A marketing system that runs on real purchase data and keeps the four things a marketing team needs in one place: predicting which customers are about to stop buying (churn prediction), basket analysis of what is bought together, recommending the right products and offers to each customer, and forecasting sales and the effect of a promotion before the money is spent. AI and machine learning help plan prices, design campaigns, and cross-sell and up-sell.
Who it is for
Marketing and sales teams who regularly have to make decisions on pricing, discounts and product bundles.
Churn prediction: know before the customer leaves
Winning a new customer costs many times more than keeping one. The model reads buying behaviour, such as the time since the last purchase, falling frequency and a changing basket size, then scores each customer's risk with a reason anyone can read, so the team can send the right offer to the right person before they actually leave.
- Customers split into active, at-risk and dormant, updated daily.
- CLV, so you know whom to invest in keeping first.
- At-risk lists sent to a campaign in one click.
Basket analysis: what is bought together
This section is a table showing which products tend to be bought together (bread with butter and cartons of milk, for example). It is the starting point for Bundle or Combo promotions, and for reshaping the mix of a basket to improve margin. Instead of guessing what customers are likely to buy together, the marketing team can read it straight off actual behaviour.
Recommendation: the right offer for each person
The recommender learns from each customer's real purchase history and from customers like them, then picks the product, discount or bundle most likely to be bought. It works on the website, in LINE OA chat and in email campaigns, and is measured against a group that got no recommendation, so you know the extra sales really came from it.
Price Sensitivity & Optimization: finding the right level of discount
A bigger price cut does not always mean better total sales. The system uses AI to analyse the effect of discounting, measuring how sales change by discount level and by promotion code, then points to the Optimal Discount, the level at which total sales are highest. It can also assess past promotions retrospectively, so the next campaign is worth more than the last.
- Impact measured by discount level and by promotion code
- Finds the Optimal Discount at which total sales are highest
- Reviews past promotions to plan the next round
Prediction: see the result before spending
Before a promotion goes out, the system forecasts the likely sales and margin from past promotions and seasonality, so the team can choose how much to cut, for whom and when, then check the real result against the forecast afterwards.
What it does
- Churn prediction: which customers are likely to stop buying, and why
- Customer lifetime value (CLV) and RFM segmentation
- Real-time basket analysis of what is bought together
- Personal product and offer recommendations (next best offer)
- Sales and promotion effect forecast before launch
- The optimal discount level
- Target lists sent straight to LINE OA, email or the chat system
- Campaign results measured against a group that got no offer
Want something like this?
Tell us your situation and we will work out what needs adapting to fit it.
