As of 13 August 2026, AI can only partly build a sales dashboard for your small business.

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 the sales spreadsheet or export and make a copy containing the date, order ID, product or service, channel, region, sales amount, refund or cancellation fields and cost fields if available.
   2. Remove passwords and unnecessary personal data, then check that dates, order IDs, currencies and amount columns use consistent formats before uploading the copy.
   3. Write down the definitions you want for total sales, orders, average order value, returns and any margin measure, including whether refunds and VAT are included.
   4. Upload the copy and your definitions to Polymer, or another dashboard tool, and paste the prompt with the reporting period and available fields filled in.
   5. Read the tool's data-quality findings and correct duplicate rows, missing dates or inconsistent categories in the source copy before rebuilding the dashboard.
   6. Compare each summary card and monthly total with a pivot table or spreadsheet calculation from the source data, then trace the three sample records requested by the prompt.
   7. Apply every dashboard filter and check that the filtered totals match a separate calculation, then share or publish the dashboard only after recording the metric definitions and source date.

   Prompt

   Build a sales dashboard from the sales data I provide below. Use only the supplied data and do not invent values, categories or business rules. First inspect the columns and list any missing, duplicated, inconsistent or suspicious records. Ask questions before building anything if the data structure or metric definitions are unclear.
   
   Business context:
   - Business type: [business type]
   - Reporting period: [date range]
   - Currency: GBP
   - Sales channels: [channels]
   - Desired users: [users]
   
   Required measures:
   - Total sales: use [gross or net sales] and explain the choice.
   - Number of orders: count [order ID definition].
   - Average order value: define it as [definition].
   - Returns or cancellations: use [field and definition].
   - Gross margin: include only if the supplied data contains reliable cost figures.
   
   Create a dashboard specification with:
   1. A clear title and reporting period.
   2. Summary cards for the agreed measures.
   3. A sales trend over time.
   4. A breakdown by product or service, channel and region where those fields exist.
   5. Filters for date, channel and the other useful fields available.
   6. A table showing the underlying records or totals behind each chart.
   
   For every measure, state the exact source column, formula, filters and treatment of blanks, refunds and duplicates. Flag anything that cannot be calculated reliably. Give me the dashboard layout and, if the tool supports it, the exact steps or code needed to build it. Finish with a verification checklist that compares the dashboard totals with the source data and includes three sample records to trace manually.
   
   Sales data:
   [paste the data or describe the connected source here]

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