The vocabulary, defined.
The customer-value terms we use, in plain language. Forward CLV, how it differs from historical LTV, the probabilistic models underneath, and the distinctions that decide where a dollar should go.
The core metrics
- Forward CLVForward-looking customer lifetime value
- The probabilistic estimate of what a customer will contribute from today forward, over a bounded window (typically the next 12 months), net of returns and discounts. It answers the only question that should drive spend: what is this customer worth from here. Distinct from historical LTV, which measures what a customer has already spent. Forward CLV is scored per customer, not averaged across a cohort.
- Customer Lifetime Value (CLV)The field-standard term
- The total profit a customer is expected to generate over the relationship, adjusted for margin and the probability they remain active. Originates from Fader and Hardie's 2005 probabilistic-modeling work. Tacet uses CLV, not 'customer equity,' which is a separate and easily confused academic construct.
- DeltaCLVThe change in forward CLV
- The change in a customer's forward CLV over a defined experiment window. Tacet's primary experiment-outcome metric: a positive DeltaCLV means an intervention actually increased a customer's expected future profit, measured against a control group rather than claimed. When DeltaCLV is zero, the intervention is retired.
- CLV ConcentrationWhere value sits in the base
- How forward CLV is distributed across a customer base. Value is rarely spread evenly. A thin top tier, often the top 10 to 15 percent of customers, tends to hold the majority of forward value. Concentration is what tells an operator where retention and acquisition dollars should go, and it replaces flat demographic segmentation.
- P(Active)Probability a customer is still alive
- The estimated probability that a customer is still an active buyer, given their own purchase rhythm weighed against how long they have been silent. Because ecommerce churn has no cancellation event, P(Active) is how you put a number on an invisible decision. A frequent buyer gone quiet scores low; an infrequent buyer with the same gap scores high, a distinction recency thresholds and RFM buckets cannot make. Paired with Forward CLV, it sorts a lapsed list into who to revive, retarget, or sunset.
- Expected contributionThe number that survives the P&L
- The margin-adjusted forward value of a customer, net of the cost of goods, returns, and discounts. It is the denominator for capital-allocation decisions, and it is not the same as gross revenue or engagement. A customer can top the revenue leaderboard and still be underwater on contribution once returns land.
The models underneath
- The models underneathProbabilistic, not heuristic
- Forward CLV is produced by probabilistic models in the Fader and Hardie lineage: one estimating how often a customer buys and when they lapse, another estimating what they spend when they do. Both are fitted per customer, not per cohort, which is what makes the output a score rather than a bucket. What separates Tacet is the layer on top: returns and discounts netted into the score itself, so the number is contribution rather than revenue.
- Holdout validationProving the model before trusting it
- The discipline of proving a model on data it has not seen. Fit the model on an earlier window of order history, predict the most recent months you have held out, then compare the prediction to what actually happened. A model earns the right to direct a dollar of spend only by first proving it on a past it never saw.
Distinctions worth keeping straight
- Predictive vs. descriptive CLVForward-looking vs. backward-looking
- Descriptive CLV, often called historical LTV, reports what customers have already done. Predictive CLV, or forward CLV, estimates what they will do next. Most dashboards are descriptive: they chart the past. Forward CLV is predictive, and only a predictive number can be acted on before the money is spent.
- Net of returns and discountsWhy revenue overstates value
- Revenue is booked at checkout. Returns land weeks later, after acquisition, shipping, and processing costs are already spent, and discounts came off the top. Scoring a customer 'net of returns and discounts' subtracts both, so the number reflects real contribution rather than gross sales. In categories with high return rates, the netting is the whole story.
The model behind these terms is documented on the methodology page, and the plain-language pitch is on the FAQ.
See these numbers on your own customers.
The free diagnostic scores every customer in your Shopify store by forward CLV, net of returns and discounts, and walks you through where your value concentrates. You keep the report either way.
Read-only access · No deck required · You keep the analysis either way