Home · Business · IT, Data & Security · Data analysis & reporting

PARTLY

As of 13 August 2026, AI can only partly forecast sales for your UK small business.

This still needs a person who signs their name to it.

Can you do it?

15 minutesto a draft.

1 hourto something you’d act on.

Cost, all in£0

Skill neededchat-fluent

Who has to check ityou

What the alternative costsAkkio is a no-code AI analytics and prediction tool for business data; a human analyst is the alternative when your data needs business judgement.

If this goes wrong: you order too much stock, schedule too much labour or plan cash around a forecast that missed a change in demand.

What to actually do

  1. Hand it to a person

    The route this page recommends

    A person who owns the outcome does this end to end, worth it when the failure is dear.

  2. Use a tool built for this

    Second choice

  3. Do it yourself

    The distant third

    A chat interface, chat-fluent skill, and roughly 1 hour until you can act on the result.

    How to actually do it

    1. Open your sales system or spreadsheet and export dated sales records with the amount, units, product or service category and any refunds or cancellations.
    2. Add a notes sheet containing price changes, promotions, closures, new or discontinued products, known contracts and other events that affected demand, while removing unnecessary customer personal data.
    3. Check that dates use one consistent format, sales amounts use pounds sterling, categories have consistent names and duplicate or cancelled transactions are labelled before saving the file.
    4. Paste the prompt into ChatGPT, Claude or Gemini and attach the cleaned spreadsheet and notes, replacing the bracketed forecast period and category description with your own details.
    5. Answer the model's data questions, then ask it to produce the forecast, low and high scenarios, assumptions and backtest without filling gaps with estimates.
    6. Compare the backtest and every stated assumption with your original records and business notes, then change the forecast or its context where the comparison exposes an error.
    7. Use the checked forecast as a planning input alongside your cash position and operational constraints, and send it to your accountant or business adviser if a major borrowing, hiring or stock commitment depends on it.

    Prompt

    Forecast sales for my UK small business using only the data and facts I provide. Do not invent missing figures, dates, products, customers or causes. First identify missing, inconsistent or duplicated data and ask concise questions before forecasting. Then produce: 1) a forecast for [forecast period] by [product, service or other useful category], 2) a central estimate plus a plausible low and high scenario, 3) the assumptions behind each scenario, 4) any trend or seasonal pattern supported by the data, 5) a backtest showing how the method would have performed on earlier periods where possible, and 6) a plain-English explanation of the main uncertainties. Use pounds sterling and the same sales measure throughout. Separate confirmed facts from assumptions. Do not present the forecast as a guarantee and do not recommend stock, staffing, borrowing or spending decisions without stating what additional information would be needed. Data: [paste your dated sales records, including amounts, units and categories]. Business context: [opening hours, prices, promotions, closures, known contracts, planned changes and any other factors affecting demand].

    Open it prefilled in ChatGPT or Claude, or copy it into Gemini, which takes no prefill link.

What it gets wrong

  • AI cannot know about an upcoming competitor, customer loss, supplier problem or local event unless you provide that context.
  • It cannot turn incomplete or inconsistent sales records into reliable evidence by producing a more polished chart.
  • It cannot decide whether a past spike was a genuine change in demand or a one-off event without your business knowledge.
  • It cannot take responsibility for stock, staffing, borrowing or cash-flow decisions based on the forecast.
  • It cannot make a forecast trustworthy merely by giving it a narrow-looking range or a confident explanation.

What caps this at PARTLY: judgement under ambiguity, verification cost and stakes of error.

How we scored this

Five axes, each scored nought to two by hand: ten means AI carries the task cleanly, and the thresholds that turn a total into YES, PARTLY or NO are published in the methodology. Each axis name links to its definition.

AxisScore (0–2)
Output2
Inputs1
Verification1
Liability1
Effort delta2
Total7 / 10

The methodology and its thresholds are published in full.

FAQ

Can AI predict my small business sales?
Partly. AI can turn suitable historical sales data into a forecast and scenarios, but it cannot know future market changes or correct missing business context. Treat the result as planning evidence, not a guarantee.
Can AI forecast sales in Excel?
Yes, if the spreadsheet has consistent dates, sales measures and categories. AI can analyse the file or create formulas and charts, but you still need to check the calculations and the assumptions against your records.
What data do I need to forecast sales?
Start with dated sales records showing amounts, units and product or service categories. Add prices, refunds, promotions, closures, new or discontinued offerings, known contracts and any other event that changed demand.
Can I trust an AI sales forecast?
Only to the extent that its data, assumptions and past-period tests stand up to checking. You carry the consequences of decisions based on it, so a forecast that affects borrowing, hiring or major stock commitments deserves review by an accountant or business adviser.

Nearby answers

All Data analysis & reporting rulings

Assessed by gpt-5.6-luna (gpt-5.6-luna) on 2026-08-13, second-checked by an independent model. Wrong somewhere? Email [email protected] and it gets re-checked.

The newsletter

AI news, new answers and product picks, straight to your inbox.