Picture your weekly Monday management meeting. The numbers exist, buried in a spreadsheet with 11 tabs that only one person knows how to navigate. So someone screenshots a few cells into WhatsApp, someone else reads totals off a printout, and the first twenty minutes go to establishing what the numbers are instead of why they moved.

What that meeting needs is a dashboard: one screen with the headline figures on top, a revenue trend, the top products, a breakdown by location — and filters, so when somebody asks “what does that look like for May, just for the Montego Bay store?” the answer is a click, not a follow-up email.

Until recently, getting one meant weeks of work in specialist software, plus an analyst to build it and licence fees to keep it alive. That is why most small businesses never got one. Artificial intelligence has now collapsed that whole project into a single instruction.

One prompt does the work

Here is the entire method. Export your sales data to one clean sheet — date, product, location, amount. Give it to your AI assistant with a prompt like this:

“You are a world-leading HTML dashboard designer. Build me an interactive HTML dashboard from this sales data. I want summary cards for total revenue, growth versus last month, and average sale value; a monthly revenue trend; my top ten products; and a breakdown by location. Let me filter everything by month and by location. Deliver the finished dashboard as a downloadable HTML file.”

One prompt does the work: ask for summary cards, a monthly
revenue trend, top products, and filters — all in plain English. (Branded graphic by PGH Consulting, LLC)

A few minutes later you are looking at a working dashboard — charts drawn, totals calculated, filters live — delivered as a single HTML file you can download, open in any browser, and share with your team. Not a mock-up or a description of a dashboard: the actual thing, ready to click.

Claude, Copilot, or ChatGPT — use whichever you already have

The remarkable part is that this is not one product’s party trick. The three assistants most Jamaican professionals already use can all do it.

Claude (by Anthropic) turns your data into a polished interactive page — charts, summary cards, and working filters — that opens right in the conversation and can be shared with your team. If presentation matters, this is the strongest of the three.

Use what you already have: Claude, Copilot, or ChatGPT can each
build a working dashboard from one prompt — the tool matters less than
the instruction. (Branded graphic by PGH Consulting, LLC)

Microsoft Copilot lives inside Excel and the wider Microsoft 365 suite. Point it at your table and ask for the analysis and charts; if your company runs on Microsoft and the data already sits in Excel, it is the shortest path.

ChatGPT (by OpenAI) accepts an uploaded spreadsheet in the chat and builds the charts and interactive views from it — turn on Canvas to preview the dashboard live before you download it. If ChatGPT is the assistant you already pay for, you do not need anything else.

The tool matters far less than the prompt. Say what you want on screen, what you want to filter by, and who it is for. That sentence is the skill — and it transfers across all three.

What this looks like in practice

The first question changes: with the dashboard on screen, the
meeting moves from what the numbers are to why they moved. (Branded graphic by PGH Consulting, LLC)

Take a retailer with three locations — Kingston, Mandeville, Montego Bay. On Friday afternoon the owner exports the month’s sales from the point-of-sale system, hands the file to her preferred assistant with the prompt above, and spends ten minutes checking the totals against the source sheet. On Monday the meeting opens with the dashboard on the screen. Nobody asks what the numbers are. The first question is “why is Mandeville down eight per cent?” — which is the conversation the meeting was always supposed to be.

“The first twenty minutes of the meeting used to go to what the numbers are. A dashboard spends them on why.”

A note on security

Before any real sales data goes into any AI tool, ask the unglamorous question: where does this file go? With all three assistants, your upload travels to the provider’s servers. Each of them offers business plans — Claude for Work, Microsoft 365 Copilot, ChatGPT Business and Enterprise — under which your data is not used to train their models; on personal plans, you should check the privacy settings and switch training off before uploading anything sensitive. And regardless of plan: a sales dashboard rarely needs customer names, account numbers, or staff details, so strip them out before the file leaves your machine. Aggregated sales by product and location tells the story without exposing anyone.

What it will not do for you

The dashboard is only as good as the spreadsheet underneath it. If your export has duplicated rows or a misspelt location name, the AI will faithfully chart the mess. Check a handful of numbers — total revenue, one product, one location — against the source before anyone else sees it. And remember what a dashboard is: it shows you the business; it does not run it. The judgement about what to do when Mandeville is down eight per cent is still, happily, yours.

Four steps before your next meeting: build the dashboard fast,
but verify the figures before anyone acts on them. (Branded graphic by PGH Consulting, LLC)

What to try this week

1. Export your last 12 months of sales to one clean sheet with four columns: date, product, location, amount — and remove customer names first.
2. Give it to Claude, Copilot, or ChatGPT with one prompt: summary cards, monthly trend, top ten products, breakdown by location, filterable by month and location.
3. Before sharing, check three numbers against the source spreadsheet — total revenue, one product, one location.
4. Open your next management meeting with the dashboard on screen, and note which number people ask to click into first. That is your next report.

Always verify the figures against your source data before acting on any dashboard.

Peta-Gaye Hardy is the founder of PGH Consulting, LLC, where she helps finance and operations teams adopt AI in practical, low-risk ways. She writes the weekly AI in Finance & Business column and is based between Jamaica and the United States. Learn more at www.pghconsultinggroup.com. Follow on Instagram @pghconsultinggroup.

Disclosures: This article is informational and does not constitute investment, tax, legal, or accounting advice. AI tools can produce errors and figures should always be verified against source data. Some features described require paid subscriptions. The author has no commercial relationship with Anthropic, Microsoft, OpenAI, or any product mentioned and was not compensated by them. Readers should consult a qualified professional before acting.