Platform · Sales forecasting
Ecommerce sales forecasting built from the customers you already have.
A trend line can’t tell you that your best cohort is slipping. LiftSignals builds its revenue forecast from customer-level predictions: who is likely to order again, how often and for how much, so the forecast changes when your customers do, and you can see why.
No credit card · Read-only access · Works with Shopify, WooCommerce, BigCommerce, Amazon, Square or a CSV
Definition
What is sales forecasting?
Sales forecasting estimates how much revenue a business will make over a coming period. Most ecommerce forecasts extend last year’s trend. A customer-based forecast adds up what each existing customer is predicted to spend, then adds expected new customers on top.
A CUSTOMER-BASED FORECAST
Forecast revenue = Σ predicted spend of existing customers + expected new-customer revenue
Returning revenue comes from each customer’s predicted orders and order value. New-customer revenue comes from your recent acquisition pace and seasonality.
| Spreadsheet trend | LiftSignals sales forecast | |
|---|---|---|
| Built from | Last year’s monthly totals | Each customer’s predicted spend |
| Explains changes | No | Shows which customers moved the number |
| Uncertainty | One line | A likely range |
| Updates | When someone rebuilds it | Every night |
Why trend lines miss
Your forecast says growth while your best customers are quietly leaving.
01
Top-line trends hide cohort decay.
New-customer growth can mask returning-customer decline for months, until it doesn’t.
02
A single number invites false confidence.
Inventory and spend plans built on one line have no room for the range that actually exists.
03
Nobody can explain a miss.
When the month comes in low, a trend line can’t say whether it was fewer new buyers or lapsing old ones.
How LiftSignals does it
A revenue forecast you can trace back to customers.
Returning revenue
The sum of each existing customer’s predicted orders and spend, from the same models behind lifetime value and churn.
New customers
Expected first orders from your recent acquisition pace and seasonal patterns.
Range
A likely range around the forecast, not just a single line.
Drivers
A breakdown of what moved the forecast since last week: customers lapsing, customers rising, new-customer pace.
Refresh
Recalculated every night as orders arrive.
REVENUE FORECAST · NEXT 13 WEEKS
Down $9,400 since last week: 62 high-value customers moved to high churn risk.
Illustrative read-out. Your forecast is built from your own customers, and the range is shown in the app.
What to do with it
Four decisions a customer-based forecast improves.
01
Plan inventory around a range
Order to the likely range, not a single line, so a soft month doesn’t leave you overstocked.
02
Catch a revenue gap early
When the forecast drops because valuable customers are lapsing, start the win-back before the month closes.
03
Set acquisition targets that close the gap
Compare returning revenue with your goal to see how much has to come from new customers.
04
Explain the number to your team
Show which customers moved the forecast this week, so the conversation is about causes, not just totals.
Proof on every play
Then prove the plays moved the forecast.
Every audience you send carries an automatic holdout: a random slice of those customers who don’t get the message. 30 days later you see what the campaign caused, not what it claimed, in dollars, with the confidence interval printed next to it.
Is it right for you?
Who customer-based forecasting is for.
A good fit if you
- Get a meaningful share of revenue from returning customers
- Plan inventory or ad budgets weeks ahead
- Have at least about 500 customers and 90 days of orders
- Want to know why a forecast moved
Not the right tool if you
- Need financial forecasting for your whole P&L
- Sell mostly one-time purchases to new customers
- Have just launched with little order history
- Need store-by-store retail forecasting
FAQ · sales forecasting
Sales forecasting questions, answered.
See the customer model underneath in how to calculate customer lifetime value.
How do you forecast ecommerce sales?
The simplest way extends past trends. A stronger way adds up what existing customers are predicted to spend and adds expected new-customer revenue, which is how LiftSignals does it.
How accurate is the forecast?
It comes with a likely range. The app shows how past forecasts compared with what actually happened in your store.
Can it explain why the forecast changed?
Yes. Each week shows what moved it: customers lapsing, customers rising in value and the pace of new customers.
Does it account for seasonality?
Yes. Seasonal patterns from your order history are part of both the returning and new-customer parts.
How much history do I need?
About 500 customers and 90 days of orders to start. Seasonal effects are clearer with a full year of history.
Can I export the forecast?
Yes, as a CSV, so it can go into your planning spreadsheet.
RELATED CAPABILITIES
Customer lifetime value →
Predicted 12-month value for every customer.
Churn prediction →
See who’s drifting away before they’re gone.
Marketing mix modeling →
Where the next ad dollar should go.
See next quarter’s revenue, customer by customer.
Connect your store this afternoon. Your first forecast arrives overnight.