There are at least five types of error related to analytics. These are instrumentation error, algorithm error, transposition error, statistical error, interpretation error. 1. Instrumentation Error When the instrument is measuring a phenomenon incorrectly. This is not to be confused with a human mistaking what an instrument really actually measures. Rather, this is when the instrument itself is only recording half of something. Or not measuring something at all. It’s akin to saying that thermometer is broken. Instrumentation has varying degrees of accuracy. For instance, the unique http cookie is subjected to fault as a result of a deteriorating cookie retention curve. The instrument continues to work just fine – it’s just that user behavior has changed, affecting its accuracy.[…]

An excellent analysis done by Allan Engelhardt, back in 2006 I suppose, talks about the 3/2 rule of employee productivity. The Coles notes version is that when you triple the number of employees, you cut their productivity in half. Check out the diagram below. Pretty scary right? Naturally, the story is much more complex than portrayed. Some sectors have mild slopes, like technology companies. Arguably, they’re using technology to flatten out the productivity slope. But it’s still slightly negative. Naturally, larger companies scale, so they still make more profit overall. Small companies are very good at doing many things. They become less good as they become large. And then ultimately, they stop being really, really good at anything at all.[…]

Konrad von Finckenstein, chairman of the CRTC, went before committee yesterday and made the remark: “The vast majority of Internet users should not be asked to subsidize a small minority of heavy users.” I take issue with Finckenstein’s statement. For one, the vast majority of Internet users subsidize a number minority. Urban customers, who are comparatively cheaper to connect with bandwidth, pay more to subsidize rural customers, who are comparatively more expense to connect. Didn’t the CRTC pass fees last year, forcing the vast majority of us to subsidize the viewing habits of the small minority of people who watch the CBC, Flashpoint, and DeGrassi? Isn’t the CRTC mandating the subsidization of something called “Canadian New Media”? Seems to me[…]

If you’re following a Canadian tech entrepreneur or scientist on Twitter, you might be noticing the #stopthemeter hashtag and statements questioning something called the #CRTC. The CRTC is the regulatory body responsible for regulating radio, television, and Internet Service Providers in Canada. On paper, Canada has six or seven major teleco’s. These are divided up by region. Telus and Shaw compete in the West. Rogers and Bell compete in Ontario. Videotron and whoever competes in Quebec. There are regional variants and government monopolies in the smaller provinces. Canada is peppered with duopolies – which are, in effect what economists might call natural monopolies. A complicating factor is that Canada is a massive country with a very dispersed rural population. These[…]

The space is tremendously fragmented because social itself if fragmented, unstructured, and ill behaved. Broadly, there’s ‘listening’, which has its origins in the PR space, and then there’s marketing performance, which has its origins in the analytics space. Although there are nearly 250 (+) ‘listening’ companies out there, none of them will have a solution that fits your unique set of circumstances, biases, and needs. You are simply not a PR person. Web analysts entering into social should be prepared to confront fragmentation and complexity on a level that they have yet to experience. If you work in an enterprise with more than 150 people, you will rapidly reach a stage where you will not be able to keep pace[…]

ETL stands for Extract, Transform and Load. They’re the three vital steps most analysts do before Analyze, Investigate and Storytell. Most of the time, the ET’ing is done for us. You log into a tool and hit export. The Loading part, getting the data into a format where it can be statistically analyzed or presented in a culturally acceptable way, is longer. And it’s where we spend too much time. But not today. I’m unpacking a tricky T problem. In an attempt to fully automate an algorithm further and unlock an area of possibility, involves a tricky operation of flattening lists of lists of lists, which, sadly for me, are also composed of lists. It’s tricky. There are functions that[…]

An insight is: New information Executable Causes action Profit results With that piece of jargon unpacked, the next one is ‘convenient reasoning’. Convenient reasoning is: An existing heuristic, hunch, feeling, belief, or instinct The seeking of validation or evidence Evidence to the contrary or modifying the position will be rejected Convenient reasoning differs from a hypothesis. A hypothesis is rejected if it’s proven wrong. No amount of evidence to the contrary will ever deter a convenient reasoner. Building cases in support of a project, plan, or prospect is an incredibly important skill. Rallying persuasive evidence is a key part of that. A whole industry was built around the provision of convenient facts. It’s an essential skill. Perhaps there would be[…]

Just as some people define they’re identity by what they buy, some people define themselves by the tools that they use. There’s a certain cache about using the ACH. Or being an OCP. Or knowing enough to choose select instead of forward regression. Or the use of Bayesian methods. Coremetrics against Omniture. Google Analytics over Webtrends. R over SPSS. Graffle over Visio. And so on. There’s a large degree of tool centricity in three communities: web analytics, data mining, and marketing science. The irrational judgements about people in each of those communities, based on the degree of sophistication of tools, is dangerous. Worse – it’s detrimental. It’s detrimental because it narrows your view. For one, different tools are right for[…]

It’s warmer today. Two points from Day 3: When pressed on what DM’ers thought of web analysts, some made comparisons to ‘convenient reasoning’. The comparison wasn’t made in a nice voice. That should actually really concern you if you’re a web analyst. Broad concern about devils in details on FTC ‘do not track’ list. Reflections: I have a much better understanding of the problems that face DM’ers, compared to the problems facing WA’s. Not all of them are interesting problems. Some of them are solvable if DM’ers and analysts work together. The linkage between ‘insight’ and ‘innovation’ has been finally, for the absolute first time in my mind, been completely made. I may actually calm down about the use of[…]

Did I mention how cold it is outside? A few points from day 2: Impressed with the transparency of Bill Whymark of GE. He detailed his segmentation process, in detail, and had a large number of slides which had been redacted by legal. Words x’d out in mid sentence. My key takeaway was the ongoing gap between value proposition writing (marketing / comm strategy) and segmentation. Bill was able to prove very remarkable lifts as a result of the segmentation as a result of operational improvements caused as a result of the segments (prioritization). Pretty powerful material. A second instance of creative override of segmentation was discussed in another presentation. An automotive enthusiast/under the hood segment was to be activated[…]