AI Prompt Research Template
AI Prompt Research Template
Turn real customer tasks and keywords into a weighted prompt panel for ChatGPT, Perplexity, Gemini and AI Overviews.
Which prompts belong in your AI visibility panel, how much each matters, and how many observations each tracking cycle needs.
- Excel workbook (.xlsx), opens in Google Sheets
- PDF field guide, 9 to 11 pages
- Blank and worked CSV templates
- DataForSEO import script
Free, no email required. Version 3.0, updated November 2026.
=IF(K13="","",IF(K13<>"Ready","n.a.",F13*G13))The formula above is the real one from cell M13. Scroll it sideways to read the whole rule.
Connected to the DataForSEO API
This template is designed around one DataForSEO endpoint. Pull the data, convert it with one command, and the workbook does the rest.
DataForSEO Labs: Keyword OverviewSearch volume, monthly history, difficulty and intent for a list of keywordsdataforseo_labs/google/keyword_overview/live
Save the JSON responseThe full API response, kept as your source of truth.response.json
Run the import scriptOffline. No API calls, no credentials.dataforseo_import.py import --template 25
Paste at A13Dashboard, checks and action list update instantly.AI Prompt Research Template.xlsx
Column map for this template:Filled by the APIAudit trail (automatic)You add
What the AI prompt research template decides
AI visibility tracking is only as good as the prompts you track. Most teams start with a dozen brand-flattering prompts written in a meeting, which tells you almost nothing about how real customers find, compare and choose. Prompt research deserves the same rigor as keyword research.
This AI prompt research template builds a repeatable prompt panel. Each prompt is tied to a persona, a task, a related keyword and its search volume, plus a business relevance weight. It calculates a weighted search context so you can prioritize prompts, and the observations needed per cycle from repeat runs and platform count.
Who it's for
- SEOs starting a GEO or AI visibility program
- Brand and product marketers who want to know how AI describes them
- Agencies setting up AI tracking for clients
- Teams choosing prompts for an AI visibility tool
Where it fits in your workflow
The AI prompt research template answers one question. These three templates answer the questions around it:
- AI Citation Tracker: how often AI answers cite your site, on which platforms and prompts, measured over an explicit set of valid, observed answers.
- AI Share of Voice Calculator: how your domain's AI citation rate and share compare with competitors' within the same answer corpus.
- SEO Content Brief Template: which sections of a page are ready to publish, and what each unfinished section needs next: writing, fact-checking or editor approval.
What's inside the download
One ZIP with the workbook, a 10-column input template (blank and filled), the PDF field guide and an offline script that converts saved DataForSEO responses into the input CSV. The workbook has five tabs.
How the AI prompt tracking calculation works
Observations per cycle = repeat runs × platform count. Weighted search context = related keyword volume × business relevance. Volume is context, not a measure of prompt volume, which no platform publishes.
The key formula, exactly as it appears in the workbook
=IF(K13="","",IF(K13<>"Ready","n.a.",F13*G13))
| Result column | Plain-English formula | What it tells you |
|---|---|---|
| Observations / cycle | Runs per cycle × Platforms | Number of scheduled platform-run observations for this prompt. |
| Weighted search context | Monthly searches × Business relevance factor | Traditional-search context weighted by analyst relevance. Not AI prompt volume. |
How to use the AI prompt research template
- Collect real tasksMine sales calls, support tickets, onsite search, People Also Ask and community threads for the jobs customers are trying to get done.
- Write natural promptsPhrase each prompt the way a person would type or say it, with context. One prompt per row.
- Weight by relevance and demandLink a related keyword and its volume, then set business relevance (0 to 1).
- Plan the cycleSet repeat runs (answers vary run to run) and platforms. The Dashboard totals observations per cycle.
The worked example in the Workspace tab. Clear the lime input cells, paste your CSV at A13 (values only) and every result recalculates.
Fill the AI prompt research template with DataForSEO
Every template in this library is built to run on DataForSEO data. This one uses DataForSEO Labs: Keyword Overview: search volume, monthly history, difficulty and intent for a list of keywords.
Most of the data columns fill automatically. Add the business inputs listed below, then paste.
- Request the dataSend this request with your DataForSEO API login (HTTP Basic auth).
POST https://api.dataforseo.com/v3/dataforseo_labs/google/keyword_overview/live [ { "keywords": [ "your keyword 1", "your keyword 2" ], "location_code": 2840, "language_code": "en" } ]
- Save the full JSON responseSave the complete response, not just the task ID, next to the template folder.
- Convert it with the included scriptIt runs offline, makes no API calls and never sees your credentials.
python3 Automation/dataforseo_import.py import --template 25 --input response.json --output my-inputs.csv
- Add what the API can't knowPrompt ID, Prompt text, Persona, Decision task, Business relevance factor, Runs per cycle, Platforms
- Paste at A13Clear the demo rows, paste the CSV values at Workspace cell A13 and the dashboard updates.
| Fills automatically from DataForSEO | You add |
|---|---|
| Keyword, Monthly searches | Prompt ID, Prompt text, Persona, Decision task, Business relevance factor, Runs per cycle, Platforms |
Source reference fills automatically with the task ID, endpoint, location, language, device and snapshot time, so every row is traceable. Before a big pull, estimate the cost with the DataForSEO pricing calculator. The same steps, plus an example request file, are in the download's Automation folder.
Worked example: Northbeam Outdoor
Every template ships filled with data from Northbeam Outdoor, a fictional outdoor-gear retailer that appears across all 40 workbooks, so you can see what useful output looks like before you add your own. This workbook has 12 example rows.
Twelve prompts across four platforms with five runs each means 240 observations per cycle. That's manageable manually once a month, and the minimum needed to see real movement.
Top 10 by weighted search context
| # | Prompt ID | Weighted search context |
|---|---|---|
| 1 | P-09 | 36,450 |
| 2 | P-01 | 22,200 |
| 3 | P-05 | 9,900 |
| 4 | P-11 | 5,670 |
| 5 | P-02 | 2,880 |
Input fields
Keep the columns in this order when you prepare a CSV. A missing observation stays blank; use 0 only for a measured zero or a verified No. Fields marked DataForSEO can be filled automatically by the included script.
| CSV field | Column | Meaning and rule | Source |
|---|---|---|---|
| prompt_id | Prompt ID | Stable panel identifier. | Your records |
| prompt | Prompt text | Exact natural-language request to test. | Your records |
| persona | Persona | Specific buyer or user context. | Your records |
| task | Decision task | Decision the user is trying to complete. | Your records |
| keyword | Keyword | Exact query in the selected country and language. | DataForSEO: keyword |
| search_volume | Monthly searches | Provider monthly search-volume estimate. Use one market, language and source date. | DataForSEO: search_volume |
| business_relevance | Business relevance factor | Analyst planning factor 0 to 1 for the task, not measured AI demand. Decimal: 0.08 = 8%. | Your records |
| repeat_runs | Runs per cycle | Repeated observations planned for each platform and locale. | Your records |
| platform_count | Platforms | Number of platforms in the panel. | Your records |
| source_ref | Source reference | Task ID, export filename and observation scope. Retain the untouched source response separately. Optional. | DataForSEO: source_ref |
Reading the result honestly
A balanced panel includes brand, comparison, problem, task and decision prompts. Keep the panel definition stable.
Limits. Traditional keyword volume is not a proxy measurement of prompt frequency. Panel selection affects reported visibility.
AI prompt tracking mistakes to avoid
Tips from practice
- Feed the panel straight into the AI citation tracker.
- Keep the panel stable for at least three cycles before changing it.
- Add a few competitor-comparison prompts; they're often where you're missing.
Frequently asked questions
How do I choose prompts to track in ChatGPT?
Start from real customer tasks and questions, write them as natural prompts, and weight them by business relevance and related search demand.
How many prompts should I track?
Small sites can start with 10 to 30. Larger brands often track hundreds. Consistency over time matters more than size.
Is there search volume for AI prompts?
Not publicly. Use related keyword volume as a proxy for interest, and label it as context.
How often should I run my prompt panel?
Monthly is common. Run each prompt several times per platform each cycle because answers vary.
Use it with these templates
Each template answers one decision. These are the ones that feed into, or follow from, the AI prompt research template.
26AI Citation TrackerAI Optimization: LLM Mentions
Log answers from ChatGPT, Perplexity, Gemini and AI Overviews, and measure your citation rate by platform.
27AI Share of Voice CalculatorAI Optimization: LLM Mentions
Compare citation rate and share across competitor domains inside one fixed set of AI answers.
19SEO Content Brief TemplateLabs: Keyword Overview
A content brief built for Google and AI answers: real questions, original evidence, sources and a three-step approval workflow.
28Answer Evidence Audit TemplateOn-Page API
Check every claim for a supporting source, original value, clarity and freshness, so AI answers can quote and cite it.
01Keyword Prioritization TemplateLabs: Keyword Overview
Rank every keyword by the profit it could add per hour of work, not by search volume alone.
30AI Referral Traffic TrackerAI Optimization: LLM Mentions
Measure sessions, conversion rate and contribution per session from ChatGPT, Perplexity, Gemini, Copilot and Claude.
Part of the AI Visibility & GEO Toolkit
Get all 6 AI Visibility & GEO templates in one ZIP, or the complete set of 40 SEO and GEO templates with the 401-page field handbook.
Further reading on Optimization Theory
Built by Jourdan Rombough, technical SEO and GEO consultant with 15+ years in search. About Jourdan or get an SEO plan.






