<?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 Nikita Maloo on Medium]]></title>
        <description><![CDATA[Stories by Nikita Maloo on Medium]]></description>
        <link>https://medium.com/@nikitamaloo222?source=rss-bd1ea945185d------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*koRmX9vg5v0hfXdrMoXXEw.jpeg</url>
            <title>Stories by Nikita Maloo on Medium</title>
            <link>https://medium.com/@nikitamaloo222?source=rss-bd1ea945185d------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 03:04:02 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@nikitamaloo222/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[Powering Digital Success: The Essentials of Seamless Payment Integration]]></title>
            <link>https://medium.com/@nikitamaloo222/powering-digital-success-the-essentials-of-seamless-payment-integration-4d005f7d5267?source=rss-bd1ea945185d------2</link>
            <guid isPermaLink="false">https://medium.com/p/4d005f7d5267</guid>
            <dc:creator><![CDATA[Nikita Maloo]]></dc:creator>
            <pubDate>Tue, 05 Sep 2023 22:15:13 GMT</pubDate>
            <atom:updated>2023-09-06T22:59:12.023Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*jmUGI-J-FO-nRAoHPLMA9A.png" /><figcaption>Picture Credits — <a href="https://appinventiv.com/blog/payment-gateway-integration-the-process-and-what-nexts/">https://appinventiv.com/blog/payment-gateway-integration-the-process-and-what-nexts/</a></figcaption></figure><p>In today’s rapidly evolving digital landscape, payments have emerged as the cornerstone of online transactions. The pivotal role that payments play in shaping user experiences, driving business growth, and establishing trust cannot be understated. As we engage with mobile and web applications for shopping, bookings, subscriptions, and more, the efficiency and security of payment methods deeply impact user confidence and satisfaction. Let’s delve into some of the core considerations underpinning the integration of payments within these applications:</p><ol><li><strong>Diverse Payment Options — Fostering trust through user-centric choices</strong></li></ol><p>Offering a range of payment choices (credit/debit cards, digital wallets, bank transfers, etc.) caters to various user preferences, boosting confidence. Ensure that the selected payment methods align with the target audience’s tendencies. Adapt to regional payment habits and offer localized alternatives where necessary.</p><p>For instance, in regions with high mobile penetration, supporting mobile money can be crucial.</p><p>In regions where credit card adoption might be low, offering alternative payment methods like digital wallets (e.g., Alipay, WeChat Pay in China) or cash-on-delivery can significantly increase user accessibility and trust.</p><p><strong>2. Enhanced User Experience (UX) — Streamlining for a seamless transaction</strong></p><p>A smooth user experience boosts conversion rates and user satisfaction. Design a simple and intuitive payment flow to minimize friction during the checkout process. Provide clear and concise instructions, error messages, and confirmation notifications. Incorporate features like one-click payments or saved card details for returning users.</p><p>For instance, Airbnb’s payment process is an excellent example, enabling users to seamlessly pay for bookings without feeling like they’re leaving the platform.</p><p><strong>3. Device Compatibility — Ensuring consistency across devices</strong></p><p>Uniformity across different devices (smartphones, tablets, desktops) and screen sizes ensures a seamless experience. Implement responsive design to accommodate various screen resolutions.</p><p>For instance, an online food delivery app should provide a consistent payment experience whether users are ordering from their smartphones or desktop computers.</p><p><strong>4. Feasibility Consideration — Evaluating business implications</strong></p><p>Assess the feasibility of implementing various payment options from both technical and business standpoints. Consider the cost of integrating each method, revenue impact, and the target audience’s preferences.</p><p>For instance, a subscription-based app might prioritize recurring card payments over one-time bank transfers due to stability and user convenience.</p><p>Uber’s decision to introduce Uber Cash, a prepaid balance system, allows users to preload funds and make quick payments, enhancing the company’s cash flow.</p><p><strong>5. Performance, Speed &amp; Latency — Prioritizing swift transactions</strong></p><p>Optimize the payment process to minimize latency and ensure quick response times, providing a seamless payment experience and preventing user frustration. Implement best practices for performance optimization, such as data caching, reducing network requests, and optimizing image and resource loading.</p><p>For instance, Uber’s payment system is optimized for speed; users experience real-time fare estimates and seamless payments, even in areas with fluctuating network connectivity.</p><p><strong>6. Scalability &amp; Reliability — Anticipating growth for consistency</strong></p><p>Planning for growth ensures consistent performance. Design and Build a scalable payment infrastructure that can accommodate increasing transaction volumes as the user base expands. Implement robust error-handling mechanisms to minimize downtime and address payment failures and ensure service availability.</p><p>For instance, Payment providers like Stripe and Braintree are known for their reliable and scalable APIs, enabling businesses to handle high transaction volumes without service disruptions.</p><p><strong>7. Location and Localization — Catering to local preferences</strong></p><p>Supporting local preferences enhances user adoption. Ensure compliance with applicable legal and regulatory requirements in each operating region. Accommodate local currencies and payment methods relevant to the target market. Adapt to language preferences.</p><p>Spotify, for instance, tailors its payment options to accommodate local preferences. In some countries, it accepts payment through carrier billing, making it accessible to users without credit cards</p><p><strong>8. Testing and Quality Assurance — Thorough testing ensures reliability</strong></p><p>Conduct comprehensive testing of the payment functionality to ensure its reliability, accuracy, and compatibility across various devices, operating systems, and network conditions. Perform end-to-end testing of the entire payment flow, including scenarios like successful payments, failed payments, and error handling. Simulate real-world situations with different payment methods to validate the effectiveness and address any issues or vulnerabilities.</p><p>For instance, PayPal Sandbox and Stripe’s test mode allow developers to simulate transactions before going live, reducing the risk of bugs affecting real payments.</p><p><strong>9. Risks and Fraud Prevention — Prioritizing security is non-negotiable</strong></p><p>Implement robust security measures to safeguard user payment information and prevent unauthorized access or fraudulent activities. Utilize encryption techniques, tokenization, and secure network protocols to protect sensitive data during payment transactions. Employ fraud detection systems and algorithms to identify and prevent fraudulent transactions in real time.</p><p>For instance, PayPal (Fraud Protection Advanced) and Stripe (Radar) fraud detection tools analyze billions of transactions to identify suspicious activities.</p><p><strong>10. Data Privacy — Transparent data handling fosters trust</strong></p><p>Responsible data management builds trust. Adhere to data protection regulations such as GDPR and clearly communicate how user data will be used during payments. Follow industry standards and regulations like PCI DSS to ensure payment data security. Obtain explicit consent for data processing and storage.</p><p>For instance, Apple Pay anonymizes transaction data and uses tokenization to ensure that user card information is never stored on Apple servers or shared with merchants.</p><p><strong>11. Feedback Monitoring &amp; Optimization — Adapting based on insights is invaluable</strong></p><p>Continuously monitor user feedback and analytics to identify trends and areas of improvement. Regularly optimize the payment process based on feedback and data analysis.</p><p>For instance, Airbnb frequently updates its payment process based on user feedback, making it easier and more intuitive over time.</p><p>In conclusion, integrating seamless and secure payment systems in mobile and web applications is paramount. These considerations, spanning diverse payment options, user experience, feasibility, performance, security, and localization, collectively craft a robust payment ecosystem. By adhering to these principles and continually refining the process based on user feedback, businesses can foster trust, enhance user satisfaction, and drive lasting success in the digital landscape.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4d005f7d5267" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Securing Trust: Seller Risk Assessment in E-commerce Marketplaces]]></title>
            <link>https://medium.com/@nikitamaloo222/securing-trust-seller-risk-assessment-in-e-commerce-marketplaces-84afa22af2cb?source=rss-bd1ea945185d------2</link>
            <guid isPermaLink="false">https://medium.com/p/84afa22af2cb</guid>
            <dc:creator><![CDATA[Nikita Maloo]]></dc:creator>
            <pubDate>Fri, 28 Jul 2023 05:46:37 GMT</pubDate>
            <atom:updated>2023-08-03T06:13:13.876Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*1EMFkTORQwSbTsYv1DXH3w.png" /></figure><p>Onboarding new sellers to an e-commerce marketplace is undoubtedly an exciting endeavor, promising growth and expansion for both the platform and the sellers themselves. However, this process also brings along inherent risks that must be managed with precision to ensure a seamless and successful integration.</p><p>Let’s delve into some of the most common risks that come with onboarding new sellers and explore strategies to mitigate these challenges.</p><p><strong>Common risks that come with onboarding new sellers include:</strong></p><ol><li><strong>Fraudulent Sellers:</strong> One of the primary risks is the possibility of onboarding fraudulent sellers who may engage in deceptive practices, sell counterfeit goods, or conduct other dishonest activities on the platform.</li><li><strong>Quality and Authenticity of Products: </strong>There is a risk that new sellers may not meet the quality standards expected by the marketplace, leading to negative customer experiences and potential returns or disputes and harming the marketplace’s reputation.</li><li><strong>Compliance and Legal Risks: </strong>Sellers may inadvertently or intentionally violate laws or regulations related to product labeling, safety standards, copyrights, trademarks, or other legal requirements, exposing the marketplace to legal liabilities.</li><li><strong>Data Security and Privacy:</strong> When new sellers join the marketplace, there is a risk of data breaches or unauthorized access to sensitive customer information if appropriate security measures are not in place.</li><li><strong>Inventory Management and Stockouts: </strong>New sellers may struggle with inventory management, leading to issues such as overselling, out-of-stock products, or delays in fulfilling orders leading to dissatisfied customers.</li><li><strong>Price Competitiveness and Price Violations: </strong>New sellers may engage in price violations, such as price gouging or undercutting competitors unfairly, disrupting the overall pricing integrity of the marketplace.</li><li><strong>Customer Service and Responsiveness:</strong> If new sellers are not adequately trained or equipped to handle customer inquiries and complaints, it could harm the marketplace’s reputation and customer trust.</li><li><strong>Payment and Financial Risks: </strong>New sellers may encounter issues with payment processing or financial transactions, leading to delayed or incorrect payments.</li><li><strong>Shipping and Delivery Problems:</strong> New sellers may face challenges in managing shipping and delivery logistics, resulting in delays, lost packages, or incorrect deliveries.</li><li><strong>Product Returns and Refunds: </strong>Onboarding new sellers increase the possibility of product returns and refund requests, which could impact the marketplace’s profitability and customer satisfaction.</li><li><strong>Seller Performance and Metrics: </strong>Poor performance by new sellers, such as late shipments or low product ratings, can affect the overall performance metrics of the marketplace.</li></ol><p>If fraudulent or risky sellers successfully onboard e-commerce marketplace platforms, they can engage in various deceptive activities that negatively impact both platform’s reputation and customer experiences.</p><p><strong>Here are some of the potential fraudulent activities that these sellers may engage in:</strong></p><ol><li><strong>Fake or Counterfeit Products:</strong> Sellers may list fake or counterfeit products, misrepresenting them as genuine brand-name items, deceiving customers into purchasing lower-quality goods.</li><li><strong>Shilling and Fake Reviews: </strong>Sellers may create fake accounts to post positive reviews for their products or post negative reviews for products sold by other sellers or incentivize buyers to do the same, artificially manipulating search rankings and product visibility.</li><li><strong>Price Gouging:</strong> Some sellers may exploit high-demand situations by significantly raising prices, and taking advantage of customers during emergencies or limited supply scenarios.</li><li><strong>Unauthorized Selling &amp; Brand Misuse:</strong> Unauthorized sellers may list products without proper authorization from the brand or manufacturer, leading to potential legal issues and product quality concerns.</li><li><strong>Selling Restricted or Banned Items: </strong>Sellers may attempt to list and sell items that are restricted or prohibited by the platform’s policies.</li><li><strong>Order and Shipping Fraud:</strong> Sellers may engage in order fraud, such as falsely claiming items were not delivered, or shipping fraud, where they may manipulate shipping tracking information.</li><li><strong>Account Takeover:</strong> Fraudsters may attempt to take over legitimate seller accounts to access sensitive information, commit fraud, or damage the reputation of the original account holder.</li><li><strong>Identity Theft: </strong>Criminals may use stolen or synthetic identities to create fake seller accounts, engaging in fraudulent activities.</li><li><strong>Phishing and Scams: </strong>Some sellers may use phishing tactics to trick buyers into revealing personal information, login credentials, or financial details.</li><li><strong>Gift Card Fraud:</strong> Fraudulent sellers may offer gift cards or vouchers with no real value, deceiving customers into making purchases and causing financial harm to buyers.</li><li><strong>Vendor Credential Abuse: </strong>Third-party sellers may misuse the credentials of authorized vendors to sell unauthorized products.</li><li><strong>Unethical Dropshipping Practices: </strong>Sellers may engage in unethical dropshipping practices, leading to delayed shipments or unfulfilled orders.</li><li><strong>Bait-and-Switch: </strong>Sellers may advertise one product but deliver a different or lower-quality item to customers, misleading them into making a purchase.</li><li><strong>Fake Discounts and Sales: </strong>Sellers may falsely inflate original prices to create the illusion of significant discounts during sales events, misleading customers into thinking they are getting a great deal.</li><li><strong>Arbitrage Abuse: </strong>Sellers may exploit price discrepancies between different marketplaces to resell products at higher prices, potentially leading to customer dissatisfaction.</li></ol><p>Therefore, identifying the intention of sellers on e-commerce marketplace platforms is crucial to ensure a safe and trustworthy environment for buyers and other legitimate sellers. To assess seller intentions, marketplaces can use a combination of pre-onboarding and ongoing monitoring strategies.</p><p><strong>Here are some approaches to consider:</strong></p><ol><li><strong>Robust Seller Verification: </strong>Implement a robust seller verification process during onboarding to ensure that only legitimate and trustworthy sellers are allowed to operate on the platform. Conduct thorough background checks to ensure sellers have a credible history and to identify any previous instances of fraudulent activities or legal issues.</li><li><strong>Data Analytics &amp; Machine Learning Algorithms: </strong>Utilize data analytics &amp; advanced machine learning algorithms to analyze seller behavior and identify patterns associated with fraudulent activities.</li><li><strong>Automated Risk Assessment: </strong>Employ automated risk assessment systems to flag high-risk sellers based on various factors like product listings, pricing, and customer feedback.</li><li><strong>Customer Reviews and Ratings: </strong>Encourage users to report suspicious or fraudulent activities they encounter on the platform. Monitor customer reviews and ratings to identify sellers with a high volume of negative feedback.</li><li><strong>Seller Performance Metrics: </strong>Track seller performance metrics, such as order defect rate and late shipping rate, and cancellation rate to evaluate seller behavior and identify potential issues. Look for sellers consistently receiving negative feedback related to product quality or authenticity.</li><li><strong>Intellectual Property Violations: </strong>Regularly check for intellectual property violations to ensure sellers are not infringing on trademarks or copyrights.</li><li><strong>Data Sharing and Collaboration with Brands and Regulatory Authorities: </strong>Collaborate with brands and law enforcement agencies to share information on fraudulent sellers and take collective action against them.</li><li><strong>Customer Support &amp; Seller Feedback Monitoring: </strong>Monitor seller communication with customers to identify phishing attempts or suspicious behavior. Analyze customer support interactions to spot trends in fraudulent activities or scam attempts.</li><li><strong>Listing and Pricing Audits: </strong>Conduct regular audits of product listings and pricing to detect misleading information and price gouging.</li><li><strong>Suspension and Termination Policies: </strong>Have clear and transparent suspension and termination policies for sellers found engaging in fraudulent activities.</li><li><strong>Training and Awareness: </strong>Educate sellers about the platform’s policies and guidelines, as well as the consequences of engaging in fraudulent activities.</li><li><strong>Policy Enforcement and Penalties: </strong>Impose penalties on sellers found guilty of fraudulent behavior and take legal action when necessary.</li><li><strong>Data Security Measures: </strong>Implement stringent data security measures to protect customer information and prevent data breaches.</li><li><strong>Anonymous Buyer Tests: </strong>Conduct anonymous buyer tests to purchase products from sellers and assess product quality, authenticity, and customer service.</li><li><strong>Continuous Improvement: </strong>Regularly review and improve fraud detection and prevention mechanisms based on evolving industry best practices and technological advancements.</li></ol><p><strong>Some important signals or indicators that can aid in the above risk assessment approach are:</strong></p><ol><li><strong>Significant Sales Spike</strong>: Sudden surge in sales without a corresponding increase in customer reviews or engagement may be a red flag.</li><li><strong>Product Review Patterns</strong>: A large number of positive reviews posted within a short time or repetitive language in reviews could indicate fake or incentivized reviews.</li><li><strong>Order and Shipping Analysis: </strong>Suspicious behaviors such as frequent cancellations or changes in shipping locations.</li><li><strong>Product Pricing Analysis:</strong> Regularly review product prices to detect sudden price spikes or unusually low prices, which could be indicative of price gouging or counterfeit products.</li><li><strong>Sudden High-Volume Listings</strong>: A rapid increase in the number of listings by a seller may indicate a sudden influx of counterfeit or unauthorized products.</li><li><strong>High Volume of Negative Reviews</strong>: A disproportionate number of negative reviews or complaints about a seller’s products or services may signal issues with product quality or customer service or fake negative reviews.</li><li><strong>Mismatched Product Images</strong>: Sellers using images that do not match the actual product being sold may be selling counterfeit items.</li><li><strong>Geolocation and IP Analysis:</strong> Monitor seller account logins and activities for suspicious geolocation and IP addresses to detect multiple accounts operating from the same location.</li><li><strong>Transaction Pattern Analysis: </strong>Utilize data analytics to identify unusual transaction patterns, such as multiple orders from the same IP address or billing/shipping address.</li><li><strong>Unverified Contact Information</strong>: Sellers providing fake or unverifiable contact information may be trying to hide their true identity.</li><li><strong>Rapidly Changing Business Details</strong>: Frequent changes to a seller’s business name, address, or other details may be an attempt to avoid detection.</li><li><strong>High Rate of Order Cancellations</strong>: Excessive order cancellations by a seller could be an indicator of inventory management issues or fraudulent practices.</li><li><strong>Abnormal Shipping Locations</strong>: Shipping from unexpected or unusual locations may raise suspicion of potential counterfeit goods.</li><li><strong>Unresponsive Sellers</strong>: Sellers who are unresponsive to customer inquiries or complaints may be disreputable or engaged in fraudulent activities.</li><li><strong>Suspicious Seller Behavior:</strong> Unusual seller activity, such as changing prices frequently, offering products out of their typical category, or duplicating listings, may warrant investigation.</li></ol><p>By carefully understanding and proactively managing these risks, marketplaces can foster a secure and trustworthy environment for both buyers and sellers, ultimately contributing to sustainable growth and a positive reputation.</p><p>The key thing to remember is that identifying risky sellers is an ongoing process, and no single approach is foolproof. Combining multiple strategies and leveraging technology can help e-commerce marketplaces proactively detect and take action against sellers with malicious intentions. The timely and appropriate response to fraudulent sellers is essential in maintaining a positive shopping experience and building trust with buyers on the platform.</p><p>#viewsofmyown #FraudNugget #marketplacerisk #riskmanagement #trustandsafety #fraudrisk #fraudawareness #frauddetection #fraudprevention #fraudmanagement #riskmanagement #ecommerce #marketplace #productstrategy #sellerperformace #sellerrisk #selleronboarding #marketplaceperformace</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=84afa22af2cb" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Identify Patterns & Commonalities:]]></title>
            <link>https://medium.com/@nikitamaloo222/identify-patterns-commonalities-2882bebf4a39?source=rss-bd1ea945185d------2</link>
            <guid isPermaLink="false">https://medium.com/p/2882bebf4a39</guid>
            <dc:creator><![CDATA[Nikita Maloo]]></dc:creator>
            <pubDate>Fri, 14 Jul 2023 00:34:55 GMT</pubDate>
            <atom:updated>2023-07-14T00:54:45.259Z</atom:updated>
            <content:encoded><![CDATA[<h3>Fighting Fraud Rings</h3><ol><li><strong>Identify Patterns &amp; Commonalities:</strong></li></ol><ul><li>Take a holistic approach when analyzing different characteristics to identify patterns rather than focusing on individual factors.</li><li>Fraud rings can easily circumvent the rules only created to target specific devices, IPs, card numbers, etc..</li><li>Some fraud rings can resurface after being dormant for a while and linking current and previous instances of fraud can provide valuable data for detecting other undetected bad accounts or preventing future occurrences. Analyzing behavioral and technical attributes together helps establish connections between current and previous fraudulent activities.</li><li>Look for commonalities between the behaviors of good users and bad users, but exercise caution in flagging behaviors exhibited by 100% of users as risky.</li></ul><p><strong>2. Gather More Intel in the Detection Process:</strong></p><ul><li>When feasible and without impacting customer experience, consider allowing bad actors to continue their activities to gather additional behavioral signals and characteristics.</li><li>Instead of an immediate reaction, acquiring more information during the detection process makes it more costly and less attractive for fraudsters to attempt further fraud.</li><li>Hiding when &amp; how the marketplace is detecting fraud is important to limit the options available to fraudsters for finding workarounds.</li><li>For instance, instead of blocking an IP during login, block it at the checkout stages or later in the flow. If you block it at checkout, fraudsters wouldn’t know if it was the card, the IP, or something else that led to being blocked. In some cases, fraudsters might try 5–10 different credit cards. With that information, the platform can block all of those cards too. If the IP was blocked at the front door (during login), fraudsters could have changed or masked the IP and would have attempted to log in again.</li></ul><p><strong>3. Don’t Set &amp; Forget — Rule Governance and Maintenance:</strong></p><ul><li>Avoid creating rules for every individual fraud ring encountered, else your rule set will be a mile long, and if a fraud ring changes its behavior, you might start catching good users and ruin their experiences.</li><li>Regularly assess and maintain your rule set to ensure its viability and effectiveness, ensuring that it achieves a good hit rate without impacting genuine users.</li><li>Turn off rules that are no longer effective or are catching legitimate users.</li><li>Ongoing rule evaluation is crucial to adapt to changing fraud patterns and optimize the effectiveness of the rule set.</li></ul><p>Please remember to adapt these strategies to your specific business needs and risk tolerance, as fraud prevention approaches may vary.</p><p>#FraudNugget #marketplacerisk #fraudrings #riskmanagement #trustandsafety #fraudrisk #fraudawareness #frauddetection #fraudprevention #productmanagement</p><p>Reference — <a href="https://www.linkedin.com/post/edit/7051426701137903616/#">Marketplace Risk</a> — <a href="http://invalid//">Virtual Event Link</a> — <a href="https://www.marketplacerisk.com/webinar-series/fighting-fraud...rings">https://www.marketplacerisk.com/webinar-series/fighting-fraud...rings</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2882bebf4a39" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Push and Pull Payments: Dynamics of Fund Transfers]]></title>
            <link>https://medium.com/@nikitamaloo222/push-and-pull-payments-dynamics-of-fund-transfers-ccda94bb905e?source=rss-bd1ea945185d------2</link>
            <guid isPermaLink="false">https://medium.com/p/ccda94bb905e</guid>
            <dc:creator><![CDATA[Nikita Maloo]]></dc:creator>
            <pubDate>Fri, 14 Jul 2023 00:05:51 GMT</pubDate>
            <atom:updated>2023-07-14T00:05:51.046Z</atom:updated>
            <content:encoded><![CDATA[<p>Let’s understand Push and Pull Payments: Pros, Cons, and Differences</p><p><strong>Introduction: </strong>In the modern era of digital transactions, push-and-pull payment methods have gained significant prominence. These terms refer to different mechanisms through which funds are transferred between parties. In this article, we will delve into push and pull payments, providing a clear explanation of their functionality, outlining the advantages and disadvantages they offer, and unraveling the similarities and differences between these two methods. By the end of this article, you will have a comprehensive understanding of these payment methods and their applications.</p><p><strong>Push Payments:</strong> Push payments, as the name suggests, involve the sender initiating the transaction and “pushing” funds to the recipient. In this scenario, the sender has control over the payment process. Examples of push payments include bank transfers, wire transfers, and mobile payments.</p><p><strong>Pros of Push Payments:</strong></p><ol><li><strong>Security</strong>: Push payments offer a higher level of security as the sender is in control of the transaction. They can ensure accuracy and verify recipient details before initiating the payment.</li><li><strong>Quick Settlement</strong>: Push payments typically result in a fast settlement, allowing recipients to access funds promptly.</li><li><strong>Easy Tracking</strong>: Since the sender initiates the payment, it is easier to track and reconcile transactions for record-keeping purposes.</li><li><strong>Versatility</strong>: Push payments can be conducted through various channels, including online banking, mobile apps, and electronic funds transfer.</li></ol><p><strong>Cons of Push Payments:</strong></p><ol><li><strong>Limited Recipient Control</strong>: Once the sender initiates the payment, they have limited control over the funds. If an incorrect recipient is specified, rectifying the mistake can be challenging.</li><li><strong>Inconvenience</strong>: As push payments require the sender to initiate the transaction, it can be cumbersome in situations where frequent payments need to be made to the same recipient.</li></ol><p><strong>Pull Payments:</strong> In contrast, to push payments, pull payments involve the recipient initiating the transaction and “pulling” funds from the sender’s account. Common examples of pull payments include credit card transactions and direct debits.</p><p><strong>Pros of Pull Payments:</strong></p><ol><li><strong>Convenience</strong>: Pull payments are generally convenient for recurring payments such as subscriptions or utility bills. The recipient can automatically pull funds from the sender’s account, reducing the sender’s effort.</li><li><strong>Recipient Control</strong>: The recipient has control over the transaction, ensuring accurate collection of funds.</li><li><strong>Flexibility</strong>: Pull payments can be scheduled in advance, allowing for automated transactions at specified intervals.</li><li><strong>Reduced Errors:</strong> As the recipient initiates the payment, there is a lower risk of incorrect recipient details, ensuring accurate transactions.</li></ol><p><strong>Cons of Pull Payments:</strong></p><ol><li><strong>Security Concerns</strong>: Since the recipient has access to the sender’s account information, there is a potential risk of unauthorized charges if proper security measures are not in place.</li><li><strong>Limited Recipient Verification</strong>: In pull payments, the sender has less control over verifying recipient details before the transaction, which can lead to accidental or unauthorized payments.</li></ol><p><strong>Differences between Push and Pull Payments:</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*WWF7grjgSBcwgEGyC6DFKg.png" /></figure><p><strong>Conclusion</strong>: Push and pull payments represent distinct approaches to fund transfers, each offering its own set of advantages and disadvantages. Push payments provide greater control and security for the sender, while pull payments offer convenience and automation for the recipient. By understanding their pros, cons, similarities, and differences, you can make informed decisions regarding the most appropriate payment method for your specific needs.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ccda94bb905e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Window Functions for Analytics in SQL]]></title>
            <link>https://medium.com/swlh/window-functions-for-analytics-in-sql-f4dac0906a55?source=rss-bd1ea945185d------2</link>
            <guid isPermaLink="false">https://medium.com/p/f4dac0906a55</guid>
            <category><![CDATA[sql]]></category>
            <category><![CDATA[advanced-analytics]]></category>
            <category><![CDATA[data-analytics]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[window-functions]]></category>
            <dc:creator><![CDATA[Nikita Maloo]]></dc:creator>
            <pubDate>Sat, 11 Jul 2020 01:37:36 GMT</pubDate>
            <atom:updated>2020-12-15T04:07:22.813Z</atom:updated>
            <content:encoded><![CDATA[<h4>Learn about how to use Window Functions in SQL.</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/841/1*GIt165I6w961xQ91m7tOJA.png" /><figcaption>Image is taken from <a href="https://www.sqlitetutorial.net/sqlite-window-functions/">SQLite</a></figcaption></figure><p>This article assumes that you have some basic SQL experience.</p><p>Window functions operate on a set of rows and return a single value for each row from the underlying query. The term window describes the set of rows on which the function operates. The window is defined by the <strong>OVER</strong> clause.</p><h4><strong>Different types of Window Functions</strong></h4><ul><li><strong>Aggregate Functions: </strong>MAX(), MIN(), AVG(), SUM(), COUNT(), etc.</li><li><strong>Ranking Functions: </strong>ROW_NUMBER(), RANK(), DENSE_RANK(), NTILE(), PERCENT_RANK(), CUME_DIST() etc.</li><li><strong>Analytic Functions:</strong>LEAD(), LAG(), FIRST_VALUE(), LAST_VALUE(), etc.</li></ul><p><strong>OVER clause defines the partitioning and ordering of rows (i.e. a window) for the above functions to operate on. Hence, these functions are called window functions. The OVER clause accepts the following three arguments to define a window for these functions to operate on.</strong></p><ul><li><strong>PARTITION BY:</strong> Divides the query result set into partitions. The window function is applied to each partition separately. Multiple fields need to be separated by a comma. If the PARTITION BYis not specified, then the grouping will be done on the entire table and values will be aggregated accordingly.</li><li><strong>ORDER BY:</strong> Sorts the rows within each partition. If ORDER BY is not specified, ORDER BY uses the entire table.</li><li><strong>ROWS or RANGE clause:</strong> Further subsets the rows within the partition by specifying start and endpoints within the partition. The default for ROWS or RANGE clause is RANGE BETWEEN UNBOUND PRECEDING AND CURRENT ROW</li></ul><h4>Window Function versus Group By</h4><ul><li>Window functions do not reduce the number of rows in the output unlike GROUP BY function.</li><li>Window functions can retrieve values from other rows, whereas GROUP BY functions cannot.</li><li>Window functions can calculate running totals and moving averages, whereas GROUP BY functions cannot.</li></ul><p>Let’s dive into the examples. The data I will be using here is the Customer Purchase Dataset <strong>(sqlbank3)</strong> that can be found <a href="https://github.com/nikitamaloo/Window-Functions-SQL/blob/master/Retail_Data.csv"><strong>here</strong></a><strong>.</strong></p><pre><strong>SELECT * FROM sqlbank3;</strong></pre><p>Here is how the data in the table looks like.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*tRcofcq3Gsf7gAh8MaiyHA.png" /><figcaption>Each INVOICENO is an order by a customer that can contain multiple items (STOCKCODE)</figcaption></figure><p>Starting with a simple SQL statement.</p><h4>SELECT, FROM, WHERE, GROUP BY, HAVING, ORDER BY, LIMIT</h4><p>When we write SQL statements they should be in the same order as mentioned below.</p><pre>SELECT *<br>FROM &lt;condition&gt;<br>WHERE &lt;condition&gt;<br>GROUP BY &lt;condition&gt;<br>HAVING &lt;condition&gt;<br>ORDER BY &lt;condition&gt;<br>LIMIT</pre><p>In terms of execution, however, the flow is different. That means when the SQL statements run on the server-side it is read in the order as mentioned below.</p><pre>FROM (it selects the table that is required)<br>WHERE (it filters the rows based on condition)<br>GROUP BY (it aggregates the result that was filtered)<br>HAVING (additional level of filtering for aggregated values)<br>SELECT (it selects the appropriate columns from the result)<br>ORDER BY (it arranges values in ascending or descending order)<br>LIMIT (it controls how many rows are outputted in the query)</pre><pre><strong>SELECT</strong> Quantity, <strong>COUNT</strong>(<strong>*</strong>) <strong>AS</strong> Quantity_Counts<br><strong>FROM</strong> sqlbank3<br><strong>WHERE</strong> UnitPrice <strong>&gt;=</strong> 5<br><strong>GROUP</strong> <strong>BY</strong> Quantity<br><strong>HAVING</strong> Quantity_Counts <strong>&lt;</strong> 450<br><strong>ORDER</strong> <strong>BY</strong> Quantity_Counts <strong>DESC<br>LIMIT</strong> 10;</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/382/1*frtYQNl6coSgG1Zrf1xc6g.png" /><figcaption>The output of the above query</figcaption></figure><p>Alright, let’s begin with the window functions.</p><p>window_function_name(&lt;<em>expression&gt;</em>) OVER ( &lt;<em>partition_clause&gt;</em> &lt;<em>order_clause&gt;</em>)</p><h4><strong>1. AVG(), SUM()</strong></h4><p>The <strong>OVER</strong> clause is used to determine which rows from the query are applied to the function, what order they are evaluated in by that function, and when the function’s calculations should restart.</p><pre><strong>SELECT</strong> InvoiceNo,</pre><pre><strong>AVG</strong>(UnitPrice) <strong>OVER</strong>(<strong>PARTITION</strong> <strong>BY</strong> InvoiceNo) <strong>AS</strong> AvgUnitPrice,</pre><pre><strong>SUM</strong>(Quantity) <strong>OVER</strong>(<strong>PARTITION</strong> <strong>BY</strong> InvoiceNo) <strong>AS</strong> TotalQuantity</pre><pre><strong>FROM</strong> sqlbank3;</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/504/1*38XpPQ7YcW7feClmKgUs1w.png" /><figcaption>The output of AVG(), SUM()</figcaption></figure><p>This result shows the average unit price<em> (AvgUnitPrice)</em> and total quantity <em>(TotalQuantity)</em> for each order <em>(InvoiceNo).</em></p><h4><strong>2. </strong>AVG(), SUM(), COUNT(), MAX(), MIN()</h4><pre><strong>SELECT</strong> InvoiceNo, StockCode, Description,UnitPrice,</pre><pre><strong>AVG</strong>(UnitPrice) <strong>OVER</strong>(<strong>PARTITION</strong> <strong>BY</strong> InvoiceNo) <strong>AS</strong> AvgUnitPrice,</pre><pre><strong>SUM</strong>(Quantity) <strong>OVER</strong>(<strong>PARTITION</strong> <strong>BY</strong> InvoiceNo) <strong>AS</strong> TotalQuantity,</pre><pre><strong>COUNT</strong>(StockCode) <strong>OVER</strong>(<strong>PARTITION</strong> <strong>BY</strong> InvoiceNo) <strong>AS</strong> Number_Of_Items_Per_Invoice,</pre><pre><strong>MAX</strong>(UnitPrice) <strong>OVER</strong>(<strong>PARTITION</strong> <strong>BY</strong> InvoiceNo) <strong>AS</strong> Max_UnitPrice_Per_Invoice,</pre><pre><strong>MIN</strong>(UnitPrice) <strong>OVER</strong>(<strong>PARTITION</strong> <strong>BY</strong> InvoiceNo) <strong>AS</strong> Min_UnitPrice_Per_Invoice</pre><pre><strong>FROM</strong> sqlbank3;</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*GaCWYBhi4f168vUMctoZlg.png" /><figcaption>The output of AVG(), SUM(), COUNT(), MAX(), MIN()</figcaption></figure><p>This result shows the average unit price <em>(AvgUnitPrice)</em>, total quantity <em>(TotalQuantity),</em> the number of different items per order <em>(Number_Of_Items_Per_Invoice),</em> the maximum unit price of an item <em>(Max_UnitPrice_Per_Invoice),</em> and the minimum unit price of an item <em>(Min_UnitPrice_Per_Invoice)</em> for each order <em>(InvoiceNo).</em></p><h4><strong>3. </strong>ROW_NUMBER()</h4><p><strong>ROW_NUMBER</strong> function assigns a number to each record in the row. We can combine <strong>ORDER</strong> <strong>BY</strong> and <strong>ROW_NUMBER</strong> to determine which column should be used for the row number assignment.</p><pre><strong>SELECT</strong> InvoiceNo, StockCode, Description, UnitPrice,</pre><pre><strong>ROW_NUMBER</strong>() <strong>OVER</strong>(<strong>PARTITION</strong> <strong>BY</strong> InvoiceNo <strong>ORDER</strong> <strong>BY</strong> UnitPrice <strong>DESC</strong> ,StockCode <strong>DESC</strong>) <strong>AS</strong> Rank_of_UnitPrice_Per_Invoice,</pre><pre><strong>COUNT</strong>(StockCode) <strong>OVER</strong>(<strong>PARTITION</strong> <strong>BY</strong> InvoiceNo) <strong>AS</strong> Number_Of_Items</pre><pre><strong>FROM</strong> sqlbank3;</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*XelNSoC0X1OWQvkQ6x94jQ.png" /><figcaption>The output of ROW_NUMBER()</figcaption></figure><p>This result shows the ranking of the items in each order <em>(InvoiceNo)</em> based on the decreasing order of their unit price.</p><h4><strong>4.</strong> ROW_NUMBER() (Top Five)</h4><pre><strong>WITH</strong> CTE <strong>AS</strong> (</pre><pre><strong>SELECT</strong> InvoiceNo, StockCode, Description,UnitPrice,</pre><pre><strong>ROW_NUMBER</strong>() <strong>OVER</strong>(<strong>PARTITION</strong> <strong>BY</strong> InvoiceNo <strong>ORDER</strong> <strong>BY</strong> UnitPrice <strong>DESC</strong> ,StockCode <strong>DESC</strong>) <strong>AS</strong> Rank_of_UnitPrice_Per_Invoice</pre><pre><strong>FROM</strong> sqlbank3)</pre><pre><strong>SELECT</strong> <strong>*</strong> <strong>FROM</strong> CTE</pre><pre><strong>WHERE</strong> Rank_of_UnitPrice_Per_Invoice <strong>&lt;=</strong> 5;</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/820/1*qrGtcnQVF_1Uou5uRZVQtQ.png" /><figcaption>The output of ROW_NUMBER() (Top Five)</figcaption></figure><p>This result shows the top 5 items from each order <em>(InvoiceNo)</em> that have the highest unit price.</p><h4><strong>5. </strong>ROW_NUMBER(), RANK(), DENSE_RANK()</h4><p><strong>RANK</strong> is similar to <strong>ROW_NUMBER</strong> except it will assign the same number to rows with identical values, skipping over the following number.</p><p>There is also <strong>DENSE_RANK</strong> which assigns a number to a row with equal values but doesn’t skip over a number.</p><pre><strong>SELECT</strong> InvoiceNo, StockCode, Description, UnitPrice,</pre><pre><strong>ROW_NUMBER</strong>() <strong>OVER</strong>(<strong>PARTITION</strong> <strong>BY</strong> InvoiceNo <strong>ORDER</strong> <strong>BY</strong> UnitPrice <strong>DESC</strong>) <strong>AS</strong> RowNumber_of_UnitPrice_Per_Invoice,</pre><pre><strong>RANK</strong>() <strong>OVER</strong>(<strong>PARTITION</strong> <strong>BY</strong> InvoiceNo <strong>ORDER</strong> <strong>BY</strong> UnitPrice <strong>DESC</strong>) <strong>AS</strong> Rank_of_UnitPrice_Per_Invoice,</pre><pre><strong>DENSE_RANK</strong>() <strong>OVER</strong>(<strong>PARTITION</strong> <strong>BY</strong> InvoiceNo <strong>ORDER</strong> <strong>BY</strong> UnitPrice <strong>DESC</strong>) <strong>AS</strong> DenseRank_of_UnitPrice_Per_Invoice,</pre><pre><strong>AVG</strong>(UnitPrice) <strong>OVER</strong>(<strong>PARTITION</strong> <strong>BY</strong> InvoiceNo) <strong>AS</strong> AvgUnitPrice,</pre><pre><strong>COUNT</strong>(StockCode) <strong>OVER</strong>(<strong>PARTITION</strong> <strong>BY</strong> InvoiceNo) <strong>AS</strong> Number_Of_Items</pre><pre><strong>FROM</strong> sqlbank3;</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*yqkQQkD99BtWlz_Fmk9ULQ.png" /><figcaption>The output of ROW_NUMBER(), RANK(), DENSE_RANK()</figcaption></figure><p><strong>Notice the 3 different types of columns.</strong></p><ul><li><strong>ROW_NUMBER:</strong> Nothing new here, we are merely adding value for each row in the query.</li><li><strong>RANK:</strong> Here, we give a ranking based on the values but notice we do not have the rank 5. Instead, we have two 4s, and we skip to rank 6.</li><li><strong>DENSE _RANK:</strong> Similar to <strong>RANK</strong> but instead of skipping the rank 5, we include it.</li></ul><h4>6. NTILE()</h4><p><strong>NTILE</strong> function helps to identify what percentile (or quartile, or any other subdivision) a given row falls into. The syntax is <strong><em>NTILE(# of buckets)</em></strong>. In this case, <strong>ORDER BY</strong> determines which column to use to determine the quartiles (or whatever number of tiles you specify).</p><p><strong>NTILE</strong> creates tiles based on the following formula:</p><blockquote><strong>No of rows in each tile = No of rows in result set / No of tiles specified</strong></blockquote><pre><strong>WITH</strong> CTE <strong>AS</strong> (</pre><pre><strong>SELECT</strong> CustomerID, <strong>SUM</strong>(Revenue) <strong>AS</strong> TOTAL_REVENUE</pre><pre><strong>FROM</strong> sqlbank3</pre><pre><strong>GROUP</strong> <strong>BY</strong> CustomerID</pre><pre><strong>ORDER</strong> <strong>BY</strong> CustomerID,TOTAL_REVENUE)</pre><pre><strong>SELECT</strong> CustomerID, TOTAL_REVENUE,</pre><pre>NTILE(4) <strong>OVER</strong>(<strong>ORDER</strong> <strong>BY</strong> TOTAL_REVENUE) <strong>AS</strong> REVENUE_QUARTILES</pre><pre><strong>FROM</strong> CTE;</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/856/1*Cfw-1vvqmYKeI0grpGkHvw.png" /><figcaption>The output of NTILE()</figcaption></figure><p>This result shows that customers with <em>Total_Revenue</em> &lt;= 668.43 fall in Quartile 2 and customers with <em>Total_Revenue</em> &gt;= 668.56 fall into Quartile 3.</p><h4>7. PERCENT_RANK(), CUME_DIST()</h4><p>The <strong>CUME_DIST</strong> function calculates the relative rank of the current row within a window partition using the following formula: (number of rows preceding or peer with current row) / (total rows in the window partition).</p><p>The <strong>PERCENT_RANK </strong>function calculates the percent rank of the current row within a window partition using the following formula: (x — 1) / (number of rows in window partition — 1) where x is the rank of the current row.</p><pre><strong>SELECT</strong> InvoiceNo, StockCode, Description, UnitPrice,</pre><pre><strong>RANK</strong>() <strong>OVER</strong>(<strong>PARTITION</strong> <strong>BY</strong> InvoiceNo <strong>ORDER</strong> <strong>BY</strong> UnitPrice <strong>Desc</strong>) <strong>AS</strong> Rank_of_UnitPrice_Per_Invoice,</pre><pre><strong>COUNT</strong>(StockCode) <strong>OVER</strong>(<strong>PARTITION</strong> <strong>BY</strong> InvoiceNo) <strong>AS</strong> Number_Of_Items,</pre><pre>PERCENT_RANK() <strong>OVER</strong>(<strong>PARTITION</strong> <strong>BY</strong> InvoiceNo <strong>ORDER</strong> <strong>BY</strong> UnitPrice <strong>DESC</strong>) <strong>*</strong> 100   <strong>AS</strong> Percent_Rank,</pre><pre>CUME_DIST() <strong>OVER</strong>(<strong>PARTITION</strong> <strong>BY</strong> InvoiceNo <strong>ORDER</strong> <strong>BY</strong> UnitPrice <strong>DESC</strong>)   <strong>AS</strong> Cume_Dist</pre><pre><strong>FROM</strong> sqlbank3;</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*b3ojl_bSsAdnBj9ZKiCl1A.png" /><figcaption>The output of PERCENT_RANK(), CUME_DIST()</figcaption></figure><h4><strong>8.</strong> LAG(), LEAD()</h4><p><strong>LAG</strong> is a window function that outputs a row that comes before the current row.</p><p><strong>LEAD</strong> is a window function that outputs a row that comes after the current row — essentially the opposite to <strong>LAG</strong>.</p><pre><strong>SELECT</strong> CustomerID, date_trunc(&#39;MONTH&#39;, InvoiceDate) <strong>AS</strong> Month, <strong>SUM</strong>(Revenue) <strong>AS</strong> Monthly_Revenue,</pre><pre><strong>LAG</strong>(Monthly_Revenue, 1) <strong>OVER</strong> (<strong>PARTITION</strong> <strong>BY</strong> CustomerID <strong>ORDER</strong> <strong>BY</strong> Month) <strong>AS</strong> Previous_Month_Revenue,</pre><pre><strong>LEAD</strong>(Monthly_Revenue, 1) <strong>OVER</strong> (<strong>PARTITION</strong> <strong>BY</strong> CustomerID <strong>ORDER</strong> <strong>BY</strong> Month) <strong>AS</strong> Next_Month_Revenue</pre><pre><strong>FROM</strong> sqlbank3</pre><pre><strong>GROUP</strong> <strong>BY</strong> CustomerID, Month</pre><pre><strong>ORDER</strong> <strong>BY</strong> CustomerID, Month;</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/794/1*QPEHXo_0l-eMgKrFWIHr9Q.png" /><figcaption>The output of LAG(), LEAD()</figcaption></figure><p>This result shows the revenue generated from the customer each month <em>(Monthly_Revenue)</em> and the month before <em>(Previous_Month_Revenue)</em> and after <em>(Next_Month_Revenue)</em> that whenever they made the purchase.</p><h4><strong>9.</strong> <strong>PARTITION BY (RUNNING TOTAL)</strong></h4><pre><strong>WITH</strong> CTE <strong>AS</strong> (</pre><pre><strong>SELECT</strong> CustomerID, date_trunc(&#39;MONTH&#39;, InvoiceDate) <strong>AS</strong> Month,</pre><pre><strong>SUM</strong>(Revenue) <strong>AS</strong> Monthly_Revenue</pre><pre><strong>FROM</strong> sqlbank3</pre><pre><strong>GROUP</strong> <strong>BY</strong> CustomerID, Month</pre><pre><strong>ORDER</strong> <strong>BY</strong> CustomerID, Month)</pre><pre><strong>SELECT </strong>CustomerID, Month, Monthly_Revenue,</pre><pre><strong>SUM</strong>(Monthly_Revenue) <strong>OVER</strong> (<strong>PARTITION</strong> <strong>BY</strong> CustomerID <strong>ORDER</strong> <strong>BY</strong> Month <strong>ASC</strong> <strong>ROWS</strong> <strong>BETWEEN</strong> <strong>UNBOUNDED</strong> <strong>PRECEDING</strong> <strong>AND</strong> <strong>CURRENT</strong> <strong>ROW</strong>) <strong>AS</strong> RUNNING_TOTAL_REVENUE_PER_CUSTOMER</pre><pre><strong>FROM</strong> CTE</pre><pre><strong>ORDER</strong> <strong>BY</strong> CustomerID, Month<strong>;</strong></pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/734/1*fNopdc9u-amU9uLTTlr0ug.png" /><figcaption>The output of PARTITION BY (RUNNING TOTAL)</figcaption></figure><p>This result shows the running total of revenue for each customer <em>(RUNNING_TOTAL_REVENUE_PER_CUSTOMER)</em> for the months they have made a purchase.</p><p>Window Functions are very useful for performing data analysis calculations and address an important need compared to the GROUP BY clause.</p><blockquote>Thanks for reading this article. I hope it was useful. If you find any errors or if you have any feedback, please let me know. I would be happy to learn from you.</blockquote><p>We can connect on <a href="https://www.linkedin.com/in/nikitamaloo/"><strong>LinkedIn</strong></a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f4dac0906a55" width="1" height="1" alt=""><hr><p><a href="https://medium.com/swlh/window-functions-for-analytics-in-sql-f4dac0906a55">Window Functions for Analytics in SQL</a> was originally published in <a href="https://medium.com/swlh">The Startup</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>