This week I thought I would write about the goldmine of data that our customers sit on and how best to use it with the help of some simple AI prompts.

‍

The MYPOS Connect portal has numerous reports that detail sales transactions. On the surface, these are standard reports: product codes, descriptions, classes, receipt numbers, values, tax, user, and device for example. The critical ingredient though is the receipt code so we can do some basket analysis.

Unlike simple sales summaries, this dataset (the reports) is transaction-level.

‍

With some AI help you can move from:

“What sold?”

to:

“What sold together?”

“What patterns exist in buying behaviour?”

“What drives higher basket values?”

AI spots patterns, they might not be obvious to us. For example:

• Hidden product pairings

• Staff-driven sales differences

• Device/location-based behaviour

• Missed upsell opportunities

Here are some examples of how to do this in simple steps

1. Run the Sales by transaction report for a decent period of time (a month, 3 months, a year).

2. Save the report output as is to excel on your desktop

3. Remove your name to anonymise it slightly

4. Open your AI (ChatGPT, Claude, Gemini, Mistral etc) and add the excel file with one of these prompts :

Prompt 1 - to show us what products are most frequently bought together

Analyse this transaction-level dataset grouped by receipt code. Identify the most common product combinations purchased together and rank them by frequency.

Prompt 2 - to show us which combinations lead to the highest basket value?

Identify product combinations that are associated with the highest average transaction value. Highlight which items contribute most to increasing basket size.

Prompt 3 - to look for missed cross-sell opportunities

Identify products that are often purchased individually but could logically be paired with other commonly co-purchased items. Suggest potential cross-sell opportunities.

Prompt4 - to see if staff behaviour impacts sales?

Analyse sales by user and identify differences in average basket value, product mix, and upsell behaviour between staff members.

Prompt5 - to look and see if device or location affect sales patterns?

Compare transaction patterns by device and identify any differences in basket size, product combinations, or transaction value.

Prompt6 - to see what are the strongest product categories driving revenue?

Analyse sales by product class and identify which categories contribute most to revenue, margin (if available), and basket frequency.

Because MYPOS captures exactly the structured dataset AI needs to generate meaningful insights - without complex integrations or data engineering, you have this now in your system ready to go.

Give it a try and let us know how you get on at support@myposconnect.com

‍