Getting started
Open the file or system where you keep your sales records (your POS system, a spreadsheet, your accounting software). Export a transaction history that has at least three pieces of information per row: a customer identifier, a date, and an amount spent.
Save it as a plain-text CSV file, or copy the relevant rows to your clipboard.
Go to the LoyaltyLoopAI app page. No signup, no login — just a blank text area and a file upload button.
Paste: Paste your transaction data directly into the text area. Or Upload: Click the "Upload CSV file" button and select your file.
The expected column format is:
C001,2026-05-14,42.50
C001,2026-03-02,38.00
C002,2026-07-28,95.00
The tool recognises several common column names. If your headers are different from the example, check that they contain recognisable keywords (like customer, date, amount).
The analysis runs in your browser — no data is sent anywhere. In milliseconds you will see summary cards (how many lapsing, at risk, and steady customers) and a full table with each customer's recency, frequency, and spend details.
Click "Show math" next to any customer to see the exact arithmetic that determined their status.
Frequently asked questions
Does LoyaltyLoopAI store my customer data?
No. All analysis runs entirely in your browser using JavaScript. Your transaction data never leaves your computer — it is not sent to any server, not stored on any database, and not accessible to the operator. Closing the tab or refreshing the page removes all data from memory. See the privacy page for full details.
What format does my data need to be in?
Comma-separated values (CSV) with a header row and one transaction per line. Three columns are expected:
- customer_id — any identifier for the customer (name, code, email prefix, loyalty card number)
- date — transaction date in YYYY-MM-DD format
- amount — numeric value for the transaction (e.g. 42.50)
The tool recognises alternative column names such as customer, transaction_date, total, spend and others — but the simplest approach is to use customer_id,date,amount as your header row.
What do "lapsing", "at risk", and "steady" actually mean?
The tool uses two simple threshold checks, both shown in the results:
- Lapsing — last purchase was more than 60 days ago
- At risk — last purchase was 30–60 days ago
- Steady — last purchase was within the last 30 days
These are configurable thresholds in the source code (60 days and 30 days). They are not predicted by a model — they are simple arithmetic based on recency. Every customer's detailed numbers (days since last visit, total visits, average interval) are shown so you can apply your own judgment.
Do I need to create an account or sign up?
No. There are no accounts, no signup, no login, no email verification. The tool is fully functional with zero registration. There is nothing to sign up for — the analysis is a static page with no backend.
Is the tool really free? Are there hidden limits?
Yes, the core analysis is free. Every feature that currently exists is included with no rate limits, no file size limits (for reasonable files), and no upsells. The planned Pro plan is listed on the pricing page as a preview of infrastructure-dependent features that do not yet exist. There are no hidden charges.
Can I get email alerts when a customer is about to lapse?
No. The tool has no ability to send email, SMS, or any form of notification. It is a one-shot analysis tool: you paste data, see results, and the session ends when you close the tab. There is no email sender connected to this service, and scheduled or automated checks are not supported. This limitation is by design — the tool is intentionally simple and runs entirely in your browser without a server.
What browsers does the tool support?
Any modern browser with JavaScript enabled: recent versions of Chrome, Firefox, Safari, and Edge. Internet Explorer is not supported. The tool is designed for desktop and tablet screens — it works on mobile browsers but the table layout is more readable on a wider screen.
Known limitations
No data persistence
Everything is ephemeral. Refreshing or closing the page destroys all analysis and input data. There are no accounts, no saved histories, and no way to return to a previous session. If you want to keep your results, take a screenshot or copy the table data before closing the tab.
No email, SMS, or push notifications
The service cannot send any outbound messages. There is no mail server, no SMS gateway, and no push notification infrastructure. Alerts, digests, reminders, and scheduled re-checks are not available — not on the free tier, not on the planned Pro tier. The product is a self-service analysis tool, not a monitoring service.
Pro features are not available
The Pro plan described on the pricing page (historical trend view, customer list export, larger CSV support) requires server-side infrastructure — accounts, a database, data persistence. That infrastructure does not exist yet. No launch date is set. The checkout form on the pricing page is a demonstration only and does not process real payments.
No CRM or POS integration
The tool does not connect to any external system. You must export or copy your data from your POS, accounting software, or spreadsheet and bring it to the tool manually. There is no API, webhook, or plugin.
No mobile app
There is no native mobile application. The website works on mobile browsers but is optimised for desktop and tablet usage.
Simple recency-based thresholds
The analysis uses straightforward recency thresholds (30 and 60 days). It does not use machine learning, predictive models, or behavioural clustering. Customers with frequent small purchases and customers with infrequent large purchases are treated by the same arithmetic. The detailed numbers are shown so you can apply your own judgment, but the classification itself is intentionally simple.
No contact method for support
Currently, there is no mailbox, contact form, or support channel. The operator has not yet established a way to receive customer support inquiries. See the "Contact us" section below.
Contact us
The operator of this service has not yet set up a contact method. When one is established (email address, contact form, or other channel), it will be listed here and in the website footer.
In the meantime, the most helpful thing you can do is explore the tool and the information on this page. Most questions are answered by the FAQ above.
If you are the owner of this site and need to set up a contact route, record the requirement in needs.md in the project root.