Platform
The predictive analytics platform that scores every customer, acts on it and proves it worked.
A customer intelligence platform that runs everything a data team would build for a growing store, every night: who each customer is worth, who’s leaving, who’s about to buy and where to spend next. Every prediction becomes an audience, and every audience comes with a holdout that shows what it caused.
Every plan runs every model · No credit card · Read-only access
Ten capabilities
A customer intelligence platform for every retention and growth decision.
Each capability answers one question about your customers and hands the answer to the channel that should act on it. They share one nightly-scored customer record, so nothing disagrees with anything else.
01
Know your customers
Predictions on every customer, every night.
02
Act on them
Scores turned into audiences and offers.
03
Prove and plan
What worked, and where to spend next.
How it works
From connected store to proven result, with no data team in between.
01
Connect
Read-only access to your store, or a CSV. No pixels or scripts.
02
Score
Every customer is scored overnight, then again every night.
03
Send
Any score becomes an audience in Klaviyo, Meta, Google and more.
04
Prove
Every audience keeps a holdout, so you see what it caused.
The models
The predictive models underneath, and the question each one answers.
Each model is trained on your store alone, not a benchmark. Below 500 customers or 90 days of orders, a score stays blank instead of guessing. Accuracy is checked against your own data and shown in the app.
Proof on every play
Predictions are a promise. Holdouts are the receipt.
Every audience you send keeps a random holdout. 30 days later you see what the campaign caused, not what it claimed, in dollars, with the confidence interval printed next to it.
FAQ · platform
Platform questions, answered.
Comparing tools? See LiftSignals vs Klaviyo predictive analytics.
What is a customer intelligence platform?
Software that brings everything you know about each customer into one record and adds what you can predict: their value, their risk of leaving and what they’re likely to buy next. LiftSignals builds that record from your order history and turns it into audiences.
What should a predictive analytics platform do for an ecommerce brand?
Score every customer on what they’re likely to do next, send those scores to the channels that act on them, and measure what each campaign caused. Predictions that never reach a channel, or can’t be proven, don’t change revenue.
Do I need a data scientist to use LiftSignals?
No. The models are built and run for you every night. Results arrive as plain-English segments and audiences, and analysts can export the scores if they want them.
Which models are included in each plan?
Every plan runs every model, sends to every destination and keeps a holdout on every audience. Plans differ by customer count and workspaces.
What data does LiftSignals need?
Read-only access to orders, customers and products from Shopify, WooCommerce, BigCommerce, Amazon or Square, or a CSV. Ad spend from Google, Meta and Microsoft for marketing mix modeling.
How long until I see results?
Most stores see their first scores the morning after connecting. Very large stores can take longer for the first sync.
Connect your store this afternoon. See every customer scored by morning.
Every model, every destination, every plan. 30 days free.