Cookies - of both the baked and the web variety - are a wonderful thing. From an ecommerce perspective, they enable us to gain significant insight into customer behaviour: insight that conversion hackers can use to their advantage to message customers with stuff that really resonates.
In the video below, Dr Mike Baxter discusses what online retailers can do to improve their customer retention by messaging their customers more intelligently. We should also mention that this, alongside eleven other ecommerce conversion hacks are available in a free download, which you can read here.
You should be able to identify a whole bunch of your customers by cookie - you can see from analysis of your customer data that some are about to lapse and if, per chance, they should come onto the website, do not let them leave without buying something.
There happens to be an intersection of ‘I know them’, ‘I recognise them’, ‘they're on the site’, ‘but they haven’t bought for 6 months’ - and in my customer base that means they’re about to lapse.
So message the heck out of them - onsite, retarget them, email them, before you lose them because not only do you know that they’re a customer, but you have an indication of their value and you also have got some measure of what the value would be to get them back so you can do some reasonably accurate calculations about how much it might be worth discounting, free P&P-ing them, free gifting them - whatever it takes to make them make that extra purchase before you lose them.
And of course triangulate your customers across devices. I know it’s easier said than done but one of the things that I still find some people are not doing is tagging their emails with an individual identifier so that if someone buys on their laptop and opens an email on their tablet they can be connected - you have now got to cookies pointing to the same person so they can be triangulated across devices. So you don’t need me to purchase on each device before you know who I am.
The final thing to say about customers is back-match them. It sounds like a lot of work - and admittedly it’s a lot of work to set up - but once you’ve got it working it’s a routine probably job that you run once a week. The thing is, if you imagine that you’ve worked really hard to earn yourself a ‘hero customer’ and then they get married and change their surname - is that a new customer? It certainly ought not to be. What if they move home and get a new address - is that a new customer?
When I do analysis of customer data for clients I often find massive duplication - customers with many identities within a database. So it really is worth checking. And if you keep triangulating on a regular basis you can keep up with hero customer irrespective of how they try and disguise themselves.