(This is a business case study. It will be used to guide discussions during the session: “Analytical Tools” at the Vendo Partner Conference in Barcelona on Tuesday, September 13th.)
Matt was standing in Times Square looking up. He thought about how many times he’d been there before – through the movies and TV. As a kid he’d watched Superman battle Lex Luthor in this very place. More recently Iron Man and countless others had filmed key scenes in the square. And each New Year’s Eve the ball dropped while a million people celebrated in person and hundreds of millions more joined them through TV from around the world. The crowds were always thick. Having seen it so many times before, actually being there felt like déjà vu.
Matt checked his phone. In one minute the massive billboard in front of him would switch to an announcement. It was ten years in the making. The crowds would join Matt in celebrating the birthday of his online brand. It had gone from zero to one of the biggest brands in the world and it was time to party.
He had a good idea how many people would join him. On an average day 330,000 people pass through the square, which was actually shaped more like a butterfly than a square, and 460,000 when it was busy.
All of a sudden there it was. He felt a rush of excitement. It was electric. He couldn’t stop smiling.
Matt’s online brand had been built on analytics. If this moment was online (not IRL) he would have tons of data about himself and all of the people around him. He’d know exactly how many were looking, he’d know whether they visited the site, how many of them bought and how much they spent. On the Times Square announcement he’d given a code for free membership. It was easy for him to count how many people used it. But that was all he he could track.
He felt like a successful car maker who had just bought a racehorse. He was making an homage to tradition, to how things used to be, connecting himself to those advertisers who came before him. A billboard in Times Square certainly had its charm. He could feel it. Matt celebrated the announcement with friends and colleagues who’d gathered with him to see it in person.
At the end of the night he went back to his hotel room, flipped open his laptop and pulled up the stats for his online business. It had been just a few hours since he last checked. During that time his sites had seen more online visitors than could squeeze into Times Square’s on New Year’s Eve, millions more. And he could track everyone of them. He thought to himself, “analyzing ads…impossible in real life, measurable and indispensable online.”
He imagined again if that billboard had been an online banner. He’d track people who saw it (impressions) and people who did the equivalent of typing in the special link (clicking on it). He would know the effectiveness of the banner and precisely evaluate its return for his business. He’d have demographic and geo/time data. It would help him see trends, for example, that a visitor between the hours of 9pm and midnight was three times more likely to click than someone visiting from noon to 3pm. He would learn from this data. He would optimize his ad purchasing to improve his return relative to his cost.
He would answer questions like:
- How much is a member really worth?
- Does that amount justify my expense in acquiring them?
- Do my membership options make sense as a group? Each one may not be worth as much as the other but am I effectively maximizing my conversion and cumulative return?
- How many ways can I cut this up?
He thought of all the ways that visitors were different:
- ISP or connection type
Then he thought about his costs. “Are my affiliates equally profitable? Why or why not? Are my traffic sources equally profitable? Am I comparing one to the other?”
Next he checked his risk stats. VISA was about to tighten the limits of a critical risk monitoring program relating to bank-initiated refunds, or, in other words: chargebacks. Matt needed to think outside the box in order to respect these new limits.
Banks need to cooperate for a customer to chargeback. Some banks are very enthusiastic, priding themselves on their high levels of customer support. A customer says, “I don’t recognize this charge” and the bank immediately begins the chargeback. Other banks, however, prefer not to go through the hassle and effort of a chargeback. It just depends on the bank.
Knowing the difference between these two types of banks requires large data sets. If the sample is too small the data just looks like noise. Once you have the data you can detect and prevent either the transaction or the chargeback based on the bank’s history. When you have the data you can use it to make smart decisions. You’re no longer blind.
Matt closed his laptop. He still felt the glow from his experience earlier in the evening on Times Square. He’d never forget it. Yet at the same time he felt re-assured by the stats he had just reviewed. He knew that without them he’d never have been able to afford such a big birthday party.