Every Monday.
I sit and look at a list of each item of business that I want to accomplish during the week. Then on Friday, I make a new list of all the things that I accomplished during the week. And since there is no overlap in the list… I get to re-use my Monday list every week.
Efficiency.
The Zombies are not a problem until they are…
We bucket traffic into 3 types of visitors.
Human - beings with wallets
Zombies - being who have no business being on your website
Bots - non-beings sent by beings with bigger wallets
Zombies are not often called out because most analytics tools err on the side of accuracy, whereas I rather err on the side of being practical.
Labeling traffic is not and cannot be 100% accurate, and since analytics platforms sell the illusion of accuracy, they don’t introduce functions they see as less precise.
But sometimes a baseball bat is more useful than a scalpel. For example, in baseball…
…and ecommerce CRO.
So why label zombies?
Like bots, Zombies are not on your website to shop. Well, shit, that is wrong too. Let’s rephrase that — Zombies are not shopping. They are humans that landed on your site and didn’t engage with the page. So that means one of two things:
A)
they are onlookers, the perusers, tire-kickers, the curious — in which they are not a problem at all. They are looking for their own reason, not intending to buy anything.
And for that reason, they are not applicable to the behavior we want to examine in order to come up with a hypothesis for increasing engagement/conversion/whatever.
or
B)
They are a strong indication that your store just sucks. A higher rate of zombie traffic on a particular page might be telling you that they page is confusing or that it loads too slow. Or that your obnoxious pop up irks people. Or that your ads don’t match your landing page.
So we:
Remove them from the CVR calculation = more accurate conversion rate
Separate them for traffic quality = better indication of page health
Traffic Quality by Source
This is one of my favorite features of Mouse Whisperer.
Here are the traffic quality results from a top product on our site. We tracked 1000 human visitors out of 1342 total:
Bots 20 - no problem
Zombies - 322, don’t know if high or low, but we got a baseline
CVR - 7.3%, same, we got a baseline
Just from this initial snapshot, there are a few areas that need digging into deep:
Social Channels — too many zombies. Bad traffic? Or mismatch in posts/ads and landing page
Referrals — lowest zombie rate, look into where those referrals came from
Normal but significant drop from ATC > Conv — explore ideas to cut that down a bit.
Zombies are not tracked by other AB testing tools as far as I know. Maybe they approach it differently. But however you want to slice it, your conversion rate is heavily affected by the quality of traffic, and I hope that by measuring it across various channels you have more insight and more ideas to test.
- Shep