<?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 Data³ on Medium]]></title>
        <description><![CDATA[Stories by Data³ on Medium]]></description>
        <link>https://medium.com/@DataCubed?source=rss-7f94abe4b137------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*C8RWIadTHV_LWrRI4M91Xg.png</url>
            <title>Stories by Data³ on Medium</title>
            <link>https://medium.com/@DataCubed?source=rss-7f94abe4b137------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 04:04:33 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@DataCubed/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[Optimise Your Business with Machine Learning: How Random Forest models can help you make…]]></title>
            <link>https://medium.com/@DataCubed/optimise-your-business-with-machine-learning-how-random-forest-models-can-help-you-make-be68ee78180?source=rss-7f94abe4b137------2</link>
            <guid isPermaLink="false">https://medium.com/p/be68ee78180</guid>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[predictive-analytics]]></category>
            <category><![CDATA[random-forest]]></category>
            <category><![CDATA[business-optimization]]></category>
            <dc:creator><![CDATA[Data³]]></dc:creator>
            <pubDate>Tue, 27 Aug 2019 14:07:14 GMT</pubDate>
            <atom:updated>2019-08-27T14:10:49.883Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*L4NJILdAoR4lEZac.jpg" /></figure><h3>Optimise Your Business with Machine Learning: How Random Forest models can help you make predictions</h3><p><em>In this article, Dom Owens explains how Random Forest models work, and how machine learning can generate value for your business.</em></p><h3>Rise of the machines</h3><p>If you could predict something about your business, what would it be?</p><p>In recent years, there’s been an explosion in computing power. At the same time, there’s been a vast increase in the amount of data being collected. Combined, this gives businesses the potential to use machine learning techniques to pick up on patterns hidden inside data. These patterns can then help us make predictions — valuable information for a business in rapidly-changing industry.</p><p>Think of any piece of information that you’d really like to know but don’t yet. How long it takes for a subscriber to unsubscribe, the value a new customer might bring to your business over their lifetime, the sales figures of a new offer you’re thinking of launching. All of these are valuable pieces of information when you’re trying to plan for the future.</p><p>Machine learning techniques can not only help you make predictions, but can also tell you how accurate that prediction is and how the other information you have is influencing it.</p><h3>Exploring the Random Forest</h3><p>Suppose you have a large amount of data with lots of measurements and you want to know one of two things: a number (say, the number of units a product will sell) or a class (the newspaper that a customer will buy).</p><p>The Random Forest algorithm — a set of rules for making decisions — can handle this situation elegantly, giving accurate and unbiased predictions.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/690/0*Ry4NWdoTXjSjMfrE" /><figcaption>An example of an informal decision tree</figcaption></figure><p>Let’s start from the roots. You’re probably familiar with the concept of a decision tree as a way to find the answer to a problem. We ask a question that has two possible answers — often yes or no — and follow the answer to another question. We repeat the process until we get a final answer that we’re satisfied with.</p><p>The Random Forest algorithm does the same thing. It asks a question of the data to split the current branch of the tree. It selects the best question to ask from a small group of questions, opting for the question that splits the data most evenly.</p><p>This gives us a tool for making predictions. A new set of data can be fed through the decision tree, outputting a prediction for each new observation. But — a single decision tree can veer far off course if it gets caught on irregularities in the data, leading to less accurate predictions.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/860/0*VGfapKx2uQKiA4jy" /><figcaption>Voting average</figcaption></figure><p>Thankfully, we can remedy this by using the wisdom of crowds. We choose a large subset of the original data and create a tree to fit it. Then, with a new subset each time, we fit many more trees, each capable of making predictions for new data. Since each tree fits different data, and the trees split according to random rules, each tree will be unique and give different predictions for the same data.</p><p>We then take an average of all the predictions. This gives us a more accurate prediction than one tree alone, either the most popular vote for classes, or the mean average for a predicted number.</p><p>We can even see which factors are important in influencing how the trees make their decisions. Perhaps product sales depend heavily on the weather conditions and your website traffic, but less so on how much is spent on marketing the product.</p><h3>Let’s look at an example</h3><p>Let’s consider a telecoms business using <a href="https://www.kaggle.com/blastchar/telco-customer-churn">this example dataset</a>, and see if we can predict how new customers will behave. The dataset contains information on previous customers, such as their gender, the type of contract they hold, and the charges they’ve incurred, as well as whether they’re no longer a customer. Being able to predict customer churn can be very useful for making decisions about pricing and marketing.</p><p>The Error Plot tells us that, as we fit more trees, the model becomes more accurate. This shows us that taking the average of multiple trees is really valuable.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*nLIkK0kqj1g8m9Oz" /></figure><p>The Variable Importance plot shows us how important the information we’re measuring is. Larger readings correspond to more importance. We can see that charges and type of contract are most important, while the customer’s gender and whether they have a partner are least important.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*X9tlpdVvZnYw2kxC" /></figure><p>When we try out the model on new data, the model gets predictions right around 80% of the time. It does well at identifying customers who don’t churn, but less well at identifying customers who do. Given that 73% of the customers in the data set did churn, this is a useful improvement on randomly guessing if the customer churns based on population percentages. We might improve on this by collecting more data, recording more information for each customer, or by adjusting the settings of the model algorithm.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/363/0*z5ZBaydWsA1JfJQB" /><figcaption><em>The test set error rate tells us the model gets it right nearly 80% of the time</em>.</figcaption></figure><p>Many other problems can be solved in a similar way, which brings us back to our original question — if you could predict something about your business, what would it be?</p><p>We’ve developed a new predictive modelling tool using Random Forest algorithms, which we’re using to help our clients peer into the future. Our new tool allows us to identify your most valuable clients now, and then lets us go one step further to predict which clients will be the most valuable in the future. We can predict other things too, such as customer churn, or sales figures for a new offer. And we can show you the results within days.</p><p><strong>We’re really excited about this technology. If you are too and would like to know more, email us at </strong><a href="/service/mailto:hello@data-cubed.co.uk"><strong>hello@data-cubed.co.uk</strong></a><strong> or call us on 0117 25 10 100.</strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=be68ee78180" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[‘Apocalyptic volcanic super eruption that could DESTROY civilisation is much closer than we…]]></title>
            <link>https://medium.com/@DataCubed/apocalyptic-volcanic-super-eruption-that-could-destroy-civilisation-is-much-closer-than-we-b7dcfaec9db8?source=rss-7f94abe4b137------2</link>
            <guid isPermaLink="false">https://medium.com/p/b7dcfaec9db8</guid>
            <category><![CDATA[science]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[supervolcano]]></category>
            <category><![CDATA[statistics]]></category>
            <dc:creator><![CDATA[Data³]]></dc:creator>
            <pubDate>Tue, 27 Aug 2019 14:03:30 GMT</pubDate>
            <atom:updated>2019-08-27T14:10:20.008Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/900/0*DjNVpmTwsIJKtblm.jpg" /></figure><h3>‘Apocalyptic volcanic super eruption that could DESTROY civilisation is much closer than we thought’, say experts</h3><p><em>Data can give us some truly fascinating insights into the world around us. In this article, Jonathan Rougier tells us about some of his work researching explosive volcanism.</em></p><p>The interesting thing about being a statistician is that you get to play in everyone’s backyard, as John Tukey famously said (famously if you’re a statistician!). Over my career so far I have covered topics from avalanches (Swiss Federal Institute for Snow and Avalanche Research) to extreme weather early warning systems (UK Met Office). I’ve even helped to create <a href="https://play.google.com/store/apps/details?id=com.isoperla.birdid&amp;hl=en_US">a phone app to identify birds from their songs</a>, AKA “Shazam for birds”.</p><p>One of the most interesting topics has been explosive volcanism, working with scientists in the School of Earth Sciences at the University of Bristol. They have compiled the definitive database of large explosive eruptions, LaMEVE. This database (held as a spreadsheet) represents many thousands of hours of fieldwork, and also a lot of theory, about how to invert geological traces of eruptions like ash layers back to an assessment of eruption magnitude.</p><p>One of our papers had some eye-catching results concerning ‘super-eruptions’, as reported by the Huffington Post in their article, <a href="https://www.huffingtonpost.co.uk/entry/this-is-what-super-volcanoes-mean-for-the-future-of-humanity_uk_5a6619fae4b0dc592a0b30d1?guccounter=1&amp;guce_referrer=aHR0cHM6Ly9kdWNrZHVja2dvLmNvbS8&amp;guce_referrer_sig=AQAAAEPNDTyE2T1TrpM4yoM1Q0GPF6jefk-YvsrH2YZtDwXyI079y0XkxuCHSbXE3oq7NhTiAvklKQ4AL9uuBcTZOYSX9Lf6YuAeDbI45Wdaky85AE9FaCTIP9tjG3ggU-Eqn4vYbq9jwCHbGIzhLCaXy_mXrZIw480EmrZqyVEZFZM0">‘This is what super volcanoes mean for the future of humanity’</a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/964/0*aYHiorM5MDOZCEuV.png" /><figcaption>Our paper ‘<a href="https://www.sciencedirect.com/science/article/pii/S0012821X17306519">The global magnitude–frequency relationship for large explosive volcanic eruptions</a>‘ had some eye-catching results concerning ‘super-eruptions’.</figcaption></figure><p>A super-eruption is one of the very few natural hazards which has the potential to end our civilisation, or at least put a very large dent in it (the other is an asteroid strike). There have been 3 in the last 100 thousand years, with the largest, <a href="https://earthobservatory.nasa.gov/images/7380/toba-caldera">Toba (about 75 thousand years ago)</a> being the largest explosive eruption ever recorded.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*m5eW3ULVchc_jyGk.png" /><figcaption>The Toba caldera — the site of a massive super-eruption 75,000 years ago. <br><em>Image credit: NASA/METI/AIST/Japan Space Systems, and U.S./Japan ASTER Science Team</em></figcaption></figure><p>Using a careful treatment of the records in the LaMEVE database, we found that the <a href="http://bristol.ac.uk/news/2017/november/volcanic-super-eruptions.html">return period of super-eruptions is about 20 thousand years</a> — much shorter than previously thought.</p><blockquote>“Volcanologist Who Shifted Date of Next Super-Eruption Is Surprisingly Chill.”</blockquote><p>I think the first outlet to publish was End Times Headlines, for which civilisation-ending volcanic eruptions are catnip. The Daily Mail followed shortly after, with ‘Apocalyptic volcanic super eruption that could DESTROY civilisation is much closer than we thought, say experts’. The Sun referred to me as a ‘boffin’.</p><p>My favourite article was, ‘<a href="https://www.inverse.com/article/38852-supervolcanoes-volcano-eruption-earth-kill-humans">Volcanologist Who Shifted Date of Next Super-Eruption Is Surprisingly Chill</a>,’ in the online magazine Inverse.</p><p>Peter Hess, who wrote the article, had contacted me before publication with several interesting questions, and he picked up on one of the messages in the press release, that super-eruptions show that our tenure as a civilisation here on earth is quite precarious.</p><p>The reason that I was ‘surprisingly chill’ about super-eruptions was, as he put it, “Something else will probably kill us first, seeing as the global climate is warming, antibiotic-resistant bacteria are thriving, and nuclear war is a greater threat than ever.” He might also have mentioned flu-like pandemic or a massive solar storm.</p><p>In talks, a popular question is where the next super-eruption will be. This is not really predictable, but if I had to pick one location then it would be Yellowstone in North America, which has had six very large eruptions in the last 1.3 million years, including two super-eruptions.</p><p>Another question is whether there will be any warning: my colleagues tell me not necessarily. If you had a few days warning and knew where the eruption was going to be, you could improve your chances by moving to the opposite side of the earth, although maybe not by very much unless you were already a skilled survivalist.</p><p>— — -</p><p>Jonty Rougier is a ‘full spectrum’ applied statistician. The largest part of his academic work has been in Earth and Environmental Sciences. He’s worked with several UK Government departments and agencies, and also with non-profit organisations and SMEs. You can find out more about Jonty at <a href="https://www.rougierconsulting.com/">www.rougierconsulting.com</a>.</p><p><strong>If you’d like to hear more about Jonty’s work with Data Cubed and how we’re helping clients delve further into their own data, please email us at </strong><a href="/service/mailto:hello@data-cubed.co.uk"><strong>hello@data-cubed.co.uk</strong></a><strong>.</strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b7dcfaec9db8" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[3 Steps to Great Data Visualisations]]></title>
            <link>https://medium.com/@DataCubed/3-steps-to-great-data-visualisations-1c647c29fb1a?source=rss-7f94abe4b137------2</link>
            <guid isPermaLink="false">https://medium.com/p/1c647c29fb1a</guid>
            <category><![CDATA[data-visualization]]></category>
            <dc:creator><![CDATA[Data³]]></dc:creator>
            <pubDate>Tue, 27 Aug 2019 13:58:27 GMT</pubDate>
            <atom:updated>2019-08-27T14:09:34.546Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*BigCkD2eWE4qTDHE.jpg" /></figure><p><a href="http://www.bmdata.co.uk/"><em>Bryony Miles</em></a><em> is our d3.js data viz expert. She helps us bring your data to life, developing bespoke charts and dashboards that are both insightful and beautiful.</em></p><p>In this article, Bryony shares her expertise around creating data visualisations — and the three things you need to consider.</p><h3>1. Know your goal</h3><p>Data visualisations are either explanatory or exploratory. Do you want to explain something to your audience or let them explore the data?</p><p><strong>Explanatory visualisations</strong> aim to tell a story. For example — did you ever wonder how many words members of the Fellowship spoke during all 3 extended editions of Lord of the Rings? This data visualisation tells us using the explanatory approach: <a href="http://www.datasketch.es/july/code/nadieh/">Who’s speaking in Middle Earth</a>?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/911/0*DZuY5qtp89GdttUo" /><figcaption>An example of the explanatory approach: <a href="http://vimeo.com/316625127">Who’s speaking in Middle Earth?</a></figcaption></figure><p>The <strong>exploratory approach</strong> uses charts to help us explore the data and discover insight. An example is <a href="http://vimeo.com/316625127">Stockmapper</a> — a visually dynamic map of 1800 stocks.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/893/0*bx3n8jwYxM_GX7tl" /><figcaption><a href="http://vimeo.com/316625127">Stockmapper</a> uses the exploratory approach</figcaption></figure><p>Data visualisations are normally explanatory or exploratory. But you can also opt for the <strong>“Martini Glass” approach</strong>, which starts by telling a story and then allows the user to dive in further. In <a href="http://rhythm-of-food.net/">The Rhythm of Food</a>, Google News Lab and Truth &amp; Beauty use the Martini glass approach to tell a story about key food trends over the years and allow the user to explore the results.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/901/0*cAyZNgJEnO_01D7x" /><figcaption>An example of the Martini Glass approach: <a href="http://rhythm-of-food.net/">The Rhythm of Food</a></figcaption></figure><h3>2. Choose the right charts</h3><p>Choosing the right charts is crucial. A good start is to look at your data and ask yourself a some questions:</p><ul><li>Is the data over a period of time?</li><li>Is it ordinal (numerical) or categorical?</li><li>How many periods or categories are there — a few or lots?</li><li>Are you looking at relationships, comparisons, distribution and/or composition?</li></ul><p>The <a href="https://extremepresentation.com/wp-content/uploads/choosing-a-good-chart-09-1.pdf">Extreme Presentation ‘Thought Starter’</a> is a nice place to start when deciding on which charts to use, but talking through your individual needs will lead to your unique solution.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*dqSd8v6CKhKg5gK7" /><figcaption><a href="https://extremepresentation.com/wp-content/uploads/choosing-a-good-chart-09-1.pdf">Chart Suggestions — A Thought Starter</a></figcaption></figure><h3>3. Choose your software</h3><p>Tableau and Power BI are the market leading Business Intelligence tools. They’re easy to learn and enable you to quickly analyse data and develop insightful dashboards and charts. But there’s a third piece of software that’s making an impact — d3.js or ‘D3’.</p><p>D3 is a super flexible Javascript visualisation library. There are some initial development costs, then it’s free. Its key selling points include:</p><ul><li>a wide range of charts</li><li>best practice examples</li><li>no monthly fees</li><li>the ability to customise the look and feel and give your site that extra “WOW” factor.</li></ul><h4>Which data visualisation software should you be using?</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/219/0*xMWs5v7uh6MyVFJD.png" /></figure><p><strong>CLIENT 1</strong> uses Tableau or Power BI to analyse and clean data and generate new dashboards and reports. The chart type, data format and layout of these reports is changing on a daily, weekly and monthly basis. Flexibility is key.</p><p><strong>VERDICT:</strong> We’d recommend sticking with Tableau and Power BI.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/219/0*Q9RHWXr-Vv5L0lqR.png" /></figure><p><strong>CLIENT 2</strong> discovered Tableau and Power BI, and it was a revelation. They were able to explore data in new and exciting ways and developed key dashboards and charts, which they refer to on a regular basis.</p><p><strong>VERDICT</strong>: This client could make some real savings with D3, which will also allow them to make their dashboards work exactly as they want.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/219/0*yEx7RtT-1IkfMW5M.png" /></figure><p><strong>CLIENT 3</strong> hasn’t invested in any data visualisation tools yet. They’re looking for the best solution for their company.</p><p><strong>VERDICT:</strong> There are lots of things to consider when choosing a data viz tool. We’re here to help. We’ll look at your requirements in more detail and advise whether you’d be better off with a custom or off-the-shelf solution.</p><p><strong>Not sure which data viz software to use, or keen to learn more about D3? Email us at </strong><a href="/service/mailto:hello@data-cubed.co.uk?subject=Free%20Data%20Opportunity%20Score"><strong>hello@data-cubed.co.uk</strong></a><strong> or give us a call on 0117 25 10 100.</strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1c647c29fb1a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Why Every HR Team Should Hire a Data Translator]]></title>
            <link>https://medium.com/@DataCubed/why-every-hr-team-should-hire-a-data-translator-d87dc44fb6d8?source=rss-7f94abe4b137------2</link>
            <guid isPermaLink="false">https://medium.com/p/d87dc44fb6d8</guid>
            <category><![CDATA[human-resources]]></category>
            <category><![CDATA[hr-data-analytics]]></category>
            <category><![CDATA[retention]]></category>
            <category><![CDATA[performance]]></category>
            <category><![CDATA[recruitment]]></category>
            <dc:creator><![CDATA[Data³]]></dc:creator>
            <pubDate>Thu, 09 May 2019 09:43:53 GMT</pubDate>
            <atom:updated>2019-05-09T09:43:53.884Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*uXReRq6bj50Y0xIhTm9PYg.jpeg" /></figure><h3>HR is up against it</h3><p>Businesses don’t build themselves; people build them. That’s why HR teams have such a tall order. They need to hire in the right people and then build effective long-term strategies to engage them, motivate them and ultimately drive business success. Measuring all of these tasks for effectiveness has been a long-standing challenge for HR. Data is often spread across the business — in recruitment, HR, sales and finance. And beyond siloed teams, getting actionable insight from HR data is difficult in two parts.</p><p>Firstly, HR has not always had the analytical expertise needed. In teams like Marketing and Sales, data analytics has long formed the backbone of strategy. It’s used for creating new products, testing marketing channels, informing pricing strategies and of course sales targeting. In HR, the merits of data analytics are less known. The technology has been slow to adopt, and specialist roles for analysing and visualising data are rare.</p><p>And this leads to the second challenge: HR doesn’t have a reputation of measuring success, playing a numbers game, or bringing in the big bucks. Again, these go to finance, marketing, and sales. Whilst this is fast changing, some teams need support to turn insights into action; to communicate findings to the executive decision makers and get them to act on them.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*8c333d_YNEHG4q3UDb1wTA.jpeg" /></figure><h3>Turning insight into action</h3><p>Whilst data can solve complex business challenges, it’s worthless if it’s not analysed or communicated to decision-makers in the right way. That’s where a great data translator comes in. A data translator is the person that sits between the data analysts and the decision makers. They must ‘talk the talk’ of both parties, by understanding the business needs of the organisation and at the same time being data-savvy enough to talk tech and distil it to others in an easy-to-understand manner.</p><p>The data translator takes an important commercial role. Working closely with the HR team, they must challenge all HR data and ask:</p><ul><li>How can we use this data to make money?</li><li>How can we use data to save money?</li><li>How can we use data to create a competitive advantage?</li><li>How can we use data to innovate?</li></ul><p>Without this perspective, the project risks being a waste of time where the end result could offer no demonstrable value to the business performance or offer no actionable insight. The data translator brings the ‘so what?’.</p><p>After all, many data types prefer the independence of wrangling with data rather than explaining to non-tech people the implications of the data. They would rather crunch numbers than be pulled into exec meetings to defend the data insights or why the data should be trusted. Likewise, executives can be dismissive of data that can impact those who are responsible for it — they like to know they’re in control, but can feel quite uncomfortable when they’re presented with data they don’t fully understand.</p><p>Since data translators understand the core business objectives of an organisation, they can identify business actions based on the findings of the data that neither the data scientist or executive can. They can ensure that the data insights add value to the business.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*VjSB70ybi0qvH-qYxuGhnQ.jpeg" /></figure><h3>The merits of HR Data Analytics</h3><p>For the businesses that embrace HR analytics — and find the right translators — the rewards are boundless. Using intelligent data tools, HR teams can recruit, retain and develop talent better than ever before. Imagine if you could:</p><ul><li>Recruit talent faster and more effectively by tracking recruitment timescales, costs, and drop-out rates</li><li>Boost retention by mapping staff turnover and showing areas with current and future skills shortages</li><li>Improve performance by flagging performance trends against reward and L&amp;D programs</li><li>Identify future trends in areas like gender, diversity and wellbeing by highlighting changes over time</li></ul><p>By having the right insight, and being armed with the right commercial objectives, HR data can inform every area of people strategy, including:</p><p><strong>Recruitment decisions — </strong>by understanding past and future trends on time and cost to recruit roles</p><p><strong>Performance plans </strong>— by identifying performance hot spots and problem areas against L&amp;D activity</p><p><strong>Retention activities </strong>— by targeting management training and team-building in the problem areas</p><p><strong>Reward schemes </strong>— by understanding which schemes are attracting/retaining talent and boosting performance the best</p><p><strong>Diversity &amp; inclusion strategies </strong>— by exploring how diversity varies across location, department and job role.</p><h3>Want to translate your HR data into commercial success?</h3><p>At Data³, we build HR data tools and dashboards from scratch, helping turn your people data into actionable HR strategies. To find out more about the work that we’ve done and how we can help you, visit <a href="http://www.data-cubed.co.uk/">www.data-cubed.co.uk</a> or email us at <a href="/service/mailto:hello@data-cubed.co.uk">hello@data-cubed.co.uk</a>.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d87dc44fb6d8" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[From Soft to Serious: the Changing Face of HR]]></title>
            <link>https://medium.com/@DataCubed/from-soft-to-serious-the-changing-face-of-hr-cc6a8615aee8?source=rss-7f94abe4b137------2</link>
            <guid isPermaLink="false">https://medium.com/p/cc6a8615aee8</guid>
            <category><![CDATA[diversity]]></category>
            <category><![CDATA[human-resources]]></category>
            <category><![CDATA[recruiting]]></category>
            <category><![CDATA[performance]]></category>
            <category><![CDATA[rewards]]></category>
            <dc:creator><![CDATA[Data³]]></dc:creator>
            <pubDate>Tue, 30 Apr 2019 09:16:56 GMT</pubDate>
            <atom:updated>2019-04-30T09:16:56.876Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*E9t7pkiex7906njNAF3cSQ.jpeg" /></figure><p>In every organisation, the Human Resource department has always been up against it. They have a significant role in recruiting the best talent in the market, retaining them, developing them, motivating them and generally driving business performance across the board. It’s no small order. But it’s fair to say that HR teams don’t always get the praise they deserve. They’re often an unpopular department, and this is for two reasons.</p><p>Firstly, they’re the bearer of bad news. If you’re up for a disciplinary: HR. A tricky performance review: HR. Unsuccessful in a job interview: HR. You need to be let go: you get it. HR do the jobs other departments can’t. They provide a framework and a methodology that keeps businesses safe and employee practices fair, but the job is a hard one.</p><p>Secondly, HR teams don’t have a reputation for measuring success. They’re not number crunchers or sales orientated. They’re sometimes regarded as soft. Fluffy. Taking a human approach over a mechanic, results-based one.</p><p>But these old-fashioned perceptions are changing. HR departments the world over are going through a radical change: and one thing sits at the centre of it all: data.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*b3Fexwa4trN8IGOiu5S1dw.jpeg" /></figure><h3><strong>HR analytics</strong></h3><p>HR analytics is about using data to solve an organisation’s people problems. It’s about taking standard HR processes like talent management, benefits and training and assigning value to it all; using relevant, accurate metrics to track the effectiveness and impact of everything the department does.</p><p>By looking at data in a holistic way, HR teams get a better idea of how the business is operating, and can then make more informed, strategic decisions. As an example, let’s take some common metrics. A typical HR team can tell you the annual employee turnover rate of each department. Or the total spent on reward. Or how many training programs a team has run. But bring in data from beyond HRM systems, from finance teams, or sales departments, and the HR function tells a fuller story. It can tell the business how effective HR activities are, and how in turn this affects business performance.</p><p>Analytics in HR (<a href="https://www.analyticsinhr.com/">https://www.analyticsinhr.com</a>) summarise this value chain well. By going beyond descriptive metrics to looking at impacts of activities, HR move from a cost centre to a value-adding function.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/602/1*8k6FQkIGLh5oBOnBXU6whQ.png" /><figcaption><em>Source: AIHR Blog &amp; Academy</em></figcaption></figure><h3><strong>Answering the exciting questions</strong></h3><p>This is what we do as a data lab. We look beyond the dull, vanity metrics and dig into the money makers; the metrics that show how and why a HR department is worth the budget.</p><p>By looking at data across the business — we can show insight in many areas. And in ways that take HR from soft to serious:</p><p><strong>Recruitment</strong></p><ul><li>Which department takes the longest time to fill roles?</li><li>Which roles have the highest recruitment costs and dropout rates?</li></ul><p><strong>Performance</strong></p><ul><li>Does reward affect performance levels?</li><li>How does performance vary by department?</li></ul><p><strong>Engagement</strong></p><ul><li>How does employee satisfaction relate to performance?</li><li>How does benefit participation affect performance?</li></ul><p><strong>Diversity</strong></p><ul><li>Is there a gender pay and bonus gap? Does this differ by role level or department?</li><li>Do you have a disproportionate number of employees of a certain age or gender? Does this differ by role level or department?</li></ul><p><strong>Reward</strong></p><ul><li>Are employee benefit schemes retaining the highest performing employees?</li><li>Is there a correlation between employee satisfaction and employee benefits?</li></ul><p><strong>Learning and development</strong></p><ul><li>Which department spends the most on L&amp;D?</li><li>How do promotional trends vary across your business?</li></ul><p><strong>Retention</strong></p><ul><li>Do some departments retain employees better than others?</li><li>Is the business retaining the highest performing employees?</li></ul><p><strong>Travel &amp; Expenses</strong></p><ul><li>Which department/team spends the most on travel?</li><li>How have travel &amp; expenses costs changed over time?</li></ul><p>To find out more about what we do, <a href="http://www.data-cubed.co.uk/">take a look at our website</a> and get in touch.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=cc6a8615aee8" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Power of People Data]]></title>
            <link>https://medium.com/@DataCubed/the-power-of-people-data-d14244e8c178?source=rss-7f94abe4b137------2</link>
            <guid isPermaLink="false">https://medium.com/p/d14244e8c178</guid>
            <category><![CDATA[human-resources]]></category>
            <category><![CDATA[performance]]></category>
            <category><![CDATA[hr-data-analytics]]></category>
            <category><![CDATA[recruiting]]></category>
            <dc:creator><![CDATA[Data³]]></dc:creator>
            <pubDate>Fri, 26 Apr 2019 14:49:17 GMT</pubDate>
            <atom:updated>2019-04-29T09:37:44.590Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*7-Vd2V6yvsrRROavIg5RFg.jpeg" /></figure><blockquote>How the right metrics can transform a business</blockquote><p>HR teams are sat on powerful people data that can drive change in business. This data is so powerful that it can directly impact profits, market value, market share, productivity, company reputation… you name it.</p><p>The trouble is, HR teams have often focused on metrics that simply describe how a business is performing. Or efficiency targets that only show a business’ people costs. Whilst these metrics are nice to know — they won’t always drive change or exploit the value of the department’s data.</p><p>HR teams are stuck in a data rut and it’s often because:</p><ul><li>HR data is siloed</li><li>Teams have always reported in the same way</li><li>Creating meaningful metrics is really tricky</li></ul><p>But it doesn’t have to be this way. Today, we provide two examples which show the merits of mixing multiple data sets. By looking at data in a more holistic way, we show you how to draw out more insight and impact your bottom line far easier.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*s57ZYgW6PIDcmv8c-ZRKsQ.jpeg" /></figure><p><strong>RECRUITMENT EXAMPLE</strong></p><p>Recruitment Teams often measure performance using:</p><ul><li>Time to recruit per role (e.g. 10 weeks on average)</li><li>Cost to recruit per role/s (e.g. £2000 per employee)</li><li>Candidate drop-out rate (e.g. 20% drop-out from interview to join date)</li></ul><p>Whilst these measures are important to track tactical recruitment activity, they are all short-term measures of success only…they only measure the process before joining a business.</p><p>Imagine if your recruitment teams could demonstrate that new recruits in the last two years have had:</p><ul><li>25% lower absenteeism than the average employee in their business</li><li>30% higher engagement in staff surveys and learning &amp; development activity</li><li>15% higher performance ratings</li><li>40% higher participation in employee benefit schemes</li><li>2% attrition</li></ul><p>These longer-term metrics would be a far better way of assessing the success of a Recruitment Team within a business. Using these metrics, the business can properly assess their recruitment strategy and processes…</p><ul><li>Are you really recruiting the right people?</li><li>Are you driving continuous improvements in the performance and engagement of staff?</li><li>Are you recruiting people who stay within the business and add value?</li></ul><p>Knowing these longer-term metrics, you can assess the success of your entire recruitment process from the copy on the job advert, to your website career pages, to your sift and selection process to the format of your interviews and all the way to your onboarding and induction process.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*SmgY1m4HClo3w4bHJjPBCw.jpeg" /></figure><p><strong>REWARD EXAMPLE</strong></p><p>Reward Teams often measure performance using:</p><ul><li>Workplace pension participation rate</li><li>Employee share plan participation rate</li><li>Employee benefit participation rate</li></ul><p>Whilst these measures are important to track tactical employee benefit activity, they are all short term measures of success only…they only measure the reward part of the process.</p><p>Imagine if your Reward Teams could demonstrate that employee benefit participants have:</p><ul><li>30% lower absenteeism than the average employee in their business</li><li>25% higher employee satisfaction</li><li>10% higher performance ratings</li><li>20% higher retention</li><li>15% higher learning &amp; development pass rates</li></ul><p>These business-wide metrics would be a far better way of assessing the success, or not, of a Reward and Employee Benefit programme within a business. Using these metrics, the business can properly assess their reward strategy and processes…</p><ul><li>Are your reward programmes really attracting, motivating &amp; retaining your talent?</li><li>Are your employee benefit programmes helping you achieve your business goals?</li><li>Or are they counter-productive and helping you to retain under-performers?</li></ul><p>Knowing these business-wide metrics, you can make data-informed decisions about your reward and employee benefit activities and ensure that help you to support your business goals…and they’re not just a cost for your business.</p><p><strong>Data</strong>³<strong> helps businesses to create meaningful HR metrics that really track performance so you can make quicker, smarter, data-driven people decisions. Can we help you recruit, develop and retain talent better?</strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d14244e8c178" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Data is transforming our world]]></title>
            <link>https://medium.com/@DataCubed/data-is-transforming-our-world-d46303f1963a?source=rss-7f94abe4b137------2</link>
            <guid isPermaLink="false">https://medium.com/p/d46303f1963a</guid>
            <category><![CDATA[predictive-analytics]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[big-data]]></category>
            <category><![CDATA[internet-of-things]]></category>
            <dc:creator><![CDATA[Data³]]></dc:creator>
            <pubDate>Wed, 17 Apr 2019 09:49:11 GMT</pubDate>
            <atom:updated>2019-04-17T09:49:11.085Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*IGmGz42jcR2Eodj3sMlL3A.jpeg" /></figure><blockquote><strong>Our run down of data driven technologies and how they’re spreading to the masses.</strong></blockquote><p>Machine Learning, Predictive Analytics, and the Internet of Things. Simply buzzwords in corporate meeting rooms a few years back; these technologies are now shaking our world and transforming the way we do business. As a company that helps businesses with their data every day, we see these technologies revolutionising the way companies make and save money. Whether that’s helping you get closer to your customers, predicting future challenges in your workforce, or modelling different scenarios to improve your business performance, data will have an increasingly large part to play in our business decisions.</p><p>We’ve outlined three technologies here, showing how and why they’re increasing in popularity.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*i384H_6SMsFYMMH4aPrZfw.jpeg" /></figure><p><strong>Rise of the machines: machine learning takes over</strong></p><p>In the past, we solved most<strong> </strong>business problems by processing data with logic, not learning from the data itself via algorithms. But machine learning is changing all that — significantly speeding up the time it takes to analyse huge sets of data and generate granular, accurate insight.</p><p>At its most basic — machine learning is the job of making assumptions, testing those assumptions and learning as you go — given to a computer. It means data is able to be tested and retested at speeds and capabilities no human can. It means trends can be analysed and predictions forecasted at scales we’ve never seen before.</p><p>In the past, this technology was reserved for the big corporates, the fat cats who could buy and process huge datasets. The ones that could absorb the high costs associated with machine learning software, and hire the experienced scientists able to use it. Today — things are different. Machine learning is becoming mainstream and accessible even to startups. This is down to the rapid expansion of data sources available to us, significantly improved algorithms that study the data, and substantially more-powerful computing power.</p><p>As time goes on, businesses will increasingly incorporate machine learning into their data analytics capabilities. With the rise of self-serve platforms like Tableau and Power BI, this adoption is extending to those without the expertise and budgets needed in the past.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*PrEXQDwTE0eeg0ZvcVhXPg.jpeg" /></figure><p><strong>The data crystal ball: predictive models are in high demand</strong></p><p>In a similar vein to the adoption of machine learning, predictive analytics is moving from the bench to the forefront of companies’ business strategies. Simply put; predictive analytics is the process of identifying patterns in past data and using it to estimate the values for data that you do not have. With the rise of increasingly sophisticated modelling software — businesses with relatively low budgets and experience can create accurate, insightful predictions about their future. Whether that’s future customer purchasing behaviour, marketing and sales costs or campaign success — businesses are using prediction models to inform their strategies. This marks a general trend to become insight-driven, over the traditional gut-driven approach that is fast dying out.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*qUoyEaZgru5tk2Fibybkrw.jpeg" /></figure><p><strong>A more connected world: IoT is inevitable</strong>.</p><p>International Data Corporation (IDC) estimates that our daily interactions with digital devices will grow 20-fold over the next 10 years. And it’s not hard to see why. Every day our exposure to digital homes, appliances, vehicles and wearable and implantable devices is growing. What is exciting (and a little scary) is that the data from these devices are aggregated and stored online, as more and more devices are connected to the internet. This Internet of Things (IoT) means our lives are becoming increasingly influenced and dependent on data. In fact, IDC reports that by 2025, 25% of data will be real-time. And of this data, 95% of it will be driven by IoT.</p><p>For businesses, this spells big bucks. According to Vodafone’s annual IoT report, organisations that adopt IoT report an average increase in revenue of 19%. And where they reported a reduction in costs, the average was 16%.</p><p>For businesses everywhere, large and small, the adoption of smart data-driven technologies is becoming increasingly necessary. Those that fail to exploit the resources available to them in our increasingly digital world see themselves fall against their competitors and lose face with their customers. At Data³ we make it easy for our clients to get ahead of the curve. As well as building data tools from scratch, we work on clients’ existing data projects and also upskill data teams so they can run sophisticated reporting with ease. To find out more of what we do, visit our site at <a href="http://www.data-cubed.co.uk/">www.data-cubed.co.uk</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d46303f1963a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Don’t you wish your Excel was just like me: Modern BI platform reviews]]></title>
            <link>https://medium.com/@DataCubed/dont-you-wish-your-excel-was-just-like-me-modern-bi-platform-reviews-9046a1870024?source=rss-7f94abe4b137------2</link>
            <guid isPermaLink="false">https://medium.com/p/9046a1870024</guid>
            <category><![CDATA[tableau]]></category>
            <category><![CDATA[data-visualization]]></category>
            <category><![CDATA[power-bi]]></category>
            <category><![CDATA[analytics]]></category>
            <category><![CDATA[qlik]]></category>
            <dc:creator><![CDATA[Data³]]></dc:creator>
            <pubDate>Thu, 04 Apr 2019 15:08:08 GMT</pubDate>
            <atom:updated>2019-04-04T15:08:08.764Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Y-K4bpH1IXEua6hcshfESw.jpeg" /></figure><blockquote>Power BI, Qlik and Tableau under the scope</blockquote><p>Modern analytics and BI platforms are now mainstream. For companies big and small, MS excel no longer cuts the mustard when it comes to transforming data into engaging, insight-rich visualisations.</p><p>So what are they? Quite simply, analytics and BI platforms are easy-to-use software that supports the full analytic workflow — from data prepping and ingestion to visual exploration and insight generation. What sets them apart from BI platforms of old is that they don’t require hordes of IT staff to predefine data models or store data in traditional data warehouses. Self-service and agility is the name of the game, and providers are springing up everywhere to get a piece of the pie.</p><p>As the market grows and options become plentiful, it can be difficult to know which software to choose. To lend a hand, we’ve been through the latest Gartner review of the main players, picking out three that we believe are worth your consideration.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/858/1*XXkUMUSU3UZuTDhSDUxliw.png" /></figure><p>Power BI offers data preparation, interactive data discovery, and advanced analytical options, all in one product. It’s available as a SaaS option running in the Azure cloud or as an on-premises option in Power BI Report Server.</p><p>In the analytics space, Microsoft is making a name for itself as a true Leader. It has a comprehensive and visionary product roadmap and is focused on making Power BI accessible for all. It continues to demonstrate strong uptake globally and reports very high levels of customer satisfaction.</p><h3><strong>The good</strong></h3><ul><li>It’s cheap — Microsoft is putting pressure on the market with its low cost per user, virtual server subscription and embedded capacity-based pricing. Because many companies have existing Microsoft software agreements, it’s also a small step for procurement teams to roll out the software. License cost was the second most important reason for reference customers to choose Microsoft Power BI. Microsoft also came top third of vendors for sales experience — an improvement from the year before.</li><li>It’s easy to use — Power BI makes complex analysis look easy, with users giving it top-third ratings across all aspects of ease of use. Microsoft’s “first five-minute experience” (which aims to have users register for a trial and be “wowed” within five minutes) and a robust user community contribute to this result.</li><li>It’s got ambitious plans — Microsoft is investing in a broad set of visionary capabilities and integrating them with Power BI. They’re also focusing heavily on augmented analytics like machine learning and natural language processing. The aim is to make these complex technologies adoptable by people with minimal data science skills. Important roadmap items include full support for existing Reporting Services reports, a common and open data model, and open data preparation with data flows.</li></ul><h3><strong>The bad</strong></h3><ul><li>Multiple products: Although the core of Power BI is a self-contained product, Microsoft’s roadmap spans multiple products. For example, more robust conversational analytics is supported through the Cortana personal digital assistant. Data scale-up options require multiple products.</li><li>Differences in on-premises and cloud service with Azure cloud only: Power BI Report Server enables users to share reports (not dashboards) and lacks some of the ML capabilities found in Power BI SaaS. Users also report inconsistencies in the support of different data sources. Microsoft does not give customers flexibility to choose a cloud infrastructure as a service (IaaS) offering, instead running only in Azure.</li><li>With Power BI, Microsoft has mainly focused on agile, self-service analytics. On-premises SQL Server Reporting Services meets the needs for scheduled, distributed reports. This has resulted in a two-product deployment with different capabilities and different deployment approaches. Reporting Services content authoring and report distribution in Power BI is on the short-term roadmap.</li></ul><p><em>Source: Gartner </em>Magic Quadrant for Analytics and Business Intelligence Platforms. <em>See the full review here: </em><a href="https://www.gartner.com/doc/reprints?id=1-65P04FG&amp;ct=190125&amp;st=sb"><em>https://www.gartner.com/doc/reprints?id=1-65P04FG&amp;ct=190125&amp;st=sb</em></a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*8Rh5eBgcKFhw_zBZwaGwhw.jpeg" /></figure><p>Qlik delivers data discovery and agile analytics via its lead product, Qlik Sense. The platform can be used to build customised applications via an extensive set of APIs, so it supports a huge range of data analytics use cases. Qlik’s original product, QlikView, accounts for a large portion of the company’s revenue — more than 67%. According to customers, Qlik is used primarily for decentralized analytics (70%) and agile centralized BI provisioning (66%) use cases.</p><p>In July 2018, Qlik acquired Podium Data to broaden its data management capabilities. Then in January 2019, Qlik acquired Crunch Data so it could provide conversational analytics. In 2018, it launched a new pricing model to encourage QlikView customers to move to Qlik Sense, or to add Qlik Sense to their deployments; it offers dual licenses for a nominal fee.</p><p>Qlik’s position in the Leaders quadrant is thanks to its strong product roadmap, geographic reach and market understanding. But its market responsiveness scores are lower than those of other Leaders.</p><h3><strong>The good</strong></h3><ul><li>Product features and extensibility: Qlik has added augmented features via its Cognitive Engine. Improved self-service capabilities and Insight Advisor help users find useful insights faster with automatic chart generation, as well as suggested insights that the engine thinks may be important. Qlik Core, added in 2018, is a development platform where developers can build applications for cloud or edge deployments. Qlik’s scalable in-memory engine lets customers build robust, interactive, visual applications. It also supports multiple data sources, complex data models and complex calculations.</li><li>Customer experience: Qlik’s active user community is a key contributor to customers’ high scores for user enablement. These are supported by high scores from Gartner Peer Insights reviewers relating to the quality of the peer user community. Qlik’s business-value-based messaging, Data Literacy Project campaigns, and multiple conference series help to influence the market and stimulate demand.</li><li>Product vision: Qlik continues to extend and enhance its platform as the market evolves. It handles big data with its On-Demand App Generation and will expand these capabilities with its new Associative Big Data Index. Qlik continues to execute on its augmented analytics roadmap, as well as improve its data preparation and embedded analytics capabilities.</li></ul><h3><strong>The bad</strong></h3><ul><li>Lots of products: QlikView and Qlik Sense need multiple products to accommodate specific pieces of an analytics workflow. Podium Data, now called Qlik Data Catalyst, will become a stand-alone product to deliver data cataloguing and more advanced data prep.</li><li>Migration experience: Qlik announced a new pricing model in July 2018, with an option for existing QlikView customers to start using Qlik Sense. However, Qlik’s reference customers score the vendor below average for operations, with the main contributor being low scores for migration experience. When asked to name limitations with regard to wider deployment, Qlik’s customers frequently identified functional differences between QlikView and Qlik Sense. Gartner Peer Insights reviewers also rate Qlik’s service and support as below average.</li><li>Lower momentum: Qlik reduced its headcount significantly in 2018. Compared with 2017, there has been a slight decline in interest in Qlik from Gartner clients, judging from Gartner’s client inquiry data for 2018 and social media analytics. This was particularly evident for first-time customers evaluating Qlik against other leading vendors in this market.</li></ul><p><em>Source: Gartner </em>Magic Quadrant for Analytics and Business Intelligence Platforms. <em>See the full review here: </em><a href="https://www.gartner.com/doc/reprints?id=1-65P04FG&amp;ct=190125&amp;st=sb"><em>https://www.gartner.com/doc/reprints?id=1-65P04FG&amp;ct=190125&amp;st=sb</em></a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*l2HfcLmEMJhTQeT7_9w6Gw.png" /></figure><p>Tableau offers an intuitive, interactive experience where you can prepare, analyse and present findings in your data without technical skills or coding. It comes in two forms, primarily as an on-premises software desktop application or as a cloud-based SaaS offering (Tableau Online).</p><p>In 2018, Tableau introduced a new, lower-priced Viewer role and now leads with named-user, subscription-based pricing. Tableau Prep was released to improve data preparation and profiling within Tableau Desktop — and more robust server-based scheduling capabilities are in beta testing. Tableau also acquired Empirical Systems in 2018 to broaden its augmented analytics capabilities.</p><p>Tableau’s customers report that they use it primarily for decentralized analytics (70%) and agile, centralized BI provisioning (51%).</p><p>Tableau is a Leader, thanks to the popularity of its product, high customer satisfaction scores and strong roadmap.</p><h3><strong>The good</strong></h3><ul><li>Easy visual exploration and data manipulation: Tableau can ingest data from a broad range of data sources, blend them, and visualize results using best practices in visual perception. Data can be manipulated while visualizing — such as when creating groups, bins and new hierarchies — all with a high degree of ease of use.</li><li>Customers as fans: Customers have a fanlike attitude toward Tableau; a record 17,000 users attended its 2018 annual user conference. Customers placed Tableau in the top third of Magic Quadrant vendors for customer experience, and gave it high scores for achievement of business benefits. Tableau sets the industry standard for Meetup groups, roadshows, online tutorials and availability of skills in the market.</li><li>Momentum: Tableau grew its total revenue to just over $800 million through 3Q18 — double-digit growth compared with 2017. This was despite moving to subscription-based licensing, which often impairs a vendor’s growth. Tableau remains at the top of many customers’ shortlists, and continues to expand within its installed base. The Tableau Foundation and Tableau Public have been a force in the Data for Good movement, having recently pledged $100 million in funding over the next seven years.</li></ul><h3><strong>The Bad</strong></h3><ul><li>Product gaps: Support for querying multiple fact tables and complex schemas in a single data source is absent from Tableau’s product. It also doesn’t support scheduled, bursted reports in a variety of output formats, or the promotion of content through development, testing and production processes. Support for bursted reports with output to PDF is on the short-term roadmap.</li><li>Support decline: The responses of surveyed customers, together with other Gartner research, indicate that the quality of Tableau’s product support declined in 2018. Customers’ responses put Tableau in the bottom third of vendors in this Magic Quadrant, due partly to more difficult upgrades. Hyper was a major engine replacement to boost performance — one that has not gone as smoothly as previous releases. Further, 12% of Tableau’s reference customers say poor performance remains a problem, and 13% say the product cannot handle the required data volumes (both percentages are above the average).</li><li>Sales experience, contracting and cost: Tableau did well to introduce a new, lower-priced viewer license to compete better against Microsoft in particular, but this license is only available with a subscription license. Consequently, perpetual customers have to move to a new named-user and subscription model to be able to buy this new license. These conversions can be a point of friction, which may explain why Tableau’s customers place it in the bottom third of vendors in this Magic Quadrant for sales experience. Gartner Peer Insights reviewers place it in the bottom third for price and contract flexibility. Over one-third (35%) of Tableau’s customers identified cost as a limitation with regard to wider deployment.</li></ul><p><em>Source: Gartner </em>Magic Quadrant for Analytics and Business Intelligence Platforms. <em>See the full review here: </em><a href="https://www.gartner.com/doc/reprints?id=1-65P04FG&amp;ct=190125&amp;st=sb"><em>https://www.gartner.com/doc/reprints?id=1-65P04FG&amp;ct=190125&amp;st=sb</em></a></p><h3><strong>What we say</strong></h3><p>Just as a picture paints a thousand words, a good visualisation paints a thousand insights. It’s a key part in your data journey, where you bring your important analysis to life.</p><p>Whichever solution you choose, always refer back to your business case and ask: which best supports my business requirements? To get it right, your platform needs to be suitable for the end user, fit your budget, and provide the right level of ongoing support. At Data ³, we’re fans of PowerBI and Tableau, but we’re technology and tool agnostic — we use whatever works best for the project in hand. At all stages, from building the right use cases and extracting the data, to analysing and building models, we provide feedback and ongoing support. If you’re in need of help at any of these stages, visit <a href="http://www.data-cubed.co.uk/"><strong>www.data-cubed.co.uk</strong></a> and get in touch.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9046a1870024" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[IS ONE DATA ANALYST EVER ENOUGH?]]></title>
            <link>https://medium.com/@DataCubed/is-one-data-analyst-ever-enough-84f244e62dc1?source=rss-7f94abe4b137------2</link>
            <guid isPermaLink="false">https://medium.com/p/84f244e62dc1</guid>
            <category><![CDATA[data-analyst]]></category>
            <category><![CDATA[data-visualization]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[data-consultant]]></category>
            <dc:creator><![CDATA[Data³]]></dc:creator>
            <pubDate>Fri, 22 Mar 2019 14:51:18 GMT</pubDate>
            <atom:updated>2019-03-22T14:51:18.043Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*XLZtw2dBXxMl9l01EnvRXA.jpeg" /></figure><blockquote>How applying a multi-skilled team to a data analytics project delivers results</blockquote><p>We work with businesses of all sizes…and they have a wide range of data-focused employees. The big businesses have large data teams…the SMEs have one analyst…and the start-ups have multi-skilled individuals who do everything in the business…literally everything. But, when it comes to data, we see a common mistake — asking one data person to do everything on a data project.</p><h3><strong>A trio of data skills</strong></h3><p>We believe you need a trio of data skills, on a data project, as a minimum:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/800/1*I5WujQYc7Kdmvc0PgVMLmA.png" /></figure><p>If you’re missing one of these skills on a project, it won’t work…or not as well, at least. If you miss out the data consultant, you risk the project not delivering any commercial value. If you miss out the data analyst, you risk not being able to access, combine and analyse the more complex data sources. If you miss out the data visualiser, you risk not sharing the insight from your work in an intuitive way. You need all three to be assured of a successful data project.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*0Kv5tKXP2WFRtdiRdXcCRQ.jpeg" /></figure><h3><strong>The data consultant will create the use case…</strong></h3><p>This person needs to properly understand the business, the product/s, the market, the customer/s, the distributors and the business levers that can be pulled to influence costs and revenue. This knowledge needs to be applied to creating a commercial use case for the data project, for instance:</p><ul><li>How can we use data to make money?</li><li>How can we use data to save money?</li><li>How can we use data to create a competitive advantage?</li><li>How can we use data to innovate?</li></ul><p>Without this perspective, the project risks being a waste of time where the end result could offer no demonstrable value to the business performance or offer no actionable insight. The data consultant brings the ‘so what?’.</p><p>The challenge here is looking beyond the obvious. Just because past business performance shows a certain trend, that doesn’t mean it will continue in future as there might be market, regulatory or other external influence that will impact supply and demand of a business’ product or service in future. Or, just because the business has had certain revenue streams in the past, that doesn’t mean there can’t be a new one that could be data-driven.</p><p>Data consultants are often described as interpreters — because they sit between the commercial part of the business and the technology part of the business. They need to be able to speak the language of business and interpret it into tech speak.</p><p>The data consultant needs to be the innovator — the challenger, the thought-provoker, the commercial idea generator.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*gNirH-gkxuWjB99Jexleww.jpeg" /></figure><h3><strong>The data analyst will prepare the data…</strong></h3><p>This person needs to be a coder, a programmer and confident in dealing with data in all different formats, sizes and complexity. The data analyst will need to:</p><ul><li>Extract the data from a myriad of data sources</li><li>Cleanse the data and remove or process anomalies or false data</li><li>Combine data sources by finding or creating matching criteria</li><li>Analyse the data using a range of statistical and analytical models</li><li>Prepare the data for analysis</li><li>Connect up live data feeds.</li></ul><p>Without this perspective, the project risks being limited to only the simplest data sources and the simplest analytical methods which could result in far less valuable insight. The data analyst brings the tech.</p><p>The challenge here is juggling large volumes of data, sometimes from a range of data sources and often in a wide variety of data formats. Matching and combining data can be an additional challenge where there are not unique IDs for each record, so this can often be a complex process. Then there are the analytical methods which can include creating models for value analysis, segmentation, trend analysis and predictive models.</p><p>The data analyst needs to be the data wrangler — the data connector, the problem solver, the model builder.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*w8fFdZ3gAy7q1x_qOFB-gg.jpeg" /></figure><h3><strong>The data visualiser will tell the story…</strong></h3><p>This person needs to be both customer and user orientated. They need to be focused on who will use the data tool they’re building — who are they? How do they work? What do they care about? What actions can they take using the tool? This understanding needs to be interpreted into building the data visualisation tool so the data visualiser will:</p><ul><li>Visualise the data to understand patterns and trends</li><li>Create a data story for the user</li><li>Design the interactivity of the data tool so that it matches what the user will want to see/do</li><li>Select the appropriate visualisations and graphs to tell the story in the best way</li><li>Act on feedback from the users to refine and iterate the data tool</li><li>Train the users to use the data tool.</li></ul><p>Without this perspective, the project risks having a poor end result, confusing data visualisations or a lack of understandable insight. The data analyst brings the story.</p><p>Visualisation can often be seen as the pretty part, the easy bit or the quick task. It’s the complete opposite. <strong><em>Whilst it’s really easy to make something look complicated, it’s really hard to make something look simple.</em></strong><em> </em>Getting the story right, choosing the right graphs, allowing the appropriate level of interactivity — these all require deep user understanding, masses of experience and a dedication to data storytelling.</p><p>The data visualiser needs to be the storyteller — the user investigator, the completer-finisher, the designer.</p><h3><strong>Is one data analyst ever enough?</strong></h3><p>We don’t think so. You cannot expect one person to be the data consultant, data analyst and data visualiser combined. No one person would have all these skills developed to the required level for a data project. One person cannot be all things to everyone.</p><p>Our multi-skilled data team powers our clients’ businesses on a daily basis. We always work as a trio. Sometimes in larger teams when the projects require more. But never alone. One data analyst is NOT enough.</p><p><strong>We’re Data — we use data analytics to power business performance for our clients. </strong><a href="http://www.data-cubed.co.uk/"><strong>Can we help yours?</strong></a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=84f244e62dc1" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Buzzword bonanza 2: Our plain English guide to data lingo]]></title>
            <link>https://medium.com/@DataCubed/buzzword-bonanza-2-our-plain-english-guide-to-data-lingo-be9520b29080?source=rss-7f94abe4b137------2</link>
            <guid isPermaLink="false">https://medium.com/p/be9520b29080</guid>
            <category><![CDATA[naturallanguageprocessing]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[internet-of-things]]></category>
            <category><![CDATA[digital-automation]]></category>
            <category><![CDATA[deep-learning]]></category>
            <dc:creator><![CDATA[Data³]]></dc:creator>
            <pubDate>Thu, 21 Mar 2019 09:39:20 GMT</pubDate>
            <atom:updated>2019-03-21T09:39:20.896Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*IZcJKz3761vChU1VFHfzkw.jpeg" /></figure><blockquote><strong>Digital business automation, Natural Language Processing, Deep Learning &amp; Internet of Things</strong></blockquote><p>Round up, round up! It’s time to tackle more of the data buzzwords that are popping up in meeting rooms everywhere. Say goodbye to blank expressions and scratchy heads, this is your definitive guide to the latest technologies, explained in plain English.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*-4sTsXMzGJDXMDnOZEdz3w.jpeg" /></figure><h3><strong>Digital business automation</strong></h3><p>Digital business automation is the act of taking time-intensive, manual or costly work processes and using digital technology to speed them up. It’s a popular process for modern businesses, particularly ones that have been around for a while. As we know, processes typically change over time, and organic change doesn’t always mean improvement. Digital business automation challenges you to analyse your workflows and look at what can be improved.</p><p>Some of the areas include:</p><ul><li>Tasks — summarising business tasks and assessing which ones are repetitive enough to be automated</li><li>Workflow — analysing, reorganising and streamlining processes across a business</li><li>Capture — digitising business documents, capturing and analysing the data that comes from it</li></ul><p>Data analytics often goes hand in hand with business automation. It allows you to determine whether a problem was just a blip or whether it falls into a pattern. When a workflow fails, analytics will tell you where and why it failed. Managers overseeing the process can then pinpoint areas for improvement rapidly without the need to manually analyse huge chunks of data first.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*158yf37jsCNiME5lLgLwxw.jpeg" /></figure><h3><strong>Natural language processing</strong></h3><p>Natural language processing (NLP) is a branch of <a href="https://medium.com/@DataCubed/buzzword-bonanza-our-plain-english-guide-to-data-lingo-b638eda0b149">artificial intelligence</a> that helps computers understand, interpret and manipulate human language.</p><p>A common example of NLP is when you type on your mobile and see word suggestions pop up based on what you’re currently typing. That’s natural language processing in action. It’s a little thing that many of us have taken for granted for years, but its application in the business world is far-reaching.</p><p>Take a company looking at advertising a new product. They can use google to find common search terms that their users type when they’re looking for similar products. NLP allows for a quick compilation of the data into terms obviously related to their brand and those that they might not expect. Capitalizing on the uncommon terms can give the company the ability to advertise in new ways.</p><p><em>How does it work?</em></p><p>The first step in NLP depends on the application of the system. For voice-based systems like Alexa, special technology translates words into text. Next, the system breaks each word down into categories like nouns, verbs etc. This happens through a series of coded grammar rules that rely on algorithms that incorporate statistical machine learning to help determine the context of what you said. If we’re not talking about speech-to-text NLP, the system just skips this step.</p><p>The end result is the machine’s ability to categorize what is said in many different ways. Depending on the underlying focus of the NLP software, the results get used in different ways. For instance, in our SEO example above, an application could use the decoded text to pull in keywords associated with a certain product.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*8HMoWJ-JNAuLcp8TtGQJQg.jpeg" /></figure><h3><strong>Deep learning</strong></h3><p>Deep learning is a subset of machine learning in artificial intelligence that takes mountains of unstructured data and attempts to make sense of it. It imitates the workings of the human brain in processing data and creating patterns for use in decision making.</p><p><em>Deep learning vs. machine learning</em></p><p>A common AI technique used for processing big data is machine learning, which we covered in our last buzzword blog. Machine learning is a self-adaptive algorithm that gets increasingly better analysis and patterns with experience or with newly added data. A practical example of this is a company that wants to identify payments fraud within their business. By using machine learning tools, it can process huge levels of transaction data and pick out anomalies in the data, accurately predicting over time what is a legitimate or fraudulent payment.</p><p>Deep learning is a means to do this in great depth. Where a standard program could analyse simple variables like overpayments, a deep learning function could analyse multiple data sets in a nonlinear way. The analysis could include time, geographic location, IP address, type of retailer and any other feature that is likely to point to fraudulent activity. Each data layer can be analysed simultaneously and results and patterns aggregated instantly. This allows unstructured data to be analysed and comprehended at scale and speeds impossible for humans or traditional analytical programs.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*X3TNZbdu06LWNCpHixIFxQ.jpeg" /></figure><h3><strong>Internet of things (IoT)</strong></h3><p>IoT is used to describe the process of connecting everyday devices to the Internet, in order to collect data from them, exchange data between devices or control them from a distance.</p><p>These devices can be cars, home automation systems or your everyday toaster.</p><p>Caroline Gorski, the head of IoT at Digital Catapult explains “It’s about networks, it’s about devices, and it’s about data,”. IoT allows devices on closed private internet connections to communicate with others and the Internet of Things brings those networks together. It gives the opportunity for devices to communicate not only within close silos but across different networking types and creates a much more connected world.</p><p><em>What does this mean for business?</em></p><p>Experts predict that more than half of new businesses will run on the IoT by 2020. Within industrial applications, sensors on product lines can increase efficiency and cut down on waste. One study estimates 35 per cent of US manufacturers are using data from smart sensors within their set-ups already.</p><p>But it’s not all good news. Anyone who’s purchased one of the myriad smart home products — from lightbulbs, switches, to motion sensors — will attest to the fact IoT is in its infancy. Products don’t always easily connect to each other and there are significant security issues that need to be addressed. A report from Samsung says the need to secure every connected device by 2020 is “critical”. The firm’s Open Economy document says “there is a very clear danger that technology is running ahead of the game”. The firm said more than 7.3 billion devices will need to be made secure by their manufacturers before 2020.</p><p>So there we have it. Another round of data buzzwords taken head-on. If you’re interested in finding out more about these technologies — or have a data project that you’re starting or struggling with — <a href="http://www.data-cubed.co.uk/">please do get in touch</a>.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=be9520b29080" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>