By Gary Angel|
January 15, 2017
I find I’m so excited that I can barely sit still or hold a thought in my head. I think it’s the excitement only a free man can feel. A free man at a start of a long journey whose conclusion is uncertain. I hope I can make it across the border…I hope the Pacific is as blue as it has been in my dreams. I hope. – Red, Shawshank Redemption
In July of 2006 I sat down for an interview in the second floor of a two story, nondescript, red corrugated iron office building neighbored by a scrap yard. I was a couple months removed from graduation at tiny liberal arts college in Walla Walla, Washington, but I didn’t look a day over 15. Joel, a gray curly haired Jewish grandpa figure, asked a lot of questions and his business partner Gary appeared in perpetual deep thought. The two needed to replace an analyst at SEMphonic, their three-person web analytics consultancy. Exactly, or even approximately, what that entailed, I hadn’t the foggiest.
Buried behind a dozen copies of my resume, that featured such illustrious titles as ‘lead camp counselor’, ‘photography dark room assistant’ and ‘first base umpire’, I pulled out one page of bullets from an analysis Joel requested of CyberTrader’s website. An earlier failed interview at an SEO firm taught me these web people like segmentation, so I had sliced and diced the site with the precision of a toddler with Play-Doh and a butter knife.
It turned out to be good enough. After Joel and Gary’s first choice went to work as a bank teller, I was hired. I ended my running Excel list of applications in the high 200s, canceled interviews with PR, wealth management and online marketing companies and inadvertently picked a career path.
It wasn’t supposed to be that way, of course. A first job at a tiny tech(ish) company headquartered in Novato, California should have a shelf life of 12, 18 months max. Then you move onto bigger and better things. But on day one I got a login to a tool called HBX that housed Charles Schwab’s web behavioral data. It felt so right and so wrong. How could I, a kid who knows nothing about nothing, have been handed the keys a Fortune 500’s customer behavior?
When it comes tobacco, alcohol and gambling I can give or take them, but with data I’m an addict. As my digital data dependency grew so did the industry, the tools and Semphonic. There was a Great Recession. We grew. The economy improved. We grew faster. And my measuring stick isn’t revenue, margin or headcount (we did, however, crush them all YoY) but progression and mindshare.
The X factor to the whole Semphonic equation turned out to be the person from my interview in perpetual deep thought – Gary. He’s the reason why my original 12-month job shelf life estimate was off by a factor of 10. Somewhere in between Gary dropping, without instruction, a massive deep-dive analysis on me in week one and Gary being voted 2012’s ‘Most Influential Industry Contributor” by the Digital Analytics Association, I recognized I was in the right environment to learn and thrive.
The culture Gary fostered at Semphonic was that of trying something new. Failing. Or succeeding. Tweaking slightly. Or significantly. Then trying again. I had unknowingly joined a startup. I loved it.
It’s hard to tell the difference between being ahead of the curve or setting a trend. I’ll say at Semphonic we did the former; it sounds (slightly) less arrogant. We kept the RPMs of the digital behavioral tools analytics tools in the red – always building segments, reports and analyses slightly more complex than the tool was intended to handle. But while we added moving parts to our techniques, we subtracted them from our insights. Semphonic pioneered methods of customer centric recommendation delivery, shifting entire organizations from thinking solely about the pages and site sections to thinking about those who were using them.
With the acquisition of Semphonic by Ernst & Young three and a half years ago, ideas like functionalism, use case analysis, two-tiered segmentation and digital valuation all ping ponged around with a little less velocity. That’s not a knock to EY, just the reality of going from having 49 coworkers to 230,999.
But while that river of entrepreneurial ideation was dammed down in favor of sizable pursuits and scaling capabilities, the stream of entrepreneurship was never fully arrested. With time, that gradual flow built-up and became the force behind Digital Mortar.
Over the past decade to say that I’ve been a heavy user of digital behavioral analytics tools would be putting it lightly. I’m a freakin sumo wrestler. And they’re pretty good. I had already been doing some work on digital in-store integrations (signage, experiences, ect.) when about a year back I started to get exposure to retail analytics tools. Coming from the digital world, seeing retail analytics tools made me, a number nerd analytics addict, want to cry.
Foundationally store behavioral data didn’t feel like it should be treated dramatically different than web data. In both environments, the customer starts in a certain place, go somewhere next, then somewhere else, shows more interest in some things than others and performs an action that impacts the businesses bottom line. The ‘best of breed’ retail analytics tools must not have seen it this way. Across the board, they lacked the ability to accurately track a customer end to end. I’ve pulled some pretty good insights out of pretty shitty web analytics measurement, but I’d never fathomed having to optimize a customer base using partial data from the home page of a single tiny microsite. I could see game changing segmentation, testing and optimization strategies playing out throughout physical retail, but they were dependent on the data that wasn’t being collected.
I spent the last year working from the premise that there must be a reason why retail analytics tools suck. I tried to find dead ends. Places where hardware or software limitations fundamentally restricted what was possible in physical environments. I hit speed bumps, pot holes and I nearly side swiped a freeway barrier (true story – I did all those things on a 1st date when I was 17;15 years later we’re married with a 3-year-old) but the dead ends never came.
The itch to innovate. Mastery of a digital discipline that can perfectly translate. A physical world that absorbs 93 cents out of every retail dollar spent being analytically served by deadweight. Gary, a careerlong partner and friend who after (literally) writing the book on the digital world, can’t wait to start again with a blank slate. And a product path, albeit challenging, that will not just differentiate but may change retail fate. Checkmate.
Around the age of 12, as the realization sunk in that, even though I could hold my own in the West Marin Little League, I wouldn’t make MLB, I started answering the ‘what do you want to be when you grow up question’ with simply ‘I’m going to go into business’. I mostly said it because I didn’t want to follow in my parent’s footsteps and become a Shrink. But screw it. Here I go.
Welcome to Digital Mortar.
- The Basics of Display Measurement
- Merchandising Analytics: Measuring In-Store Displays
- Unintentional Model Bias: Why Laziness and Census Data can be good proxies for racism
- Screening Off – The right answer for thinking about model bias
- The Problem of Correlation: Why you don’t need to have a variable to use it