+44(0)1689 258 210 | enquiries@e-focusmarketing.com
- Our CRM Methodology
- CRM Strategy Solutions
- CRM Resources
- Supper Club
- Contact
+44(0)1689 258 210 | enquiries@e-focusmarketing.com
Customer Drift™ is the gradual weakening of a customer relationship across four dimensions: Attention, Behaviour, Relevance and Value.
It happens well before any change shows up in churn reporting or revenue, and it is usually visible in data a business already holds and is not looking at.
Customers rarely leave with a decision and a date. They drift, and they drift in a predictable order.
Most retention reporting is built to count outcomes, so it tells you how many of your customers stayed and how many left, and it tells you both of those things after the fact.
Drift is the process that produces that outcome, and by the time the outcome is measurable in your numbers the process has usually been running for three or four quarters.
That gap is why so many retention problems arrive as a surprise inside a business that was, on its own numbers (the ones anybody was actually reading), doing perfectly well.
Ask yourself what your reporting would need to show before anyone escalated. A five-point fall in repeat rate? A quarter of missed revenue? Now ask what the same reporting would have shown eight months earlier, when the relationship actually started to weaken.
In most businesses the honest answer is that it would have shown nothing at all, because nobody was looking at the things that move first.
The order matters as much as the dimensions do, because drift moves through them in sequence and each one is significantly cheaper for you to address than the one that follows it (roughly by an order of magnitude, in my experience of auditing them).
Attention is the first thing you lose and the last thing anyone measures.
Somebody who used to open, read and act on your messages starts skimming them instead, then starts leaving them for later, then stops noticing them altogether.
On most dashboards that reads as a marginal decline in engagement, comfortably inside normal variation, which is exactly why nobody in your business ever escalates it.
AI-summarised inboxes have made this harder rather than easier, because your subject line now has to earn attention from a machine (Gemini in Gmail, Copilot in Outlook) before it ever gets the chance to earn it from a person.
So the useful question is not the one your engagement report answers. Are the same people still paying attention? Or is a shrinking group quietly carrying an average that still looks perfectly stable?
Behaviour drift is the point where the relationship changes shape without anybody – on either side – deciding anything.
Your customers visit less often, the gaps between their purchases stretch out, and their progress through your journey slows, with every one of those movements small enough to be written off as seasonality (and in most businesses it is written off as exactly that, every year, until the year it cannot be).
Behaviour is where the earliest genuinely commercial signal lives – it precedes almost every lapse, and it is measurable in data you are already collecting and almost certainly not looking at.
Relevance is the dimension where good teams make things worse with the best of intentions.
A group starts to disengage, the business responds by sending more, and the extra volume carries the same blanket message that failed to land the first time.
Sending more to a drifting group accelerates the drift rather than reversing it, which is the single most common accelerant in the audits I run.
Relevance also decays quietly on its own.
A segment that was well targeted eighteen months ago now describes people who have since changed jobs, changed circumstances and changed what they need from you, and nothing in the segment definition noticed any of it.
Personalisation does not fix this, because relevance is a question about whether your message matches where somebody actually is, and a first name in a subject line (which is what most personalisation still means in practice) answers a different question entirely.
Value is where drift finally reaches the P&L, and it is the most expensive place to discover it.
Your order values slip, your repeat rates fall, and somebody reaches for a discount, always the fastest lever to hand, which works once, teaches your customers to wait for the next one, and permanently lowers what your brand is believed to be worth.
“The market is tougher this year” is very often value drift wearing a more comfortable explanation.
By the time Value moves, the other three dimensions have been moving for the better part of a year.
Every one of them was visible in your own data, and none of them was being watched, because your reporting was built to answer a different question entirely.
That is the real cost of drift: the length of time you had to do something about it and did not know you had.
How to diagnose drift in your own business
The Customer Drift™ Framework tells you where a relationship is weakening and in which dimension, and it deliberately stops there without telling you what to do about it.
What you do about it is a separate discipline, and it is the mirror image of this one – where Drift describes how momentum leaks away, Customer Momentum describes the five decisions that build it back.
Diagnose before you prescribe.
Almost every expensive retention mistake I see starts the same way, with a business treating a symptom it never traced back to a cause.
Customer Drift™ is one of the three Discover components of the eFocus Customer Progression Model, alongside Customer Momentum and the Next Best Customer Decision Framework.
If you want to know where your own relationships are losing momentum, a Customer Momentum Review is the diagnostic engagement built to answer exactly that.
No. Churn is an outcome that has already happened and can be counted. Drift is the process that leads to it, and it is measurable months or years earlier, while the relationship can still be recovered.
Attention drift typically appears well before any change in purchase behaviour, which itself appears well before revenue moves. In practice that means the earliest signals are usually available two to four quarters before a churn report would flag anything.
Work backwards. Start where you can already see a problem, usually Value or Behaviour, and trace it back to find which dimension moved first. That is where the cheapest intervention sits.
No. It applies to any Segment a business manages deliberately: prospects, subscribers, members, donors, lapsed customers. Anywhere there is a relationship to sustain, there is momentum to lose.
Often, but it costs considerably more than it would have earlier, and the usual reflex, discounting, tends to deepen the problem it is meant to solve.