How LTV.ai forecasts a campaign's revenue and margin before you send
LTV.ai forecasts revenue per recipient and click-through for every campaign idea, then ranks ideas by expected margin, not topline revenue.
Quick answer: For every campaign idea, LTV.ai predicts revenue per recipient and click-through using models trained on about a year of send and order history. Ideas are then ranked by expected margin, so a deep-discount idea that inflates topline revenue does not automatically beat a healthier full-price one.
Why forecast at all
Most teams decide what to send, then hope. LTV.ai estimates expected value first, so the ideas that reach a marketer are the ones most likely to pay. The forecast is what lets the daily briefing rank ideas by number, rather than presenting a flat list.
What the models predict
Two predictive models work together: one estimates revenue per recipient, tuned for the reality that most recipients generate zero and a few generate a lot, and one estimates click-through. Multiplied out across the audience, they produce projected revenue, projected clicks, and projected margin. Scoring runs on the finished email, with the real subject line and body, not the raw idea, so the forecast reflects what will actually go out.
Why margin, not revenue
Ranking by topline revenue rewards discounting. Ranking by margin accounts for discount depth and cost, so the system does not quietly push the brand toward giving away margin to look good on a revenue chart. This is a direct expression of the principle that performance should be real, not vanity.
Why cross-brand training
A single brand often sends too few campaigns to train a stable model alone. Pooling across brands gives the models enough signal to generalize, while brand-specific features keep the forecast tuned to each brand.
Frequently asked questions
Is the forecast a guarantee? No. It is a prediction, validated against real outcomes and improved over time.
Does it replace judgment? No. It ranks and informs; a human approves every send.
Part of the machine learning behind LTV.ai, a series on how the platform works under the hood.
See it on your store: book a demo.