Everyone wants to get eCRM right of course. But given that there are something like 40 different definitions of what eCRM is, just knowing where to start can put the kybosh on the grandest of ambitions. In 2003 some colleagues decided that what we really wanted, given increasing investment in digital, was to prove that relationships built and nurtured online actually delivered commercial results. And that's what we set out to do.
Today, eCRM in my view is simply the science of creating lasting digital-channel relationships that can be measured commercially. In other words, our ability to measure the results of tactical campaigns is all very well, but we need to attach results to our engagement with a specific customer over the course of the life of that engagement - which ideally should last much, much longer than a tactical campaign or two! So, how do the best practitioners go about it? Here are a few of the critical steps in creating an eCRM programme.
Understand what data you hold
Most marketers have a wide variety of different pots of data – transactional, email marketing, website traffic, in-store, loyalty club, competition entries, addresses with forgotten provenance. The first thing that has to be done is to find a way to make the data consistent (for example making sure “Forename” is matched up to “First name”). Next it needs to be centralised. This might be as straightforward as giving it to an agency who will give back a rationalised database containing everything they were given. You’ll probably find that a huge amount of the data doesn’t have a useful name attached to it (web traffic, or perhaps records that only contain first names).
You’ll almost certainly find that much of the data is out of date, or plain wrong (petitions signed Mickey Mouse!). You’ll also see that you probably have an awful lot of it.
The first task is to look at the data and start looking for patterns. You can do this in-house if you have a data team, or an eCRM or data agency can do this for you. Essentially they are looking for a way to segment the data into reasonable groups. For example, these groups might (if we focus on customers) include:
- Customers who spend more than £10, £20, £30, £40
- Customers who buy on particular days of the week
- Customers who respond to discounts versus those who respond to added extras
- Customers who have visited the website once before purchase versus those who visited four times.
That’s pretty straightforward. From this basic data we can probably derive a few insights which might be useful. For example, the above might show us that there is a correlation between the number of visits to the website and the value of purchases.
And there may be a correlation between the transaction value and how many times a customer has been contacted with what kind of call to action. Ideally the data will show relationships between a customer’s average transaction value (ATV) and the length of time they’ve been a customer.
So we’ve started to derive insights that can inform our strategy.
The data you already hold will probably show you some surprising things. One retail client had data that showed an ATV that was higher than expected – and almost every customer we looked at had a transaction value well below the average. Eventually we found a tiny segment of a few people who, every time there was a 50% discount offered on a DVD, would buy a thousand and eBay them for a profit. It gave us an unexpected marketing angle – we could promote specific titles to this specific segment without broadcasting it to everyone and wasting one of our (limited) contact opportunities – meaning we could send everyone else an offer that would be perfect for them instead.
So understanding what data you hold is critical. Eventually the holy grail of eCRM is a ‘“single customer view’”, where there’s just one database that all the sales, marketing, web and long-term data is fed into. For many that’s nigh-on out of reach, and in our experience of working with household name brands it’s not necessary in order to have a hugely successful eCRM programme generating massive revenues.
Turn the insights into a strategy
I’ve already mentioned some of the benefits of understanding the data – often it tells you what you need to do. In one of the above examples if four visits to the site makes for higher ATVs, then the marketing activity should concentrate on driving four visits. OK, that’s a bit simplistic perhaps, but you get the idea – eCRM basically works on this principle. Look at the data, see what it tells us to do, find out why (a whole subject in itself, which I won’t go into here), then establish a marketing programme that gets people to do more of it.
So we identify the different segments, and what each delivers commercially over time, say a year, and eventually a lifetime (this is called the ‘Customer Lifetime Value’, or CLTV). Then we prioritise. There’s little point in spending 10% of the eCRM budget targeting 10% of the base if that 10% only contributes 2% of the revenue. The segmentation tells us what to do. If we have limited resources, then broadly speaking spend the money on the segment with the highest CLTV!
And execute
Email is the cheapest way of reaching lots of different segments with relevant tailored messages at precisely the best time, as shown by the data.
It’s easy to allocate different creative executions based on the segmentation rules, and it’s straightforward to transfer data to email suppliers (unless you’re going via an eCRM agency who will manage the selection and broadcast process for you). It’s also – critically – easy to see the results in real time. Opens and click-throughs are OK as indicators, but properly integrated analysis is is central to any serious eCRM programme, and there are specialists in providing this kind of straight-through tracking. Ideally of course you’re tracking from email to sale and repeat purchase, and adding this information to the central database so you can see long-term trends and adjust your eCRM programme accordingly. This gets your by-segment ROI, and your CLTV, and informs refinements to the segmentation and comms strategies.
Great eCRM incorporates email, community hub sites, tactical campaigns (to grow specific segments or to fulfil business needs, like filling up under-booked hotels), sometimes even mobile. It’s tested, constantly, and honed until it produces reliable results time after time. It actually removes uncertainty and takes the guesswork out of marketing. Great eCRM can have a fantastic effect –like McCain Foods who changed engagement rates amongst brand resistors from 14% to 63% in under a year, or the improvement in ROI of 32% for The Sun’s Fantasy Football game this season. Done well, eCRM revolutionises marketing.
An abridged version of this article appeared in UTalkMarketing
