How to turn your Instagram analytics into an interactive dashboard with ChatGPT

By Thrive · 15 October 2026 · 6 min read

Your Instagram analytics can help you decide what to post next. The challenge is making sense of the figures and turning them into something useful.

After using ChatGPT to turn my LinkedIn analytics into an interactive dashboard, I wanted to share how you can apply the same approach to Instagram, with charts, audience information, content comparisons and a four-week content plan.

The aim is to put the insights inside the dashboard, so you can explore the numbers and see what they might mean for your content.

Here’s how you can try it with your own analytics, including a prompt you can copy and paste.

Step 1: Export your Instagram analytics

Use Meta Business Suite on a desktop browser and select the account connected to your Instagram profile. Make sure you have access to its insights.

Open Content, or the Content section within Insights, and look for Export data. Select Instagram, choose the correct account and reporting period, then generate and download the export. The file is typically CSV, which you can upload directly to ChatGPT. Menu labels and available reports can vary.

Try a 90-day period where available, or choose a shorter period with enough content to compare. Keep a note of whether your export uses lifetime content metrics or results within the selected dates.

If audience or follower information is available in a separate report, include that too. A content export may not contain everything. Keep the original files and let ChatGPT identify what each one includes.

Step 2: Upload the file and request the complete dashboard

Start a new ChatGPT conversation and attach your Instagram analytics CSV or Excel files. If you have several reports, upload them together and explain the reporting period.

The brief matters here. Asking for “some insights” leaves a lot open to interpretation. Describe the dashboard you want, what it should include and how you want to use it.

The following prompt combines the analysis, interactive dashboard and content plan into one request.

Copy and paste this prompt

Use my attached Instagram analytics exports to build a working interactive dashboard containing my performance data, insights and a four-week content experiment.

Read all files and relevant sheets before building it. Identify what each report covers, remove duplicate records where appropriate, and use only the actual figures provided. Include the following sections:

1. Performance overview

Show the reporting period and available headline figures for views, reach, content interactions, profile visits, total followers, follows and unfollows. Keep these metrics distinct and preserve the export’s definitions. Compare equivalent recent periods only where the data supports this. Calculate net follower change only if the required data is available.

2. Activity over time

Create interactive charts for the time-series metrics available in the files. Include date filters, daily and weekly views where supported, and hover details. Add short explanations of notable peaks and changes. Do not invent daily values from period totals or sum daily reach and label it unique reach for the whole period.

3. Content performance

Create a searchable, sortable table containing available publication dates, captions or titles, links, content formats, views, reach, likes, comments, saves, shares and other reported interactions. Add filters for Reels, carousels, images and Stories only where those formats are identifiable in the export.

Help me compare content with high reach against content with more interactions relative to its reach. Where the data allows, calculate interactions divided by reach, multiplied by 100. Clearly state which interactions are included, avoid double-counting totals and their components, and do not describe the result as the percentage of unique people who engaged. Keep rates based on views separate from rates based on reach.

4. Audience information

Visualise the audience categories actually supplied, such as locations or age ranges. Distinguish followers from reached or engaged audiences, and label each population clearly. If audience data is missing, explain this rather than inventing a breakdown. Do not assume these people are customers or buyers.

5. Insights and interpretation

Put the insights inside the dashboard alongside the relevant data. Explain the findings in plain English and suggest what they might mean for future content. Clearly separate observations from possible explanations. Only draw conclusions about topics or formats when the supplied data identifies them.

6. Four-week content experiment

Include a weekly plan with suggested topics and formats, example angles, the finding behind each suggestion and what I should measure. Present these as experiments to test. If my business goals or audience are unclear, flag that assumption and make the plan easy to adapt.

Design and functionality

Make the dashboard polished, mobile-friendly and easy to navigate. Use readable charts and clearly labelled sections. Ensure the filters, search and sorting work.

Provide a working interactive preview where supported, or a self-contained HTML file I can download and open in my browser.

Accuracy

Check the calculations against the source files. Keep missing values distinct from zero. Account for different content ages, incomplete coverage and partial reporting days. Distinguish lifetime content metrics from metrics within a selected reporting period. Separate paid and organic results where the files allow; otherwise flag that limitation.

Do not invent captions, follower attribution, sales or reasons for performance. Do not imply the dashboard is connected live to Instagram. Include a short explanation of the data’s limitations.

Deliver the dashboard with the insights and content plan built in.

Step 3: Check the figures and explore the findings

Before using the recommendations, compare a few headline figures and individual post results with your original spreadsheet.

Then explore the dashboard.

For example, you might find that a Reel reaches more people while a carousel generates more saves relative to its reach. That is a comparison to investigate in your own data, rather than a result to assume in advance.

These measures give you different perspectives on performance. Reach helps you understand how widely content travelled. Saves, shares and interactions relative to reach can help you decide what to test again, provided you compare the same definitions.

The figures cannot tell you exactly why someone responded, but they can give you better questions to investigate.

Step 4: Refine the dashboard through conversation

You can ask ChatGPT to explain a finding, change a chart or adjust the content plan.

For example:

▪️ “Show me the evidence behind this recommendation.”

▪️ “Which patterns depend heavily on one unusually successful post?”

▪️ “Make the four-week plan realistic for posting twice a week.”

▪️ “Add a section where I can record the results of each experiment.”

If something looks wrong, point it out and ask ChatGPT to check it against the source file.

Step 5: Repeat with your next export

This dashboard is based on the file you uploaded. It does not automatically receive new Instagram data.

When you have another reporting period to review, upload the new export in the same conversation and ask ChatGPT to update the dashboard and compare equivalent periods.

Keep a note of what you tested, too. Changes in topic, format or posting frequency provide useful context that may be missing from the export.

This is a practical way to use AI: taking your own business data and making it easier to understand, explore and act on.

The next step is to try the content experiments and see what the next set of numbers tells me.