<?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 Alexandra Ugolnikova on Medium]]></title>
        <description><![CDATA[Stories by Alexandra Ugolnikova on Medium]]></description>
        <link>https://medium.com/@alexandraugolnikova?source=rss-9a3ce45a35a4------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*YdM_CYhsV3ClClnq.</url>
            <title>Stories by Alexandra Ugolnikova on Medium</title>
            <link>https://medium.com/@alexandraugolnikova?source=rss-9a3ce45a35a4------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Wed, 18 Nov 2026 06:23:56 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@alexandraugolnikova/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[Cookbook of Finest Data Analytics]]></title>
            <link>https://medium.com/@alexandraugolnikova/cookbook-of-finest-data-analytics-1be76406245c?source=rss-9a3ce45a35a4------2</link>
            <guid isPermaLink="false">https://medium.com/p/1be76406245c</guid>
            <category><![CDATA[data-analytics]]></category>
            <category><![CDATA[data]]></category>
            <category><![CDATA[sarcasm]]></category>
            <category><![CDATA[data-science]]></category>
            <dc:creator><![CDATA[Alexandra Ugolnikova]]></dc:creator>
            <pubDate>Mon, 04 May 2020 17:40:33 GMT</pubDate>
            <atom:updated>2020-05-04T18:35:26.018Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>Follow these 12 recipes to establish the best worst data practices in your team.</strong></p><p><em>Proven to increase understanding of data of your business from “I am on a path to know more, I just need a little bit” to “I will never understand my data and users, we are all doomed. DOOMED”.</em></p><p><em>Helped multiple data teams decrease the quality of their work and double the delivery speed.</em></p><p><em>Contributed to a healthy ambiance of mistrust across stakeholders and analysts, analysts and analysts, investors and business owners</em>.</p><p><strong>1. The best stories are lies.</strong></p><p>‘Choose your story — data will follow’, they say. And if the data does not match the story… just do the analysis again.</p><p><strong>2. If you can’t convince them, confuse them.</strong></p><p>Even if your stakeholders came to you with some vague numbers in mind, it is your sacred duty as a good data samaritan to ensure that they leave more confused than ever.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/609/1*UOg2PdcQjhAVQ4errF_Chw.gif" /></figure><p><strong>3. Fake it until you make it.</strong></p><p>‘Just one more aggregation, just here, what, these numbers look weird. This must be an outlier. And this one too. And you, I just don’t like you. Go away. Go away you little bastards, don’t pollute my otherwise perfectly solid and consistent results. Aha beautiful, now I am ready to share it with the team.’</p><p><strong>4. Data is like gouda, the older the better.</strong></p><p>Haven’t we all had this deep inner battle for integrity when after spending our precious efforts on analysis just to realize that the data is outdated thus making our findings obsolete?</p><p><strong>5. People are nothing, numbers are everything.</strong></p><p>‘What, that 10 digit identifier is actually a human being whose behavior it is my job to analyze? Wait a minute, that was not mentioned in the contract! Now I need to care about their feelings and desires? It is already hard enough to do that in my own relationships, and now I need to understand what a bunch of users want at scale?’</p><p><strong>6. Your hypothesis is everything. Never ever give it up!</strong></p><p>Several A/B tests showed that your hypothesis is not significant but you are still holding on to it as a dog to its bone — loyalty is a rare find these days. Your team is happy to have you!</p><p><strong>7. Handmade is always better.</strong></p><p>It is good to write the same code over and over again. Why generate something scalable for your data manipulations that can save your team hours of effort if every time can be like the first one?</p><p><strong>8. Don’t judge data by where it comes from. We all have a past.</strong></p><p>Not sure how your KPIs are tracked and where they are coming from? Is there a high chance that they are biased towards one type of a user cohort? Are you basing your business plan on it? That’s right, time to open that bottle of champaign!</p><p><strong>9. Databases are tough nuts. They don’t need anyone to understand them.</strong></p><p>Who said that you need to understand what the fields of a database actually mean? There are so many fields, that’s a lot of understanding. Are you crazy? ‘I didn’t know that this boolean value means that this customer actually doesn’t exist anymore but I’ve always counted them as one.’ One for all and all for one!</p><p><strong>10. You need Machine Learning. ASAP.</strong></p><p>You invested half a million dollars in machine learning but after 6 months realized that there are fundamental problems with data and it was all in vain? No big deal, your investors are still impressed because you are using the learning of the machines! You are on the right path to greatness, my friend!</p><p><strong>11. Numbers never lie!</strong></p><p>It is important to keep in mind that your data knows better than you. Forget common sense — data is your best friend. In fact, your only friend in this scary world.</p><p><strong>12. Leave storytelling for your improvisation class.</strong></p><p>You did a fancy analysis, you are a rock star. Who cares if anyone understands what those numbers actually mean and why you are talking about them?</p><p><strong>And remember, no matter what happens you can always blame it on data quality!</strong></p><p>Thanks to Sam and Nina for all the funs and ideas for this cookbook.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1be76406245c" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>