor,” and “sapBackgroundColor.” The bottom design presents a darker theme with a pink square for the company logo and similar labeled color elements, but with different shades. The image shows customizable brand elements for a user interface." src="https://cdn-images-1.medium.com/max/668/1*jPhym8wRO4n_GxfGJIalYA.png" /><figcaption><em>Changing the main brand colors by replacing the default brand and base colors of the theme</em></figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/728/1*3WFcwEOzvrGoID5zPmwrow.png" /><figcaption><em>Applying a detailed brand identity with the UI Theme Designer</em></figcaption></figure><h3>Accessibility and Inclusivity</h3><p>As mentioned above, accessibility and inclusivity are now part of the definition of sustainability instead of being seen as separate items. What sets the SAP Design System apart from many other design systems is that these concepts are built into its core. Here are some examples:</p><h4>High-Contrast Ratio</h4><p>All of SAP’s visual themes have been designed to comply with SAP’s UX Consistency Product Standard, which is based on the international <a href="https://www.w3.org/WAI/standards-guidelines/wcag/">Web Content Accessibility Guidelines</a>. All applications are available in the Horizon visual themes High Contrast Black and High Contrast White, which offer features to support people with low visibility and color blindness. They have a contrast ratio of 7:1 between their different elements.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ZRZHF6rZjF_XARMxfP2UXg.png" /><figcaption><em>High Contrast Black and High Contrast White</em></figcaption></figure><h4>Screen Reader Best Practices</h4><p>Our designers are aware that there are users that access SAP applications via screen. So the designers implement accessibility assets in their designs to make them fully inclusive. These include keyboard and reading focus to logically related UI elements, and keyboard navigation.</p><h4>Multi-Sensory Messaging</h4><p>Important messages shouldn’t only be conveyed via one sensory characteristic, but via at least two: for example, color, shape, size, or position. Messages, for example, aren’t only identifiable by their colors, but also by their icons and text.</p><h4>Accessible Typography</h4><p>The SAP font 72 is continually enhanced to improve legibility: For example, its wide apex results in large counters, which means that letters such as O, P, d and e are easier to read. Also, open terminals, such as the tail ends in l and y, reduce the overall density of words.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/880/1*q2EZi6oH1E4Av8npgmmYjQ.png" /><figcaption><em>Typography example showing the larger counters and open terminals</em></figcaption></figure><h3>Conclusion</h3><p>Sustainable UX mainly means sharing and reusing assets, making processes as efficient as possible, reducing waste of time, resources, and energy, and ensuring that people know about these processes and resources, and have access to them. All of these points above require a specific mindset and framework in place. If, on top of that, you want the users to be able to continue to behave sustainably, you need to enable them to do so, too. Thirdly, there is the aspect of equity: that everybody, no matter their circumstances, should be able to use the products. Making sure that an input field sports certain accessibility features is no big deal. But the consequence that a person can then use the application with a screen reader is. Sometimes one small step means a huge societal shift.</p><p><em>Experience matters. Follow our journey as we transform the way we build products for enterprise on </em><a href="http://www.sap.com/design"><em>www.sap.com/design</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3243757c656e" width="1" height="1" alt=""><hr><p><a href="https://medium.com/sap-design/the-sap-design-system-as-a-mode-for-sustainability-3243757c656e">The SAP Design System as a Mode for Sustainability</a> was originally published in <a href="https://medium.com/sap-design">Experience Matters</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Future of Digital Commerce: A Collaborative Journey of Design and Innovation]]></title>
            <link>https://medium.com/sap-design/the-future-of-digital-commerce-a-collaborative-journey-of-design-and-innovation-157b63b10d69?source=rss-faa705307dff------2</link>
            <guid isPermaLink="false">https://medium.com/p/157b63b10d69</guid>
            <category><![CDATA[ecommerce]]></category>
            <category><![CDATA[ux]]></category>
            <category><![CDATA[innovation]]></category>
            <category><![CDATA[stanford-university]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[SAP Design]]></dc:creator>
            <pubDate>Mon, 23 Sep 2024 15:47:26 GMT</pubDate>
            <atom:updated>2024-10-08T06:35:43.296Z</atom:updated>
            <content:encoded><![CDATA[<h4>At the crossroads of design and technology, “Design Forward: A Collaborative Vision for the Future of Digital Commerce” united visionary minds to redefine the future of ecommerce. Hosted at the Stanford d.school with SAP Design, this event illuminated how AI and design thinking can create meaningful, human-centered digital experiences</h4><p><em>Written by Malini Leveque, VP of Business Insights &amp; Advanced Research at SAP</em></p><figure><img alt="A shopping cart in mid-motion tilts to the right on a soft pink background. Three turquoise shopping bags with white handles float above and in front of the cart, giving the impression of lightness and movement. The scene evokes a sense of digital or online shopping with a minimal, modern aesthetic." src="https://cdn-images-1.medium.com/max/1024/1*lPjitrs3FgGO5emMJOopOA.jpeg" /></figure><p>The event served as a springboard for groundbreaking discussions on the intersection of design, technology, and human connection. Held at Stanford’s d.school, this event provided the perfect environment for bold ideas and collaboration, aiming to reshape the future of ecommerce through AI and cutting-edge design thinking.</p><p>Against the dynamic backdrop of Stanford’s fall quarter, the d.school became a hub of innovation. Attendees dove into hands-on sessions and dialogues that challenged traditional perceptions of digital commerce and its future direction. A key highlight of the event was a vibrant panel discussion, moderated by Justin Ferrell from Stanford’s d.school, featuring insights from industry leaders who shared a collective vision of transformation.</p><p>This panel didn’t just reflect individual expertise — it wove together the key themes of how AI and design are driving the evolution of ecommerce. Kyle Wong, CEO of Pixlee and StartX, emphasized the role of social media as a powerful force in building authentic connections between brands and consumers. In a world where user-generated content is becoming a cornerstone of marketing, Wong described how platforms like Instagram and TikTok allow brands to foster deeper, more meaningful engagements. But these shifts were only the beginning.</p><p>Marc Kirmoyan, Lead Product Manager at SAP Commerce Cloud, expanded the conversation by exploring the rise of hyper-personalization in B2B ecommerce. He painted a vivid picture of digital storefronts that are continuously adapting — real-time data and AI transforming product recommendations and pricing models to meet the specific needs of each client. The vision wasn’t just of smart, responsive platforms, but of a world where the boundaries between businesses and customers blur, driven by dynamic customization.</p><figure><img alt="Quote by Lisa Kay Solomon that reads: ‘The role of design is not just to adapt but to shape and redefine future consumer experiences." src="https://cdn-images-1.medium.com/max/800/1*Ga2WohYQ4sgN7lgiNvVkzg.png" /><figcaption>Quote by Lisa Kay Solomon, Author and Designer in Residence at Stanford’s d.school</figcaption></figure><p>Lisa Kay Solomon, Author and Designer in Residence at Stanford’s d.school, brought the discussion into the future by underscoring the power of human-centered design. For her, design wasn’t just about meeting user needs today but shaping the very experiences that will define tomorrow’s commerce landscape. She argued that the focus should move beyond mere customization toward fostering genuine relationships — creating spaces where consumers feel truly understood and valued.</p><p>Kyle Haskins, Product Lead at SAP, rounded out the dialogue by offering a compelling vision of how design will reshape not only interfaces but the very stories behind them. He envisioned a world where personal assistants do more than perform tasks — they anticipate needs, creating seamless, intuitive experiences that blur the line between user and technology. Effective design, he emphasized, isn’t just functional but emotional, telling stories that resonate deeply and forge lasting connections.</p><p>This rich dialogue was only the beginning. The real magic happened when attendees moved from theory to practice. Breakout groups, centered around Design Thinking methods, transformed visionary ideas into actionable concepts that spanned the spectrum of digital commerce innovation. The <strong>“Evolving the AI Shopping Assistant”</strong> group redefined how AI can intuitively interact with consumers, while the <strong>“Intelligent Catalog”</strong> group reinvented product descriptions through storytelling. Finally, the <strong>“Boundary-Breaking Product Visuals”</strong> team stretched the limits of ecommerce imagery, turning static displays into immersive, narrative-driven visuals.</p><p>The <strong>“Design Forward: Reimagining the Future of Digital Ecommerce”</strong> event showcased the longtime relationship between Stanford’s d.school and SAP , embodying Hasso Plattner’s vision of mutual learning between academia and business to create impactful innovations. Plattner, who revolutionized design thinking and provided the original funding to build Stanford’s d.school, inspired this relationship, which not only imagines the future of digital commerce but actively shapes it. The event demonstrated how the fusion of creative minds and technological expertise can craft transformative solutions that engage, inspire, and elevate digital experiences in the modern age.</p><p><em>A special thanks to Arin Bhowmick, CDO at SAP, Balaji Balasubramanian, SVP and Global Head of Commerce and Industry Cloud at SAP, and Laura McBain, Managing Director at the Stanford d.school, for their invaluable support and sponsorship.</em></p><p><em>Experience matters. Follow our journey as we transform the way we build products for enterprise on </em><a href="http://www.sap.com/design"><em>www.sap.com/design</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=157b63b10d69" width="1" height="1" alt=""><hr><p><a href="https://medium.com/sap-design/the-future-of-digital-commerce-a-collaborative-journey-of-design-and-innovation-157b63b10d69">The Future of Digital Commerce: A Collaborative Journey of Design and Innovation</a> was originally published in <a href="https://medium.com/sap-design">Experience Matters</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Researching Our AI Future with Enterprise Software Users]]></title>
            <link>https://medium.com/sap-design/researching-our-ai-future-with-enterprise-software-users-81a564c4fe8d?source=rss-faa705307dff------2</link>
            <guid isPermaLink="false">https://medium.com/p/81a564c4fe8d</guid>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[user-research]]></category>
            <category><![CDATA[ux]]></category>
            <category><![CDATA[future-of-work]]></category>
            <category><![CDATA[ai-study]]></category>
            <dc:creator><![CDATA[SAP Design]]></dc:creator>
            <pubDate>Mon, 09 Sep 2024 14:46:23 GMT</pubDate>
            <atom:updated>2024-09-09T14:46:58.818Z</atom:updated>
            <content:encoded><![CDATA[<h4>Using the practice of strategic foresight to predict how AI can enhance workforce software for SAP’s global customers</h4><p><em>Written by Halle Kho, Vice President of SAP Design Systems Cross Topics</em></p><figure><img alt="Person viewing a futuristic, translucent screen interface with circular data models and Bloomberg Forecast panel options, set against a cityscape background." src="https://cdn-images-1.medium.com/max/1024/1*ZBXZ5Z4iDP606XnuYvoS-Q.jpeg" /></figure><p>Strategic foresight is crucial in navigating and shaping the future. By anticipating trends and understanding emerging technologies, we can better prepare and adapt to the evolving landscape. In our study that we’ll talk about below, we employed strategic foresight to identify how AI could revolutionize enterprise software. Through our research, we were able to envision potential futures and craft strategies that align with the evolving needs and behaviors of our customers. With foresight, we are not only able to predict changes but also influence them, ensuring that our solutions are not just reactive but proactive and innovative.</p><p>The Strategic Foresight team within the Chief Design Office at SAP had the privilege of dedicating two years to a longitudinal study with a stable participant group of our customers. Our focus was to understand the enterprise software user’s state of mind and behavioral approach to integrating AI into their future working methods. We partnered with a diverse group of participants including Computacenter, Sanctuary Housing Association, Southern California Edison, Tennet, the University of Zurich, and Unicredit. This diversity allowed us to gather a wide range of insights, opinions, and expectations, shaping our educated perspective.</p><h3><strong>Learnings</strong></h3><p>Through our extensive evaluative research, we had many conversations with people who each use workforce software in different ways. One of our main goals was to gather as many pain points from users around their current tools and ways of work. Overwhelmingly, we heard users struggle with an abundance of tools and apps, collaborating with teammates across the globe, workflows and processes, and a lack of customization. These learnings helped us to craft a vision for how AI could be used to not just optimize parts of the workflow for users, but overhaul and transform workforce software into a more dynamic experience for all.</p><h3><strong>Augmented systems: the future of enterprise software</strong></h3><p>Based on our research, we believe that the future of enterprise software lies in AI-powered systems that offer hyper-personalized experiences, effectively bridging the gap between workers and their tasks. These augmented systems will not only streamline workflows but also enhance decision-making by providing real-time insights and proactive assistance.</p><h3><strong>Key features of augmented systems:</strong></h3><ul><li><strong>Thought gesture interaction:</strong> Enabling intuitive and natural user interactions through advanced AI.</li><li><strong>Fluid operating system:</strong> Seamlessly integrating various tools and applications into one cohesive system.</li><li><strong>Proactive project &amp; planning:</strong> Offering AI-driven project management and planning features.</li><li><strong>Actionable business intelligence:</strong> Providing real-time data analysis and insights for informed decision-making.</li><li><strong>Co-creation editing:</strong> Facilitating collaborative work environments with integrated editing tools.</li><li><strong>Live collaboration:</strong> Supporting real-time teamwork with synchronous editing and communication.</li><li><strong>Generative data visualizations:</strong> Automatically generating insightful and customizable visual representations of data.</li></ul><h3><strong>Our vision for augmented systems</strong></h3><p>With the right design approach, we can create a fully integrated system that provides a unified view across tools, eliminating lags, gaps, and lack of standardization. This could revolutionize not just the ERP software industry, but the global workforce. Automating processes will empower employees to focus on priority tasks rather than mundane ones. Enterprise software that is accessible and responsive on any hardware offers workers more flexibility and growth potential. Finally, hyper-personalization will help workers maximize their efficiency, creativity, and productivity.</p><figure><img alt="Diagram of a futuristic work setup featuring a roll-up spatial screen, BCI interfacing earpiece, foldable tablet, and interactive standing desk." src="https://cdn-images-1.medium.com/max/1024/1*YG1uh8I-RKfvZUSrp4AAvA.jpeg" /></figure><h3><strong>The future of work awaits us</strong></h3><p>The integration of AI into enterprise workforce systems is not something to fear. Rather, we should be anticipating all the ways in which we can use this powerful tool to super-charge our methods. There are considerations to keep in mind as we move ahead, such as balancing how tool integration comes together to avoid information overload or cramped interface design. In addition, this new approach is sure to challenge established work patterns, which can pose a threat of rejection before acceptance. But the reality is that AI is already permeating so many parts of our daily lives. We must have the foresight to design the right tools in order to maximize our potential for innovation.</p><p><em>Experience matters. Follow our journey as we transform the way we build products for enterprise on </em><a href="http://www.sap.com/design"><em>www.sap.com/design</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=81a564c4fe8d" width="1" height="1" alt=""><hr><p><a href="https://medium.com/sap-design/researching-our-ai-future-with-enterprise-software-users-81a564c4fe8d">Researching Our AI Future with Enterprise Software Users</a> was originally published in <a href="https://medium.com/sap-design">Experience Matters</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The 72 Font: Enhancing Digital Product Experience and Usability]]></title>
            <link>https://medium.com/sap-design/the-72-font-enhancing-digital-product-experience-and-usability-2fa48eea4a1a?source=rss-faa705307dff------2</link>
            <guid isPermaLink="false">https://medium.com/p/2fa48eea4a1a</guid>
            <category><![CDATA[font-design]]></category>
            <category><![CDATA[corporate]]></category>
            <category><![CDATA[accessibility]]></category>
            <category><![CDATA[design]]></category>
            <category><![CDATA[ux]]></category>
            <dc:creator><![CDATA[SAP Design]]></dc:creator>
            <pubDate>Mon, 09 Sep 2024 14:40:19 GMT</pubDate>
            <atom:updated>2024-09-09T14:55:25.037Z</atom:updated>
            <content:encoded><![CDATA[<h4>72, the proprietary SAP typeface, has been meticulously designed and refined to enhance digital products, accessibility, and brand, ensuring the best possible experience for all users</h4><p>Font is an incredibly important building block for both brands and digital products. Not only does the right font create compelling visuals, but it can help to effectively communicate and resonate with users. This is exactly why SAP chose to design a proprietary font, aptly named “72,” that would elevate the user experience by being digitally native, accessible, and infused with the history of the brand.</p><h3><strong>Designed with a history of innovation</strong></h3><p>In 2015, after coming up against technical constraints in using Arial as a system font, SAP decided to collaborate with typography experts at <a href="https://www.monotype.com/resources/case-studies/creating-the-typeface-for-sap-fiori">Monotype</a> to design a font that would power its vast product suite. The result is the <a href="https://www.red-dot.org/project/72-font-family-for-sap-fiori-16071-16068">award-winning</a> 72 font, designed to be optimized for digital products and built with a foundation of universal design.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fgiphy.com%2Fembed%2FtN3A1fBiciJZYTBKqe%2Ftwitter%2Fiframe&amp;display_name=Giphy&amp;url=https%3A%2F%2Fgiphy.com%2Fgifs%2FtN3A1fBiciJZYTBKqe&amp;image=https%3A%2F%2Fmedia3.giphy.com%2Fmedia%2Fv1.Y2lkPTc5MGI3NjExZHc1djh4Mjlqd2Fzbno0ZHF6ZG92NTh6c25paG5zbGx4dGdrbDdkdyZlcD12MV9pbnRlcm5hbF9naWZfYnlfaWQmY3Q9Zw%2FtN3A1fBiciJZYTBKqe%2Fgiphy.gif&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=giphy" width="435" height="248" frameborder="0" scrolling="no"><a href="https://medium.com/media/00ca78d28eb177d1e388cab41a0ce262/href">https://medium.com/media/00ca78d28eb177d1e388cab41a0ce262/href</a></iframe><p>The name ‘72’ refers to SAP’s founding in 1972. It highlights the innovative spirit that helped transform the scrappy start-up looking to revolutionize real-time computing into the ERP giant it is today. In addition to the hefty technical requirements needed to ensure the font was robust enough to power the digital products, the font was designed to align with key SAP brand characteristics, making it feel trustworthy, friendly, and modern. This is just one way in which SAP shows its commitment to creating software that is both beautiful and user-friendly.</p><h3><strong>Built for the digital world</strong></h3><p>72 is unique in that it was designed to be digitally native, meaning it’s optimized for use across the many mobile and <a href="https://experience.sap.com/fiori-design-web/typography-horizon/">web devices</a> that power the SAP software ecosystem. Through a comprehensive design process, SAP worked with Monotype to meticulously craft manually pixel-aligned font styles, ensuring crisp display even on low-resolution screens.</p><p>Recognizing the need for specialized typographic tools, SAP recently expanded the Font 72 family to include several key styles and features, each designed to address specific aspects of SAP’s UI design and functionality:</p><ul><li><strong>72 Monospaced:</strong> This style was created to give SAP’s classic UI a fresh, modern look, ensuring that all products maintain consistent and cohesive typography. It enhances the user experience by providing a clean and uniform text appearance across applications.</li><li><strong>72 Semi Bold:</strong> Introduced with the Horizon theme, 72 Semi Bold adds variety and improves the look and readability of SAP’s interfaces. This style enhances user engagement by making text stand out without sacrificing readability.</li><li><strong>72 Semi Bold Duplex:</strong> This unique style maintains the character widths of the regular weight while using the semi bold weight. This ensures the clarity of buttons and toolbars, making interactions intuitive and visually appealing. It bridges old and new design elements, improving the overall look and functionality of SAP applications.</li></ul><figure><img alt="This image showcases a visual presentation of different font styles and weights. On the left, text examples include styles such as Light, Regular, Italic, Bold Italic, Semibold, Condensed, and Condensed Bold. The center features a large number “72” repeated in different shades of blue. On the right, additional text shows font weights like Condensed Bold, Mono, Mono Bold, and Black. Geometric shapes and patterns in various blue tones are displayed throughout the design." src="https://cdn-images-1.medium.com/max/1024/1*QwV8bOYsZq8CKXub2uLzHA.png" /></figure><h3><strong>Meeting universal design standards</strong></h3><p>There are more than 200 million users worldwide who interact with SAP products daily. Due to the scale of this user base, SAP has an obligation to make its software as accessible as possible for everyone. Two key areas in which the font underscores the importance of accessible and inclusive design are legibility and multilingual support.</p><p>The font has been fine-tuned to enhance readability and ensure each letter is distinctly recognizable — a critical detail that significantly impacts accessibility. For instance, distinguishing between a lowercase ‘l’ and an uppercase ‘I’ at a glance is effortless with Font 72, thanks to its precise design.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fgiphy.com%2Fembed%2FWXloFiQ1XPODtGKXsb%2Ftwitter%2Fiframe&amp;display_name=Giphy&amp;url=https%3A%2F%2Fgiphy.com%2Fgifs%2FWXloFiQ1XPODtGKXsb&amp;image=https%3A%2F%2Fmedia3.giphy.com%2Fmedia%2Fv1.Y2lkPTc5MGI3NjExYjBnNTdvMDRjenk3N2N4aXg0eWs3dzAwNzgzcWM3Y3VoeXlubm1ndiZlcD12MV9pbnRlcm5hbF9naWZfYnlfaWQmY3Q9Zw%2FWXloFiQ1XPODtGKXsb%2Fgiphy.gif&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=giphy" width="435" height="248" frameborder="0" scrolling="no"><a href="https://medium.com/media/50117aff53676785fb3583fa1abe04a9/href">https://medium.com/media/50117aff53676785fb3583fa1abe04a9/href</a></iframe><p>Additionally, to ensure the best possible experience for users world-wide, the font supports multiple languages and writing systems, and utilizes a fallback font option in the rare cases a font is not available. This expansion, driven by direct user feedback, enhances SAP’s global reach, allowing more users to interact with the software in their preferred language comfortably and seamlessly.</p><h3><strong>Supporting the brand evolution</strong></h3><p>The latest evolution of the 72 font family is 72 Brand. This update not only unites the brand and product experience, but tailors each to the right setting. This commitment to quality typography is mirrored throughout SAP’s digital ecosystem. With the 72 Brand font, SAP unites, yet distinguishes between product interfaces and brand communication strategies. This unique variant of the Font 72 family is designed exclusively for branding, marketing, and non-product digital touchpoints. It includes design elements like angle cuts in certain characters that echo the SAP logo. With 72 and 72 Brand, SAP’s brand and product experiences are finally visually united across critical touchpoints like the marketing website, digital and print brand campaigns, and product interfaces.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fgiphy.com%2Fembed%2FnoQQ4s6cfBEfA1ANpq%2Ftwitter%2Fiframe&amp;display_name=Giphy&amp;url=https%3A%2F%2Fgiphy.com%2Fgifs%2FnoQQ4s6cfBEfA1ANpq&amp;image=https%3A%2F%2Fmedia4.giphy.com%2Fmedia%2Fv1.Y2lkPTc5MGI3NjExbzM3cjR2dmM1a2k3OHEwamxuOXNwdHBnNW1ieHN4N2gyMHFqdnJucCZlcD12MV9pbnRlcm5hbF9naWZfYnlfaWQmY3Q9Zw%2FnoQQ4s6cfBEfA1ANpq%2Fgiphy.gif&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=giphy" width="435" height="244" frameborder="0" scrolling="no"><a href="https://medium.com/media/1c1e72914efb7bbe46e0506732769335/href">https://medium.com/media/1c1e72914efb7bbe46e0506732769335/href</a></iframe><h3><strong>A cohesive digital experience</strong></h3><p>Embedded within 72 is the story of SAP’s journey and commitment to innovation. It’s not just a set of characters; it’s a reflection of the company’s values and dedication to creating inclusive and user-friendly experiences. As a font, it enhances every touchpoint across the user experience — from our marketing website and brand campaigns to our digital products and software — making every interaction with SAP clear, engaging, and memorable. And as a symbol, it serves as yet another way in which SAP is helping to make workforce software more accessible, delightful, and powerful for today’s businesses.</p><p><em>Experience matters. Follow our journey as we transform the way we build products for enterprise on </em><a href="http://www.sap.com/design"><em>www.sap.com/design</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2fa48eea4a1a" width="1" height="1" alt=""><hr><p><a href="https://medium.com/sap-design/the-72-font-enhancing-digital-product-experience-and-usability-2fa48eea4a1a">The 72 Font: Enhancing Digital Product Experience and Usability</a> was originally published in <a href="https://medium.com/sap-design">Experience Matters</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Light and Fast: Designing the New SAP Web Components Logo]]></title>
            <link>https://medium.com/sap-design/light-and-fast-designing-the-new-sap-web-components-logo-723cb66e6562?source=rss-faa705307dff------2</link>
            <guid isPermaLink="false">https://medium.com/p/723cb66e6562</guid>
            <category><![CDATA[design-process]]></category>
            <category><![CDATA[logo-design]]></category>
            <category><![CDATA[case-study]]></category>
            <category><![CDATA[ux-design]]></category>
            <category><![CDATA[design]]></category>
            <dc:creator><![CDATA[SAP Design]]></dc:creator>
            <pubDate>Wed, 31 Jul 2024 07:56:35 GMT</pubDate>
            <atom:updated>2024-07-31T08:26:58.524Z</atom:updated>
            <content:encoded><![CDATA[<h4>Explore the detailed design process behind creating a compelling logo, from rough sketches to final approval, using the new SAP Web Components dynamic flame logo as a case study</h4><p><em>Written by Luisa Kohl, User Experience Design Specialist at</em> <em>SAP Design</em></p><figure><img alt="Graphic design featuring a blue and purple flame icon in various styles. The main icon is displayed prominently on a gradient background, with smaller variations and design outlines on a dark blue background. The word ‘COMPONENTS’ is faintly visible in the background." src="https://cdn-images-1.medium.com/max/1024/1*DzfHw29aGqRVI8IodaLoMg.png" /><figcaption>Evolution of the new SAP Web Components Logo</figcaption></figure><p><a href="https://sap.github.io/ui5-webcomponents/">Web Components</a> now have their very own logo: a dynamic flame with a spark. Here’s how I went about designing this logo, from the first rough sketches to the final approved design.</p><p>Until now, the only logo used was the general SAPUI5 logo, a stylized phoenix in red or blue hues, dramatically raising its wings above its head and following you with its ever-watchful eyes. As far as logos go, it’s hard to beat that one.</p><figure><img alt="A stylized, circular logo of an orange and red phoenix on a black background." src="https://cdn-images-1.medium.com/max/502/1*0UBkEwRIzzwscQTpwAv26w.png" /><figcaption><em>Current SAPUI5 logo</em></figcaption></figure><h3>The task</h3><p>But that was the task: to come up with a unique logo for SAP Web Components. It started out as an initiative by the UI5 Web Components team. They had used the SAPUI5 logo so far to express the same level of enterprise qualities that people know from SAPUI5. After realizing confusions between the separate product offerings caused by re-using the logo, the web components team wanted us, the SAP Design System team, to refer to it in the new design if possible. The keywords “lightweight” and “fast” were chosen because they represent the main characteristics of web components. It soon became clear that the logo should become a logo for all SAP Web Components, as anything else would have added to the already existing confusion. This shows how important a well-designed logo is and in what way a logo can help create clarity.</p><h3>Brainstorming</h3><p>Now that the task was clear I could get to work. Designing is like solving a puzzle with your own creativity. You have some givens, a lot of question marks, and then you start exploring, which is where the fun begins. What I usually do as a first step is to write down my immediate associations. If I don’t have any immediate ideas, I go to <a href="https://thenounproject.com/search/icons/?q=fast">The Noun Project</a> for inspiration, it usually does the trick. Go and check it out! This time, it wasn’t necessary, though.</p><p>What images would you think of when you hear the words light and fast, bearing in mind a phoenix? As such, the phoenix, endlessly being reborn, is an ideal symbol for reusability, so I didn’t want to lose that connection. I immediately thought of:</p><ul><li>a firefly</li><li>an origami-style phoenix</li><li>a mixture of a feather and a flame.</li></ul><figure><img alt="Three square icons with black backgrounds featuring orange and red designs: the first icon depicts a stylized insect, the second icon shows a geometric bird, and the third icon displays a flame." src="https://cdn-images-1.medium.com/max/1024/1*I12ORsvpaOCsJyxRTAKzZQ.jpeg" /><figcaption><em>The 3 concepts initially presented: firefly, origami phoenix, flame/feather</em></figcaption></figure><p>I presented three of these concepts, and the flame/feather was the winner. The feather symbolizes the lightness, and the flame connects it to the phoenix. As the logo has to work on a global scale, I also checked what the flame stands for in different cultures, and I found out that frequently, a single flame is often associated with a guiding light in the darkness, which I also found very fitting.</p><h3>First iterations</h3><p>My first three iterations of this flame looked as follows:</p><figure><img alt="Three square icons with black backgrounds, each featuring a progressively more stylized orange and red flame design." src="https://cdn-images-1.medium.com/max/1024/1*QwTc6zl-8BYc4OJcwpPHtQ.png" /><figcaption><em>Iterations 1, 2, and 3</em></figcaption></figure><p>My team gave the feedback that the flame in the first image was too angular and not dynamic enough, and that the feather metaphor wasn’t really recognizable. They also thought the overall design too simple. The Web Components team found the second design clearer, but somehow the flame too long and thin. The third iteration then was almost there. In order to get the geometry right, I drew circles in Illustrator so that flames really followed the grid.</p><p>And voilà, we had the final design!</p><figure><img alt="An orange and red flame icon on a black background, overlaid with a grid of intersecting circles outlining the design geometry." src="https://cdn-images-1.medium.com/max/960/1*BVzcAgk5LCkMznGdof4FzQ.png" /><figcaption><em>Adjusting shape to a grid</em></figcaption></figure><h3>Color options</h3><p>As regards the color options, we started out with the same as the SAPUI5 phoenix: The Red to Mango (fire) and Indigo Teal (water) options were taken. Then the SAP branding color palette was added. I also checked what the logo looks like in black-and-white and in solid variants to make sure that it works in our high-contrast themes, which people with low vision like using, for example.</p><figure><img alt="Two flame icons side by side, the left one in orange and red representing fire, and the right one in blue and green representing water, each with a corresponding color gradient bar below." src="https://cdn-images-1.medium.com/max/744/1*Wx6X2IjWXare4vGflMZbhA.png" /><figcaption><em>Red to Mango (fire) and Indigo Teal (water) color options</em></figcaption></figure><p>With the rebranding, the Web Components teams can convey clearly what they stand for and attract the audiences that are interested in this product. Having this brand-new (pun intended) logo <a href="https://sap.github.io/ui5-webcomponents/">on the UI5 Web Components page</a> is of course only the first step, but a very important one. After all, that is what visual design can offer: clarity, light and fast.</p><p><em>Experience matters. Follow our journey as we transform the way we build products for enterprise on </em><a href="http://www.sap.com/design"><em>www.sap.com/design</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=723cb66e6562" width="1" height="1" alt=""><hr><p><a href="https://medium.com/sap-design/light-and-fast-designing-the-new-sap-web-components-logo-723cb66e6562">Light and Fast: Designing the New SAP Web Components Logo</a> was originally published in <a href="https://medium.com/sap-design">Experience Matters</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How to Write Good Prompts for ChatGPT]]></title>
            <link>https://medium.com/sap-design/how-to-write-good-prompts-for-chatgpt-fa174865d40f?source=rss-faa705307dff------2</link>
            <guid isPermaLink="false">https://medium.com/p/fa174865d40f</guid>
            <category><![CDATA[sap-design]]></category>
            <category><![CDATA[writing-tips]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[writing-prompts]]></category>
            <category><![CDATA[chatgpt]]></category>
            <dc:creator><![CDATA[SAP Design]]></dc:creator>
            <pubDate>Mon, 22 Jul 2024 08:07:29 GMT</pubDate>
            <atom:updated>2024-07-22T08:07:29.126Z</atom:updated>
            <content:encoded><![CDATA[<h4>Learn how to prompt effectively to get the best results from ChatGPT</h4><p><em>Written by Dr. Marieke Storm, Communications and Content Senior Specialist at SAP Design</em></p><figure><img alt="Young woman in glasses working on a laptop by a large window, wearing a blue shirt and ripped jeans, with natural light streaming in." src="https://cdn-images-1.medium.com/max/1024/1*lS1DDOAgJc_0vKvWIBFuaA.jpeg" /></figure><p>You want to write a text promoting a product launch. You use ChatGPT to formulate the text. But the outcome is not at all what you wanted — it’s bland, missing the point, and just plain boring. So with a sigh you decide to write the text yourself after all. Sounds familiar?</p><p>Writing good prompts is a skill. That’s why companies are currently hiring prompt engineers to do exactly that. This offers a whole <a href="https://medium.com/@fniazi201276/the-emergence-of-prompt-engineering-bridging-linguistics-and-ai-12a610b21932">new career perspective for students of subjects such as linguistics or philosophy</a>, since good prompt writing is based on the understanding of language and logic, as we will see.</p><h3>What you give is what you get</h3><p>Rule number one: What you get out mirrors what you put in. In other words, generic prompts produce generic answers.</p><p><strong>Example</strong>:</p><p><strong>Q:</strong> <em>Give me the most recent info on artificial intelligence. Use 50 words.</em></p><p><strong>A: </strong><em>As of 2022, AI advances include developments in natural language processing, machine learning, and autonomous driving. There’s growing emphasis on AI ethics, addressing bias, and transparency. AI also plays a significant role in COVID-19 management, from predictive modeling to accelerating vaccine development. It’s increasingly integrated into daily technology.</em></p><p>The answer ChatGPT provides us with is as generic as our question. This might be fine in some instances, but in most cases, it won’t be very useful for our text production.</p><p>Here’s a more refined approach:</p><p><strong>Example:</strong></p><p><strong>Q: </strong><em>You are an academic and you are writing a scientific paper. Your target audience are specialists in the field. Write a definition of artificial intelligence based on recent academic sources only. Indicate your sources. Use 50 words.</em></p><p><strong>A:</strong> <em>Artificial Intelligence is a subset of computer science which focuses on creating intelligent systems capable of performing tasks that require human intelligence. These tasks include learning, reasoning, problem-solving, perception, language understanding (Russell &amp; Norvig, 2020)[1].<br>Source: [1] Russell, S. J., &amp; Norvig, P. (2020). Artificial intelligence: A modern approach. Malaysia; Pearson Education Limited.</em></p><p>The answer differs significantly from the one above in style and content. And only because we’ve changed the prompt a little bit!</p><h3>Essential elements of good prompts</h3><p>What constitutes a good prompt? Good prompts start with a clear, concise and unambiguous instruction. This instruction is then supported by adding context and/or examples. Always indicate what kind of output should be created — a presentation, an e-mail to your colleagues, or a promotional text to business clients.</p><p>A good model to follow is <a href="https://www.linkedin.com/learning/how-to-research-and-write-using-generative-ai-tools/how-to-write-an-effective-prompt-for-ai?resume=false&amp;u=57692769">Dave Birss’ CREATE concept</a>:</p><ul><li><strong>C for Character</strong><br>Tell ChatGPT what character it should assume:<br><em>You are an experienced copywriter in an IT company.</em></li><li><strong>R for Request</strong><br>This is the actual instruction for ChatGPT:<br><em>Write a promotional text for the product XYZ for our website. The product has the following features …The target audience is …</em></li><li><strong>E for Examples</strong> (optional)<br>If you have good examples, add them. ChatGPT will copy them in style and content.</li><li><strong>A for Adjustments</strong><br>Give ChatGPT further (stylistic) directions, either in your first prompt or in further prompts:<br><em>Focus on…</em> / <em>Use emotional adjectives</em> / <em>Use plain English</em> / <em>Use subheaders</em> / <em>Use bullet points</em></li><li><strong>T for Type of output</strong><br>Indicate the type of text you want: a poem, a dictionary entry, a love letter, … Also add the word length.</li><li><strong>E for Extras.</strong> Add extra phrases to help get the right results.<br>For example:<br>- <em>Explain your thinking</em>: to follow ChatGPT’s logic.<br>- <em>Ignore everything before this point</em>: to start afresh.<br>- <em>Ask me questions before you answer</em>: to let the program decide where it needs more information.<br>- <em>Answer only using reliable sources and cite those sources</em>: especially for academic contexts, but the results are still not very reliable, which means that you should always check them.<br>- <em>Respond “Unsure about answer” if you are not sure about the answer</em>: to make ChatGPT indicate levels of insecurity in its answers. This helps to identify so-called hallucinations.<br>Also, always tell ChatGPT what you want it to do, and not what you don’t want it to do, because it sometimes overlooks the word ‘don’t’.</li></ul><p>So, with “create”, the following prompt <br>“<em>Summarize this text.” <br></em>turns into something like <br>“<em>You are an experienced A working for a B company. Summarize this text for a C publication. In your summary, focus on D. The target audience is E. Use X words. Use Y language.”</em></p><p>For further inspiration, you can also look at the example above again.</p><p>Write prompts until you get the result. Generative AI tools don’t work like search engines. We tend to assume that the text we get at our first attempt is the final thing, but the contrary is the case: If you get a result that is not (quite) right, rephrase and use different techniques until ChatGPT creates a good text for you. Remember: It’s called ChatGPT for a reason: chat with it. If you don’t like the output, tell the program what to change.</p><h3>Popular prompt techniques</h3><p>There are many ways to reach your goals. Here is a summary of the (currently) most popular techniques.</p><ul><li><strong>Zero-shot prompt<br></strong>The way most people use ChatGPT — writing a prompt without follow-up prompts. If you do this, you assume that the model second-guesses a lot. Usually, this is not very effective.</li><li><strong>Few-shot prompt<br></strong>With several consecutive prompts for one task, you achieve your goal in small steps. You also give it the context and ideally also examples. With this method, you’ll get better responses, because the model will be able to determine your desired tone of voice, structure, and so on. You might run into difficulties as soon as logic reasoning is required, though. In these cases, the approach described next is more fruitful.</li><li><strong>Chain-of-thought prompt<br></strong>Here, you give an example and explain the reasoning behind the answer in the example. In the same prompt, you give ChatGPT a similar task to solve in the same way (<a href="https://arxiv.org/abs/2201.11903">Wei et al.</a>). An alternative is to add “think step by step” to the prompt, which means that you’ll be able to see where ChatGPT’s logical flaw is. You can then tweak your prompt again (<a href="https://arxiv.org/abs/2205.11916">Kojima et al.</a>).</li></ul><h3>Final thoughts</h3><p>In this post, my focus was on the methods you can immediately use without any further understanding or preparation. For more complex ones such as Retrieval Augmented Generation, go to sites such as <a href="https://www.promptingguide.ai/techniques/rag">promptingguide.ai</a>. Although I tested them with ChatGPT, these tips apply to other generative AI tools, too.</p><p>I hope it has become clear that you need to invest a bit of time to try out several few shot prompts until you get a text you can work with. Sometimes you will find that writing short texts yourself is much quicker indeed. Or you’ll opt for a hybrid approach, using the tool to optimize your text rather than letting it write something from scratch for you. Whatever you do, learning its advantages AND its limitations is essential.</p><h3>Sources/further study</h3><ul><li><a href="https://arxiv.org/abs/2201.11903">Chain-of-Thought Prompting Elicits Reasoning in Large Language Models (arxiv.org)</a></li><li><a href="https://www.linkedin.com/pulse/create-unleashing-art-crafting-irresistible-ai-benjamin-c-/">CREATE: Unleashing the Art of Crafting Irresistible AI Prompts | LinkedIn</a></li><li><a href="https://www.linkedin.com/learning/how-to-research-and-write-using-generative-ai-tools">How to research and write using generative AI tools (linkedin.com)</a></li><li><a href="https://openai.com/research/language-models-are-few-shot-learners">Language models are few-shot learners (openai.com)</a></li><li><a href="https://www.promptingguide.ai/">Prompt Engineering Guide | Prompt Engineering Guide (promptingguide.ai)</a></li><li><a href="https://www.linkedin.com/learning/introduction-to-prompt-engineering-for-generative-ai/taking-gpt-3-for-a-ride?autoSkip=true&amp;leis=OCN&amp;resume=false&amp;u=57692769">Taking GPT-3 for a ride (linkedin.com)</a></li></ul><p><em>Experience matters. Follow our journey as we transform the way we build products for enterprise on </em><a href="http://www.sap.com/design"><em>www.sap.com/design</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=fa174865d40f" width="1" height="1" alt=""><hr><p><a href="https://medium.com/sap-design/how-to-write-good-prompts-for-chatgpt-fa174865d40f">How to Write Good Prompts for ChatGPT</a> was originally published in <a href="https://medium.com/sap-design">Experience Matters</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>