Solutions
Notice the quiet ones. Before they are gone.
A ready-made analysis measures each customer against their own usual gap between orders, and lists who has gone quiet, what their business is worth and a next step. Act on a row, remind them in your own words, and see whether it worked.
Against their
own rhythm
- 1Why, in one sentence
- 2Act on the row
Sounds familiar?
In the words sales teams use:
We notice a customer has stopped ordering at the quarterly review.
The big accounts get calls. The steady small ones drift away.
Our reports show who bought, not who stopped.
How it works at Tarnwick.
On Tarnwick Supply Co., sample data.
- 1
Add the analysis. Pick the system, pick "Customers going quiet", say when it runs (every day, twice a day, every week, or only when you press Update now) and who sees it (only you, every row, or each person their own rows).
Analysis & prediction - 2
The fields are mapped for you on systems with a ready pack, such as Xero, QuickBooks, Stripe, HubSpot and Odoo. On any other connection or saved query, map them yourself.
Data connections - 3
It measures each customer's rhythm. Each customer is measured against their own usual gap between orders, not one rule for all. A seasonal check against last year is there if you want it, and the thresholds are yours to change.
Analysis & prediction - 4
Each row says why, what the customer is worth a month, and a next step.
Analysis & prediction - 5
Act on a row. Send it through a follow-up flow, hand it to a team or a person, mark it handled, write a note, or ask the assistant about it.
Analysis & prediction - 6
Remind them in your own words. A company owner copies the ready flow "Remind customers going quiet" and starts it on a schedule: every morning it takes the rows the analysis marked for action, writes each customer a reminder from the text you write, asks you first, then e-mails it. Try it first: nothing is sent and nothing is changed.
Workflows - 7
See whether it worked. Of the rows you acted on, how many left the list within 30 days, and the money on them. It compares them with a small group it held back at random, and shows "not enough data yet" until 30 rows have been acted on.
Analysis & prediction
1Add the analysis
Orders (Odoo)
Customers going quiet
each person their own
every day
6Remind them in your own words
7See whether it worked
23 rows acted on. Not enough data yet: a rate shows from 30.
No rate until it
means something
Who is most likely gone.
- A prediction, trained on your own history, says when each customer's next order is due. It marks each one due, overdue or most likely stopped, recent ones first, ranked by the yearly value at risk.
- It is trained on your company's own records only, retrained every week, and sooner when it drifts.
- It needs 200 decided records with 30 of each outcome, and says how many more it needs.
- Each customer comes with three reasons, and the fixed line "These are patterns in your history, not proof."
Before a call.
The starter board "A company" shows one customer across every analysis, and what was done about it.
The starter board "The team" shows, for each account owner, what is urgent, what it is worth and what was done.
Ask the assistant what the analyses know about a customer before a meeting.
Who sees what.
- Each person can see only their own rows, by the owner column or by each reader's own sign-in.
- Share a prediction so each sales person sees their own customers, and each team lead sees the team's.
What you get.
A list each morning of who has gone quiet, why, what they are worth, and what to do.
A reminder in your own words, sent only after a yes.
An honest answer to "did it work?", with a group held back to compare.
Before you start.
- The analysis needs your customers and orders to be connected.
- Xero and QuickBooks first ask you to register an app with them.
- The same data gives the same answer: a second run the same day changes nothing. With too little data it ends as "not enough data", never a thin list.
- In the chat, asking who stopped ordering is not reliable yet. Use the analysis.
For sales managers, see Sales. Sales
Bring the names of three customers you have not heard from.
In the demo we connect one of your systems, and you ask the questions.