This page exists to be probed, not to sell.
What Forward CLV rests on, the three terms that carry it, what goes in and what comes out, and how every play is measured against a held-back share of the same customers.
Standing on decades of academic work.
Tacet's forward value rests on probabilistic models of when customers buy, when they lapse and what they spend, a line of academic work with a long record of published validation.
For how a lapse gets a number, read Nobody knows when a customer churns.
- RFM scores (recency, frequency and money spent) and rules of thumb sort customers by what they already did. They don't forecast what any of them will contribute next.
- Because a probabilistic model makes a forecast, the forecast can be checked against what customers went on to do. A bucket that predicts nothing can never be proven wrong.
- Decades of peer-reviewed validation across categories mean the assumptions are known, stated and open to being proven wrong. Not a black box.
Three terms carry the whole system.
Each has a full entry in the glossary.
Customer lifetime value
The whole relationship
The total contribution a customer produces over the whole relationship: the margin on what they kept. Measured on history, it only looks back.
In the glossary: Customer lifetime valueForward CLV
The next 12 months
What a customer is expected to contribute over the next 12 months, after product costs and expected returns. One score per customer.
In the glossary: Forward CLVIncremental profit
Measured, per play
What a play added: profit from the customers it reached, compared with a held-back share of the same customers. Shown as a range.
In the glossary: Incremental profitWhat goes in, what comes out.
The score is built from your store's own records, and it leaves some costs out on purpose.
Goes in
Shopify orders, customers, product costs and refunds, over one read-only connection.
Comes out
A Forward CLV score for every customer. With the subscription, it is written onto each Klaviyo profile as named properties, updated daily.
Not deducted
Payment fees, fulfillment and ad spend. Forward CLV is a product-margin figure, not a fully costed one.
Where the number is weaker
- Where your Shopify catalog has no product costs, dollar figures rest on an assumed margin, and Tacet says so.
- Forward CLV needs enough repeat purchases to be reliable. Where a store's history is too thin, Tacet tells you plainly.
How a play is measured.
Every play is read against a held-back share of the same customers, on rules set before it sends.
Set before launch
The primary outcome and the read date are set before launch and frozen when your admin approves the play. It concludes once, on that date. The recommendation cannot chase a flattering week.
Given a range
Incremental profit and the change in Forward CLV each carry a 95% confidence interval. It comes from re-drawing the play’s customers at random, 2,000 times per read, and seeing how far the difference moves. The uncertainty you see is earned, not styled.
Ready to say no
A verdict comes back positive, negative or inconclusive, and an inconclusive result is never shown as a win. In a controlled trial, half the finding was an email we told the client to stop sending.
Intervals belong to what we measure. A play’s verdict comes from the 95% interval on its primary outcome, set before launch: clear of zero above is positive, clear of zero below is negative, and anything else is inconclusive. Weekly reads continue to the planned final week. Forward scores are point estimates, with no decorative ±.
Why a range rather than a single number: What a confidence interval is doing on a marketing report. The one completed trial, read week by week, is on the results page.
Probe it on your own data.
The Diagnostic is free. It scores every customer on your own Shopify history, over one read-only connection. Decide for yourself before anything directs a dollar.
Read-only access · No deck required · You keep the analysis either way