🎯 What you'll have at the endYou'll follow the same 3-stage path the creamery used: tidy your data, automate the routine, then add AI predictions.

🧰 Before you start

A CRM (customer relationship manager, the system where you keep customer records), e.g. Salesforce, HubSpot or Zoho
Past orders or sales data
Ideally an implementation partner for bigger systems

⏱️ Time neededMonths (in phases)

💵 Rough costEnterprise software: Salesforce/Agentforce pricing varies by number of users

🪜 Step by step

🧑‍💻 This one is rated Advanced: you may want a tech-savvy friend or colleague to help with the setup.

🐣 Beginner mode: showing the 6 essential steps only. Extra steps are hidden; tap 🚀 Advanced at the top to see all 6.

✅ 0 of 6 steps done 📄 Printable page

1
☐
Stage 1: clean your data
Put all customers, products and past orders in one system with consistent names. This is the boring but vital bit.
Mark as done
2
☐
Stage 2: automate one routine
Pick something like order entry. Make it quick to type an order in plain words (e.g. 'Firehouse 2kg x 10').
Mark as done
3
☐
Stage 3: add predictions
Turn on AI that suggests reorders from each customer's history. Staff review and confirm in seconds.
Mark as done
4
☐
Spot forgotten items
Have the AI flag products a customer usually buys but missed this time.
Mark as done
5
☐
Plan customer service AI
Later, let an AI answer detailed product questions from your records.
Mark as done
6
☐
Measure
Track order time, errors and missed reorders before and after.
Mark as done

💬 Copy-paste prompts

Tap Copy prompt, paste it into your AI (Grok, ChatGPT, Claude or Gemini), then replace the [yellow bits] with your own details.

💬 Plan my AI roadmap

My business: [what you sell, to whom, roughly how many customers and products].
Current systems: [spreadsheets / CRM name / accounting software].
Biggest daily time-wasters: [list].
Using a 3-stage approach (clean data, automate routines, add AI predictions), give me a 6-month plan with monthly milestones, what data I need for each stage, and the cheapest tools that would work.

💬 Predict reorders from a spreadsheet

Here is our order history [paste or attach: customer, product, date, quantity]. For each customer, predict what they're likely to order next and when, and list products they usually buy but haven't ordered recently. Show results as a table.