1<!doctype html><html lang=en-gb><head><meta charset=UTF-8><meta name=HandheldFriendly content="True"><meta http-equiv=X-UA-Compatible content="IE=edge"><meta name=viewport content="width=device-width,initial-scale=1"><link rel=preconnect href=https://rsms.me/><link rel=canonical href=https://xmrit.com/about/><title>What is this? | Xmrit </title>
vendor: 1,104 bytes, line 1
1<script>!function(e,t){var n,s,o,i;t.__SV||(window.posthog=t,t._i=[],t.init=function(a,r,c){function d(e,t){var n=t.split(".");2==n.length&&(e=e[n[0]],t=n[1]),e[t]=function(){e.push([t].concat(Array.prototype.slice.call(arguments,0)))}}(s=e.createElement("script")).type="text/javascript",s.async=!0,s.src=r.api_host+"/static/array.js",(i=e.getElementsByTagName("script")[0]).parentNode.insertBefore(s,i);var l=t;for(void 0!==c?l=t[c]=[]:c="posthog",l.people=l.people||[],l.toString=function(e){var t="posthog";return"posthog"!==c&&(t+="."+c),e||(t+=" (stub)"),t},l.people.toString=function(){return l.toString(1)+".people (stub)"},o="capture identify alias people.set people.set_once set_config register register_once unregister opt_out_capturing has_opted_out_capturing opt_in_capturing reset isFeatureEnabled onFeatureFlags getFeatureFlag getFeatureFlagPayload reloadFeatureFlags group updateEarlyAccessFeatureEnrollment getEarlyAccessFeatures getActiveMatchingSurveys getSurveys onSessionId".split(" "),n=0;n<o.length;n++)d(l,o[n]);t._i.push([a,r,c])},t.__SV=1)}(document,window.posthog||[])</script>
1<script type=module src=/js/alpine.f7c503582fddf62852a93d54351b9758d5de3cd10e5f8a2152120e9106c4c076e8e6d15e07d02587c940b1a980f5ed1a1940cb42b730b357680acd0bd82833a9.js integrity="sha512-98UDWC/d9ihSqT1UNRuXWNXePNEOX4ohUhIOkQbEwHbo5tFeB9Alh8lAsamA9e0aGUDLQrcws1doCs0L2CgzqQ=="></script>
1<link rel=stylesheet href=/css/styles.css><link rel=stylesheet href=https://rsms.me/inter/inter.css><link rel=icon href=https://xmrit.com/favicon.ico><link rel=apple-touch-icon sizes=180x180 href=https://xmrit.com/apple-touch-icon.png><link rel=icon type=image/png sizes=32x32 href=https://xmrit.com/favicon-32x32.png><link rel=icon type=image/png sizes=16x16 href=https://xmrit.com/favicon-16x16.png><link rel=manifest href=https://xmrit.com/site.webmanifest><meta property="og:title" content="What is this?"><meta property="og:description" content="What are XmR charts and why would you want to use them?"><meta property="og:type" content="article"><meta property="og:url" content="https://xmrit.com/about/"><meta property="og:image" content="https://xmrit.com/sample_xmr.png"><meta property="article:section" content><meta property="article:published_time" content="2023-11-17T15:57:29+07:00"><meta property="article:modified_time" content="2023-11-17T15:57:29+07:00"><meta itemprop=name content="What is this?"><meta itemprop=description content="What are XmR charts and why would you want to use them?"><meta itemprop=datePublished content="2023-11-17T15:57:29+07:00"><meta itemprop=dateModified content="2023-11-17T15:57:29+07:00"><meta itemprop=wordCount content="1487"><meta itemprop=image content="https://xmrit.com/sample_xmr.png"><meta itemprop=keywords content><meta name=twitter:card content="summary_large_image"><meta name=twitter:image content="https://xmrit.com/sample_xmr.png"><meta name=twitter:title content="What is this?"><meta name=twitter:description content="What are XmR charts and why would you want to use them?"><meta name=twitter:site content="@xmrit_official"></head><body class='max-w-screen-2xl mx-auto'><header class="col-span-12 pt-3 px-4 border-b pb-5 mb-5 border-slate-400" x-data="{ isOpen: false }"><div class="flex items-stretch md:items-center justify-between"><a class='flex items-center' href=/><img class="w-16 h-16 mr-1" src=https://xmrit.com/xmrit-logo.png><div class="flex flex-col justify-start"><h1 class="text-2xl md:text-4xl font-sans font-bold tracking-tight text-gray-700">Xmrit</h1><h3 class="text-xs md:text-sm font-sans -mt-1 italic">by Commoncog</h3></div></a><button type=button class="lg:hidden bg-white rounded-md p-2 inline-flex items-center justify-center text-gray-400 hover:text-blue-800 focus:outline-none focus:ring-2 focus-ring-inset focus:ring-white" aria-expanded=false @click="isOpen = !isOpen"> 2<span class=sr-only>Open main menu</span><svg class="h-5 w-5" xmlns="http://www.w3.org/2000/svg" fill="none" viewBox="0 0 24 24" stroke-width="2" stroke="currentcolor" aria-hidden="true"><path stroke-linecap="round" stroke-linejoin="round" d="M4 6h16M4 12h16M4 18h16"/></svg></button><nav class="mt-5 md:mt-0 hidden lg:block text-sm md:text-base"><a class="hidden lg:inline px-2 md:px-4 py-3 hover:text-blue-700" href=/articles/>Articles</a> 3<a class="hidden lg:inline px-2 md:px-4 py-3 hover:text-blue-700" href=/opensource/>Open Source</a> 4<a class="hidden lg:inline px-2 md:px-4 py-3 hover:text-blue-700 underline decoration-indigo-500 underline-offset-4" href=https://shop.xmrit.com/metrics-masterclass>Metrics Masterclass</a> 5<a class="hidden lg:inline px-2 md:px-4 py-3 hover:text-blue-700" href=/manual/>User Manual</a> 6<a class="hidden lg:inline px-2 md:px-4 py-3 hover:text-blue-700" href=/about/>What is This?</a> 7<a class="px-2 md:px-4 md:ml-4 py-3 text-white bg-blue-800 hover:bg-blue-700 rounded-lg" href=/t/>Go to Tool→</a></nav><div x-cloak x-show=isOpen x-transition:enter="duration-150 ease-out" x-transition:enter-start="opacity-0 scale-95" x-transition:enter-end="opacity-100 scale-100" x-transition:leave="duration-100 ease-in" x-transition:leave-start="opacity-100 scale-100" x-transition:leave-end="opacity-0 scale-95" class="absolute top-0 inset-x-0 p-2 transition transform origin-top lg:hidden z-50"><div class="rounded-lg shadow-md bg-white ring-1 ring-black ring-opacity-5 overflow-hidden"><div class="px-5 pt-4 flex items-center justify-between"><div><a href=/><img class="h-8 w-auto" src=https://xmrit.com/xmrit-logo.png alt></a></div><div class=-mr-2><button type=button class="bg-white rounded-md p-2 inline-flex items-center justify-center text-gray-400 hover:bg-gray-100 focus:outline-none focus:ring-2 focus:ring-inset focus:ring-cyan-600" @click="isOpen = !isOpen"> 8<span class=sr-only>
8Close menu</span><svg class="h-6 w-6" xmlns="http://www.w3.org/2000/svg" fill="none" viewBox="0 0 24 24" stroke-width="2" stroke="currentcolor" aria-hidden="true"><path stroke-linecap="round" stroke-linejoin="round" d="M6 18 18 6M6 6l12 12"/></svg></button></div></div><div class="pt-5 pb-6"><div class="px-2 space-y-1"><a class="block px-3 py-2 lg:px-2 lg:py-0 rounded-md text-base font-medium sm:font-normal text-gray-900 hover:bg-gray-50 lg:hover:text-purple-700 lg:hover:bg-white" href=/articles/>Articles</a> 9<a class="block px-3 py-2 lg:px-2 lg:py-0 rounded-md text-base font-medium sm:font-normal text-gray-900 hover:bg-gray-50 lg:hover:text-purple-700 lg:hover:bg-white" href=/about/>What is This?</a> 10<a class="block px-3 py-2 lg:px-2 lg:py-0 rounded-md text-base font-medium sm:font-normal text-gray-900 hover:bg-gray-50 lg:hover:text-purple-700 lg:hover:bg-white" href=/manual/>User Manual</a> 11<a class="block px-3 py-2 lg:px-2 lg:py-0 rounded-md text-base font-medium sm:font-normal text-gray-900 hover:bg-gray-50 lg:hover:text-purple-700 lg:hover:bg-white" href=/opensource/>Open Source</a> 12<a class="block px-3 py-2 lg:px-2 lg:py-0 rounded-md text-base font-medium sm:font-normal text-gray-900 hover:bg-gray-50 lg:hover:text-purple-700 lg:hover:bg-white underline decoration-indigo-500 underline-offset-4" href=https://shop.xmrit.com/metrics-masterclass>Metrics Masterclass</a></div><div class="mt-6 px-5"><a href=/t/ class="block text-center w-full py-3 px-4 rounded-md text-slate-900 border border-slate-900 hover:bg-blue-900 hover:text-white">Go To Tool</a></div><div class="mt-6 px-5"></div></div></div></div></div></header><main class='relative bg-white'><section class='grid grid-cols-12 grid-flow-row gap-1 md:gap-2 px-2 sm:px-6 mx-auto'><div class="col-span-12 md:col-span-8 md:col-start-3 py-3 px-2 mb-3"><h1 class="text-5xl md:text-6xl font-bold tracking-tight">What is this?</h1></div><div class="col-span-12 md:col-span-8 md:col-start-3 pt-3 px-2 pb-5 mb-5"><h2 class='text-slate-800 border-b border-slate-300 max-w-prose pb-1 mb-4 toc-header'>Table of Contents</h2><div class="js-toc mb-3 lg:mb-0 pb-3 border-b border-slate-300 max-w-prose"></div><button title="back to top" class="fixed right-5 bottom-5 bg-slate-700 hover:bg-slate-600 text-white leading-none shadow-md rounded-full h-16 w-16 flex flex-col items-center justify-center" onclick='window.scrollTo({top:0,behavior:"smooth"})'> 13▲<span class="block text-sm">Top</span></button><article class="prose lg:prose-lg pgcontent mt-6"><p>Xmrit (pronounced <em>‘xam-mer-it’</em>) is a free tool by <a href=https://commoncog.com>Commoncog</a> to help 14business operators quickly create and share XmR charts.</p><h2 id=what-is-an-xmr-chart>What is an XmR chart?</h2><p>An XmR chart helps you analyse your data. In a sentence: it helps you separate signal from noise. It is particularly useful to business operators.</p><p>To understand this, letâs talk about a problem with interpreting data that everyone should be familiar with.</p><p>Letâs say that you open a business dashboard to look at some metric. The metric goes up and down.</p><p><img src=New-Subscribers-Variation.png alt="A squiggly line chart showing new subscribers"></p><p>Thatâs nice. Now what?</p><p>What does the chart tell you? If it goes up, is this good? If it goes down, is this bad? What should you do in response to what you see?</p><p>If youâre like most folks, you <em>probably donât know</em>. You’ve probably had the experience where you open an analytics tool and felt a little … <em>confused</em>. You donât know what to investigate. You donât know if a number going up is good, or if a number going down is bad, or whether itâs all just business as usual.</p><h3 id=understanding-variation-is-a-big-deal>Understanding variation is a big deal</h3><p>Hereâs a fact: <em>all</em> business metrics display some variation. Itâs perfectly normal for numbers to wiggle up and down on a routine basis.</p><p>The question you want to ask is this: when a metric changes, is this a signal? Or is this simply routine variation â that is, noise? Should I investigate, or can I safely ignore it?</p><p>If you don’t know how to answer this question, <strong>you won’t be able to become data driven</strong>.</p><p>In fact, this ‘problem of dealing with variation’ leads you to a number of other potential problems:</p><h3 id=problem-1-you-dont-know-if-youve-improved>Problem 1: You don’t know if you’ve improved</h3><p>Letâs say that you want to test if a new sales process is more effective. You make a change to the way Marketing Qualified Leads are vetted and wait a few weeks. Then you take a look at your data:</p><p><img src=prospect_calls_change_1.png alt="A time series of prospect calls showing no discernible change in variation"></p><p>Did it work? Hmm. Maybe wait a few more weeks?</p><p><img src=prospect_calls_change_2.png alt="A time series of prospect calls showing no discernible change in variation, even after a few additional data points"></p><p>Did it work? Did it fail? Hmm. Perhaps itâs made an impact on some other metric? You take a look at a few other sales metrics but theyâre all equally wiggly. You discern no clear pattern; you chalk it up to a âmaaaybe?â and move on to the next idea. You donât get any feedback on your moves. Youâre like a blind archer, shooting arrows into the dark.</p><h3 id=problem-2-you-waste-time-chasing-noise>Problem 2: You waste time chasing noise</h3><p>Your boss opens up the sales meeting and says âSales is down 21% for the month! This is very bad! Weâre no longer on track to hit our quarterly targets!â</p><p>Youâre told to look into it. You spend a huge amount of time investigating, and it turns out that ⦠you canât find anything wrong. Perhaps this is just routine variation?</p><p>Your boss doesnât understand that, of course. He doesnât understand that routine variation exists (if he did, he wouldnât have asked you to investigate). So you make up some explanation, he accepts it, and then next month the number goes up again and everyone breathes a sigh of relief.</p><p>
14Until the next time the number goes down. Then you get yelled at again.</p><p>This all quickly becomes kabuki theatre.</p><h3 id=problem-3-you-may-set-dumb-slas>Problem 3: You may set dumb SLAs</h3><p>Youâre in charge of data infrastructure. You have a maximum latency SLA (Service Level Agreement) for some of your servers. Every week your latency metrics are piped into a Slack channel, with a % change.</p><p>Every two months or so, the latency for a key service violates your SLA. You get yelled at. âWHAT WENT WRONG?â your boss sends over the channel, tagging you.</p><p>You investigate but canât find anything wrong. It doesnât cross your mind that perhaps the processâs natural wiggling will â for some small % of the time â violate the SLA.</p><p><img src=stupid_sla.png alt="A time series with a horizontal green line that represents an SLA drawn across it. The time series breaks above the green line at three points."></p><h3 id=the-end-result>The end result</h3><p>As a result of these and other problems, you stop taking metrics seriously because you learn so little from them.</p><p>Or you learn to dread ânumber go downâ, because you know youâll be asked to investigate. You donât feel in control of your numbers. You feel like they kick you around.</p><p>And, finally, sadly, management learns that âyell at employees when number goes downâ often 15results in the number going up … although whether this is by random chance is anybodyâs guess.</p><h2 id=but-what-if-theres-an-alternative-hellip>But what if thereâs an alternative â¦</h2><p>Letâs pause for a moment and imagine: what if there’s a magical tool that can consume a series of numbers and then tell you what to investigate.</p><p>For instance, it might tell you: âSomething went wrong with sales last month! 16Investigate now!â</p><p><img src=something_wrong_in_sales.png alt="Time series with title ‘Something Wrong in Sales Last Month’"></p><p>Or perhaps it might tell you to ignore things because everything is just 17routine variation.</p><p><img src=predictable_process_xmr.png alt="An XmR chart that shows a predictable process"></p><p>Well, this is what XmR charts do. And theyâre <em>radically</em> simple to understand.</p><h2 id=how-xmr-charts-work>How XmR charts work</h2><p>This is what an XmR chart looks like:</p><p><img src=forum_pageviews.png alt="A screenshot of an X chart and an MR chart"></p><p>XmR charts are so named because they consist of an ‘X’ chart (the ‘X’ variable, 18or the metric you care about), and a ‘Moving Range’ chart, which shows 19differences from point to point.</p><p>XmR charts have only three basic rules.</p><h3 id=rule-1-process-limit-rule>Rule 1: Process Limit Rule</h3><p>Rule one states that if a point lies outside the limit lines (the blue lines), 20on either the X chart or the MR chart, <em>something unusual is going on</em>.</p><p><img src=rule_1.png alt></p><p>You should investigate!</p><h3 id=rule-2-quartile-limit-rule>Rule 2: Quartile Limit Rule</h3><p>Rule two states that if you have a run of three out of four successive points that is closer to the limit lines (blue dotted lines) than the centre line (red dotted line), then this is a <strong>moderate</strong> source of exceptional variation and you should investigate.</p><p>Another way of thinking about this is that you need three out of four successive points beyond a quartile limit (the dotted line in between the limit and centre lines).</p><p>In the image below, three out of four of the yellow points are closer to the lower process limit line (in blue dotted line), then the average line (in the centre, in red).</p><p><img src=rule_2.png alt></p><p>This is a moderate source of exceptional variation. Something odd has happened. You should investigate!</p><h3 id=rule-3-runs-of-eight>Rule 3: Runs of Eight</h3><p>Rule three states that if you have data points in a row on one side of the average (red) line, this is a weak source of special variation and you should investigate.</p><p><img src=rule_3.png alt></p><h2 id=wait-why-does-the-xmr-chart-work>Wait, why does the XmR Chart Work?</h2><p>Here’s the intuition behind why XmR charts work: all processes show some amount of routine variation, yes? We may characterise this variation as drawing from some kind of probability distribution. What kind of probability distribution? Well, we don’t really care.</p><p>What the XmR chart does is to detect <em>the presence of more than one probability distribution in the variation observed in a set of data</em>. We do not care about the exact nature of the probability distributions present, only the number of them.</p><p>If your process is predictable, the variation it shows may be said to be âroutineâ, or to draw from just one probability distribution. On the other hand, when youâve successfully changed your process, the data youâll observe after your change will show different variation from before. We can say that your time series will show variation drawn from two probability distributions. Also: if some unex
20pected, external event impacts your process, we may also say that weâre now drawing from some other probability distribution at the same time.</p><p>The XmR chart does this detection by <em>estimating</em> three sigma around the average line. For the vast majority of real world processes, most routine variation will fall within three sigma of the average.</p><p>If a point falls outside the three sigma limits, something exceptional is going on â thereâs likely another source of variation present. The other two detection rules are run-based (they depend on sequential data points in a time series) and are designed to detect the presence of moderately different probability distributions. For the vast majority of real world distributions, XmR charts will have a ~3% false positive rate (<a href=https://spcpress.com/pdf/DJW220.pdf>source</a>). This is more than good enough for business experimentation.</p><p>This intuition helps explains both the power as well as the limitations of the XmR chart.</p><h2 id=where-can-i-find-more-information>Where can I find more information?</h2><p>We intend to add a <em>lot</em> more information to xmrit.com in the near future. If you want to <em>use</em> XmR charts in practice, you should probably skim through the <a href=/manual/>User Manual</a>, which includes some practical tips for XmR chart usage (in addition to explainers on how to use the Xmrit tool).</p><p>If you’d like an explainer of how XmR charts might lead to company-wide data-driven operations, you’ll probably want to read the essay <a href=https://commoncog.com/becoming-data-driven-first-principles/>Becoming Data Driven, From First Principles</a> over at Commoncog.</p><p>If you’d like more advanced information about XmR charts, your best source of information is likely (in increasing order of complexity):</p><ol><li><a href=https://www.spcpress.com/book_understanding_variation.php>Understanding Variation</a></li><li><a href=https://www.spcpress.com/book_twenty_things.php>20 Things You Need To Know</a></li><li><a href=https://www.spcpress.com/book_making_sense_of_data.php>Making Sense of Data</a></li><li>Donald Wheeler’s <a href=https://www.spcpress.com/djw_columns.php>Quality Digest Columns</a></li></ol></article><div class="mt-20 border-t border-slate-400 pt-5"><h2 class=text-2xl>Would you like to learn more?</h2><div class="not-prose cta-box mt-8 text-base lg:text-lg flex flex-col items-center max-w-prose"><h2 class="text-xl font-bold uppercase text-center my-5">The Free Xmrit Email Course</h2><div class=py-2><p class="mb-3 text-base lg:text-lg">Want to quickly get started with XmR charts? You'll learn …</p><ul class="pl-5 list-disc"><li class=mb-3>How to use XmR charts to <strong>take action in your business</strong>.</li><li class=mb-3><strong>Four major ways</strong> to use an XmR chart!</li><li class=mb-3>When XmR charts <strong>don't work so well</strong>.</li><li class=mb-3>When you <strong>can and cannot trust</strong> your limit lines.</li><li class=mb-3>And more …</li></ul></div><div class=text-center>
20<script async data-uid=fd831d6682 src=https://xmrit.ck.page/fd831d6682/index.js></script>
20</div><p class="text-center text-slate-800 text-sm">One week. No spam. Just the basics.</p></div></div></div></div></section></main><footer class="col-span-12 pt-8 px-4 border-t pb-5 mb-5 border-slate-400"><div class="flex flex-col md:flex-row items-start justify-between"><p class="text-gray-500 order-last md:order-first"><a class=hover:text-gray-800 href=/>Xmrit</a> by <a class=hover:text-purple-600 href=https://commoncog.com>Commoncog</a> © 2026 //<br><a class=hover:text-gray-800 href=https://postcognito.com>Postcognito Pte Ltd</a></p><nav class="mb-5 md:mb-0 flex flex-col md:flex-row"><a class="text-gray-500 hover:text-gray-800 mx-0 md:mx-3 my-1 md:my-0" href=/articles/>Articles</a> 21<a class="text-gray-500 hover:text-gray-800 mx-0 md:mx-3 my-1 md:my-0" href=/opensource/>Open Source</a> 22<a class="text-gray-500 hover:text-gray-800 mx-0 md:mx-3 my-1 md:my-0" href=/integration/>Integration</a> 23<a class="text-gray-500 hover:text-gray-800 mx-0 md:mx-3 my-1 md:my-0" href=https://forms.gle/93hdRwdjJYDxnvSx5>Send Feedback</a> 24<a class="text-gray-500 hover:text-gray-800 my-1 md:my-0" href=/privacy/>Privacy Policy</a> 25<a class="text-gray-500 hover:text-gray-800 mx-0 md:mx-3 my-1 md:my-0 underline decoration-indigo-500 underline-offset-4" href=https://shop.xmrit.com/metrics-masterclass>Metrics Masterclass</a></nav></div></footer>
25<script src=https://cdnjs.cloudflare.com/ajax/libs/tocbot/4.18.2/tocbot.min.js></script>
25<script>tocbot.init({tocSelector:".js-toc",contentSelector:".pgcontent",headingSelector:"h1, h2, h3, h4",hasInnerContainers:!1,ignoreSelector:".formkit-header > h2"});const eltoc=document.querySelector(".js-toc");if(!eltoc.childNodes.length){eltoc.className="js-toc";const e=document.querySelector(".toc-header");e&&e.remove()}</script>
25</body></html>
Line numbers count LF bytes from the start of the resource, as the search results do. Vendor segments are library code the classifier recognised; they are stored but not indexed. Bytes are shown as Latin1 characters, one per byte.