Conversion Secret #2: No More One-Night Stands

Every year I spend the entire first week of September going to birthday parties.

Do the math.

No one had a baby AT the New Year’s Eve party I went to this year. Does that mean no one got pregnant that night?

Anything worth anything in this world has a gestation period. This is especially true for decisions regarding purchase, adoption, and loyalty of products/brands/companies/services.

Introducing the least valuable analytics report in the world:

People love this chart. They swear by it, they check it every morning. Unfortunately it is meaningless and - worse - misleading. Why? Submitted for your approval are the following scenarios:

1) You’re killing it and the chart is tanking.

The denominator for the metric in this chart is visits (nb. Avinash Kaushik rightly suggests viewing this report with Visitors or Unique Visitors as the numerator instead of the default behavior of many tools).

So, your visits going up faster than your daily purchase volume would lead to a decline in your conversion rate. OH NOES! WHERE WENT TEH MONAYZ?

Someone, somewhere just said: “hold on. if I’m getting more traffic that isn’t converting as well as my current traffic, isn’t that a problem?” The short answer is “maybe” and the long answer is “there is a chance that your entire measurement worldview is borked, any chance you can clear your afternoon so we can grab a sarsaparilla and discuss this?”

To find the answer we should drill into this possibility of higher volume and lower conversion rates being a positive thing. Hmmm, what if …

2) Your new users now have a crush, but are playing hard to get.

Daily conversion rate over time reports measure love at first sight, not likelihood of dating or marriage.

A scenario: you have embarked on a quest for new and profitable channels for user acquisition. Maybe the dynamics of your SEM campaigns has changed radically in the last 18 months (did I just hit a nerve?) and you need to find better ways to spend your marketing dollars.

You try social media campaigns. Your traffic goes up. The traffic doesn’t appear to be converting, so your conversion rate tanks and you get yelled at by someone. You tuck your tail between your legs and go back to AdWords to resume the self-flagellation. 

What if you knew that:

  1. Traffic you initially acquired from Facebook actually converted really well.
  2. These fine people started converting after 1 month of familiarity with your brand, and it took 12 months for the entire group of converting users to mature and eventually convert.
  3. These people interacted with your content and social media properties many times before purchasing.
  4. These people tended to promote your brand to their friends in a variety of explicit and implicit ways, and in turn was actually generating compound growth of new leads coming to learn about you.

These insights are completely different from those that conventional web analytics tools reveal. How would you get numbers on this? You’d need, one might imagine, a trend graph (daily/weekly) of revenue generated by users who originally came to your site from a certain domain? Then what else, a report on how long it took them to convert (maybe week by week or month by month for the cohort)?

If you had that data maybe you would turn right around and double down, since you have a handle the expected lifetime value of this traffic.

The answer is lifecycle analytics. Stay tuned.

49 notes

Show

  1. willfleiss reblogged this from markitecht
  2. dcancel reblogged this from markitecht
  3. markitecht posted this

Blog comments powered by Disqus