It’s a big week for analytics in Toronto. There’s a growing industry of digital intelligence / analytics, professionals in southern Ontario. It’s a brilliant and welcoming industry. This is the week when we get together, share knowledge, and welcome newcomers. The eMetrics Summit, the conference of the Digital Analytics Association (use the promo code BERSPK for a discount to the summit), will also mark second major Southern Ontario Chapter meeting. There will be case studies from TD, CBC, Bombardier, Intuit, The New York Times, TVO, Hyatt, and Maple Leaf Entertainment. Zoe Morawetz (TD) is showing us how they execute digital segmentation. Gareth Cull (Mozilla), Mark Dykeman (BMO) and Tim Ashby (CM) will be sharing which technical traps to avoid, Greg[…]

This piece from McKinsey highlighted the inflated expectations of big data analytics – “…expectations of senior management are a real issue…but too often senior leaders’ hopes for benefits are divorced from the realities of frontline application. That leaves them ill prepared for the challenges that inevitably arise and quickly breed skepticism.” The listicle (et tu, McKinsey?) summarized below, is somewhat related to that concern: 1. Data and analytics aren’t overhyped—but they’re oversimplified 2. Privacy concerns must be addressed—and giving consumers control can help 3. Talent challenges are stimulating innovative approaches—but more is needed 4. You need a center of excellence—and it needs to evolve 5. Two paths to spur adoption—and both require investment (automation and training) In a fit of[…]

There are varying concerns about what constitutes a causal model, the degree to which data is biased, certainty that the model is predictive about the future, and, that the model itself is a truthful depiction of nature. Over the course of the past two weeks I’ve talked with many people about their perspectives – data scientist, developers, technologies, product managers, brand managers, statisticians, consultants, professors, executive producers, and founders. We’ve talked about everything from why analysts and their customers won’t accept narrow models, why it’s far easier to summarize data than it is to describe the relationships in it, and the intractable differences between what is performance reporting and what constitutes an insight. The verdict is not in. There are varying beliefs[…]

This is a lot of inside baseball. The motivation is to share information while acknowledging that it’s wildly anecdotal. It’s directed at data scientists thinking about business. The Facts Andrew and I founded Authintic in late 2012. We landed three great customers. We met between 1,600 and 1,900 well wishers, competitors and prospective customers. Five major market hypotheses were tested. Revenue was earned and value was generated. Authintic was acquired by 500px in early 2014.   The Feels Thrilled. Very excited. And a tad skeptical about the lessons learned. People are terrible about extracting causal factors from an experience. I’m people. So I reckon that applies to me too. A sample size of 1 isn’t authoritative. It doesn’t constitute proof, or evidence[…]

The listicle is an amazing communication device. A listicle schema for communication – always in the form of a list. Sometimes that list is random, but, often ordered. I continue to be in awe of the ongoing effectiveness of the listicle. Lists are effective communication devices in analytics. Why not listicles? Lists Effective analytics dashboards are filled with lists. “The top 10 performing landing pages” “The top 5 posts” “The top 7 competitor ads…they don’t want you to know about!” Lists are visually compact and editorial appropriate. An executive might scan a list for the top performers and the bottom performers. An analytics executive might scan a list for the top 20% and verify that it accounts for 80% of[…]

The Circa app (As of January 2014) is notable for the choices the designers made. And the choices they made. The color palette is consistent. The leading is consistent and generous. Upcoming information is faded and effectively previews content. The  app can be used with gestures from one thumb, making it great for one thumb use. Just the right number of stories are presented on each day. They made quite a few good choices. They chose to hide most social sharing under a button, instead of surfacing all the options directly within the app. They chose to invest in making good recommendations about related content. They chose to invest in designing an elegant right rail breadcrumb that both respects the[…]

“The End of Facebook” trumpeted the headline. 46 points in 46 minutes on Hacker News. “Facebook Screws Social Media Marketers!” trumpets Business Insider. “Facebook is losing teens” states Global Web Index. Here we go with the bandwagon. Hop on! Only that this time isn’t going to be quite like the last time(s). Teens have fled to their smartphones They’re computers they can control. They’re computers that aren’t tied to the family room, where parents can seen them. Small screens offer a degree of privacy and intimacy that larger screens, even the tablet, just can’t replicate. Facebook saw that a long time ago and snapped up a few cool startups. Ditto Twitter. Ditto Google. And the rest of us are behind[…]

Speaking at a data conference is hard. Programming data conferences is hard. It’s damn hard. It’s hard to predict who’ll show up in your audience. It’s hard to predict if what you’ve planned to say will align with your audience. It’s even harder to predict if who you’ve chosen to talk will align to who is likely to show up. Image below most certainly related. I’ve had incredibly patient mentors when it comes to this. And I’m still optimizing. And I still find it tough. Heuristics for speakers: Shilling doesn’t work You never close a sale during a presentation. Worse, you turn leads off. Putting your ad first puts the audience last. If it’s about causing awareness of your product[…]

It’s the results, genius! It’s the results. The purpose of any sort of data analytics or data science is to get results. It isn’t about the spreadsheet that comes three weeks after the campaign. It isn’t about sandbagging numbers. It isn’t the few slides in the Quarterly Business Review. It isn’t even data entertainment. It’s the results. Great! So what’s the deal? Why is so much time expended on activities that don’t directly tie to getting results? Analytics Maturity It’s because of maturity, or the sum of experiences that an organization/culture chooses to remember. Very good models of analytics maturity exist. Stephane Hamel has a great one. Stances inform tools and tools cause experiences. Where you stand affects which, if[…]

Planning is preparation of the mind. It’s impossible to quantify every variable, every assumption, and every potential future state. Attempting to do so will simply boil the ocean and frustrate everybody around you. Analytics leaders tend to be very specific types of folk. Here are a few heuristics that might be useful for us in particular. Backcasting Backcasting is primarily an expression of preferences. The exercise almost always begins with an enunciation of a preferred, desirable, future state. Consider the following statement: “By 2016, we will be a 1 billion dollar company.” Such a statement, be it vision statements, stretch goals, or just goals, are typically not based on any sort of forecast. It’s entirely possible, and very likely, that[…]