Customer health scores

A 0–100 customer health score computed from real Shopify order history — recency, frequency, and monetary value, with recency measured against each account's own buying rhythm.

Last updated

The short version

  • Every customer gets a 0–100 health score built from recency, frequency, and monetary value — the classic RFM model, computed from your real Shopify orders.
  • The recency half is measured against that customer's own order cycle, so a quarterly buyer is not punished for behaving like a quarterly buyer.
  • Scores decay in real time. A score is re-derived every time it is read, so an account that goes quiet gets worse on its own instead of waiting for an overnight rebuild.
  • Health is also written to your Shopify customer records as a report dimension, so you can group your own Shopify reports by it.

One number for “how is this account doing?”

A wholesale sales team carries a rough version of this number in their heads for the twenty accounts they know well. The health score is that instinct, computed for all of them.

It is a 0–100 composite of the three signals that actually predict whether a business customer keeps buying:

  • Recency — how far through their normal reorder cycle they are.
  • Frequency — how consistently they order.
  • Monetary value — what they spend, relative to the rest of your customers.

This is the RFM model, which has been the standard in direct and catalog selling for decades. What RevLogic changes is the recency half.

Recency measured against the customer, not the calendar

Most scoring systems measure recency as days since last order. For a B2B store that is close to useless, because your customers do not share a buying rhythm. A restaurant supply account might reorder weekly; an equipment dealer might reorder twice a year. Thirty days of silence means panic for one and nothing at all for the other.

RevLogic computes each customer’s average order cycle from their own history and scores recency against that. Being 20 days into a 30-day cycle is healthy. Being 20 days into a 14-day cycle is not. Same elapsed time, opposite conclusions — and the score reflects it.

Scores that decay on their own

A health score that only changes when data is rebuilt is quietly misleading: an account can go silent for a fortnight and still show the number it earned when it was thriving.

RevLogic splits the score in two. The stable part — frequency, spend, the shape of the buying history — is computed when a customer’s orders change. The time-sensitive part — recency, days overdue, the resulting priority — is re-derived every time the score is read. Nobody has to trigger anything. An account that stops buying gets visibly worse as the days pass, which is the entire point of watching it.

What a score is for

Health scores are a sorting and triage tool, not a scoreboard. In practice a sales team uses them three ways:

  1. Spot the slide. A large account whose health has dropped is the highest-value phone call in the building.
  2. Segment the base. Group customers by health band to see how much of your revenue sits with healthy accounts versus slipping ones.
  3. Report on it in Shopify. RevLogic writes a health band onto each Shopify customer record — strong, steady, slipping, or critical — so you can group your own Shopify reports by it. See RevLogic in Shopify Analytics.

What it does not do

The score does not contact your customers, assign them a grade they can see, or make decisions on your behalf. It ranks a list so that the limited number of calls a sales team can make in a day go to the accounts where they matter most. The judgment stays with the person holding the phone.

Where the score is used to decide that an account is actively slipping away, that becomes churn risk and a winback list.

Frequently asked questions

What goes into a RevLogic health score?
Three components, combined into a single 0–100 number. Recency is how far through their normal reorder cycle a customer is, so lateness counts against them relative to their own rhythm. Frequency is how consistently they order. Monetary value is what they spend, measured against the rest of your customer base rather than an absolute dollar figure. Higher is healthier.
Is a health score the same as churn risk?
No, though they move together. Health is a broad 0–100 summary of how good a customer relationship is. Churn risk is a narrower judgment about whether this account is in the process of leaving, expressed as high, medium, or low. A large, frequent buyer who has just slipped past their reorder date can still have a decent health score while carrying high churn risk — and that combination is exactly the call worth making.
How often are health scores recalculated?
The static part of the score — frequency, monetary value, and the shape of a customer's buying history — is rebuilt when their orders change and on a scheduled full pass. The time-sensitive part, including recency and how overdue an account is, is re-derived every time the score is read. That means a customer's health visibly degrades as they go quiet, rather than staying frozen until the next rebuild.
Can I see health scores in Shopify's own reports?
Yes. RevLogic writes a health band — strong, steady, slipping, or critical — onto each Shopify customer record as a report dimension. You can group, filter, and chart by it in Shopify Analytics and ShopifyQL alongside your own sales data. It is on by default and can be switched off in Settings.
Does a new customer with one order get a health score?
They get a score, but a limited one. A single order gives RevLogic no cadence to measure, so there is no purchase cycle and no meaningful recency judgment yet — the account is scored on what is known and simply is not flagged as overdue, because there is no rhythm for it to be late against.

In the help center