Data Assumptions, Part 1: The Big Three

My grade-school teacher would help the class learn to spell words with mnemonics: abbreviations (RHYTHM: rhythm helps your two hips move), rhymes (“I before e, except after c…”), and components (lieutenant: lie u ten ants). We especially loved the trick involving the word “assume.” Remember what happens when you assume?

It’s funny how exercises like those pour negative connotations on a useful word like “assume.” It was reinforced at home, as my mom was always saying, “Don’t assume. You have to be 100 percent positive.” The whole notion of being sure, making certain, and leaving nothing to chance seemed to be at the core of my early education.

As a professional marketer and business leader, I now make assumptions every day. That includes sizing up the market, assessing competitors’ actions, determining information’s veracity, and gauging my team’s behavior. If I had to wait until I had every piece of data and 100 percent certainty, I’d never accomplish anything.

If assumptions are unavoidable, the key to relying on them is knowing when to use them and when to dismiss them. We often assume the data that feed our marketing decisions are correct, but they may not be. We may be forced to gut-check our assumptions and recognize when we’re committed to 100 percent accuracy and when we’re not.

Big Assumption No. 1: Analytics Data Is Accurate

Many of us have been assuming this for a long time. Your Web team members come to the table repeatedly to ensure that everything’s properly tagged and that they’re capturing everything of interest happening on the site.

Unfortunately, they often report that data included the site’s internal visitors instead of excluding them or that the 20 new offers posted last month weren’t tagged exactly the same as the offers from the previous month. In other words, you assumed one thing and found something else to be true. Did you get bitten by your assumptions here?

Big Assumption No. 2: Offline Conversion Rates Are Correct

How do you know what happens when a customer leaves your Web site and heads to the store? It’s easy to see conversions online, but when people venture into the real world, patterns quickly muddy. Even in tightly controlled situations where customers are referred to specific offline fulfillment channels (e.g., auto dealers), the data trail from click to purchase can break down at any number of places.

Big Assumption No. 3: A Lead Value Is Valid

How much is a customer really worth to you? Some businesses have done a comprehensive job of determining this through customer acquisition costs and average customer revenue (e.g., in the mobile phone space marketers know the average revenue per user precisely). You’re likely building very elaborate ROI (define) models for your marketing programs that are based primarily on this assumption about a lead’s value.

Could the value be wrong? How wrong? Is it too high or too low? What should you do?

Only three big assumptions and everything you know about your business could be dead wrong. It seems like there has to be a better way.

Next time, I’ll share other big assumptions and help you see why everything will be OK. I’ll walk through a sample model, based on multiple assumptions, useful and relevant for driving a multibillion-dollar business. Until then, assume the best about the data you already have.

Shane is off this week. Today’s column ran earlier on ClickZ. Be sure to check out part two!

Related reading

Report: millennials are more likely to share an ad (but also to mute it)
A graph showing a small pile of money at one end and a larger pile at the other, with an ascending line connecting the two.
hp
A cartoon depicting web analytics. It shows a bubble with the letters WWW in it, surrounded by "click! click!". An arrow leads from the bubble to a notepad with graphs and charts on it. Another arrow leads to a chart on a piece of paper, with a lightbulb next to it, which leads on to a spanner with the words "tweak! tweak!" next to it. In the bottom left corner is a box with four bullet points in it: gather, report, analyse and optimise.