Should I use sessions or unique visitors to work out my ecommerce conversion rate?

[joe-ainswoth]

Hi all

First question here but I've been lingering in the shadows for a while.

As part of my companies digital marketing plan for the next financial year we are looking at benchmarking against certain KPIs.

At the moment I simply report our conversion rate as Google Analytics displays it. I was incorrectly under the impression that it was reported as unique visits / total orders but I've now realised it's sessions / total orders. At my company we have quite a few repeat purchasers. So, is it best that we stick to the sessions / total orders conversion rate? My understanding is multiple sessions from the same visitor would all count towards this conversion rate and because we have repeat purchasers these wouldn't be captured under the unique visits / total orders method? It's almost as if every session we would have to consider that we have an opportunity to convert. The flip side of this is that on some of our higher margin products customers may visit multiple times before making a purchase. I should probably add that I'll be benchmarking data based on averages from the 1st April - 31st of March which is a financial year in the UK.

The other KPI we will be benchmarking against is visitors. Should we change this to sessions if we will be benchmarking conversion rate using the sessions formula? This could help with continuity and could also help to reveal whether our planned content marketing efforts are engaging users.

I hope this makes sense and thanks for reading and offering advice in advance.

Joe

[RyanPurkey]

Matthew makes great points. I'd add to this that having conversions tied to membership data makes it all the more person specific. This is why you'll here numbers like 74% conversion rate for Amazon Prime members (see: https://www.internetretailer.com/2015/06/25/amazon-prime-members-convert-74-time). Aside from better tracking you can begin to see the value for Amazon in having members...

1. Similar to Facebook they're collecting user data per person and building a massive user base aside from just sales.
2. Better tracking.
3. Higher conversion rates.
4. Top of mind branding.
5. Upselling
6. And so on...

You get the idea. That's why when you go to Amazon.com the only pop-up or animated prompt you'll see on the home page is to "sign-in".  Obviously, this could be something out of scope for your project currently, but food for thought down the road.

[Matthew_Edgar]

I'd argue there is value in looking at and benchmarking both numbers, though you might not get an accurate picture of both through Google Analytics. You want to know how many sessions ended up in an order, regardless of how many repeat customers there were that converted. As you said, every visit could end up in an order (you could get a little more detailed and segment to clarify just how many sessions qualify) and you want to know just how true that is. At the same time, you want to know how many unique people placed an order as well and repeat order rates.

Here is the tricky part. Google Analytics is pretty good at telling you how many sessions resulted in an order (the conversion rate you see in goal reports is goals per session). With the Time to Purchase report, you can get a fairly decent idea of sessions it took for those higher margin products. Now the other side: unique users. Users is wonky in how it is calculated (for instance one customer uses different browsers/devices or your customer deleted their cookies) so knowing how many users converted won't always give you the number you are after and, in my mind, it isn't reliable enough to benchmark.

What I do to get at the number of unique customers and orders per customer is use other tools (CRM, order system, etc.) to track that number--those systems are designed around people not sessions, so you are going to get a far more accurate picture of how many unique people placed an order. That is your benchmark, but you can map order dates and/or transaction IDs to GA so that you can understand traffic patterns for repeat customers and how they might differ.

Hope that helps.