Companion resource for the September 22, 2026 LinkedIn draft. Practical templates and an operating workflow.
<aside> 🚀 🚀 FREE STRATEGY CALL Want to grow your product through viral reach? We help teams across iGaming, mobile apps, AI products, and SaaS build creator-led organic distribution generating tens of millions of views at CPMs starting from $0.03.
<aside> 🚀 Book a free strategy call
</aside>
On the call you'll receive:
Build a repeatable audit of whether AI answers recommend your product, then improve the pages and evidence behind those answers. Despite the title, this is an organic recommendation workflow, not a guide to buying ad placement. A citation is not an endorsement, a visit is not a purchase, and no optimization guarantees inclusion.
GOAL: Build a small, commercially relevant prompt set rather than chasing generic mentions.
ACTION: Review recent sales questions, support tickets, search queries and comparison-page visits. Select 12 questions across discovery, comparison, constraints and switching. Preserve the customer's language and record the country and audience. Do not upload personal customer details into public AI tools.
TOOL/RESOURCE: Your support inbox, search analytics and a spreadsheet.
EXPECTED OUTPUT: A 12-row prompt register, with four high-intent questions marked as priorities.
COMMON MISTAKE: Testing only your own brand name; that measures recognition, not discovery.
| Intent | Example for a language app | Page that should answer it |
|---|---|---|
| Discovery | Which apps help adults practise spoken Spanish? | Use-case page |
| Comparison | Which is better for live speaking practice: an app or a tutor? | Balanced comparison |
| Constraint | What can I use offline for 15 minutes a day? | Feature and limitations page |
| Switching | What should I check before replacing my current language subscription? | Switching checklist |
GOAL: Separate a real visibility gap from a one-off response.
ACTION: Run the same questions in the AI products your buyers use. Record product, model or mode when shown, date, country, browsing availability and whether the session was fresh. Repeat priority questions three times in fresh sessions; this is an operating sample, not a statistically representative market estimate. Save the full answer and visible source links. Never ask a leading follow-up that tells the system to recommend you.
TOOL/RESOURCE: A browser and the audit sheet below.
EXPECTED OUTPUT: A baseline with recommendations, citations and factual errors separated.
COMMON MISTAKE: Counting a brand in a warning or negative comparison as a positive recommendation.
| Date | Prompt ID | Product/mode | Run | Recommended brands | Our brand recommended? | Our URL cited? | Incorrect claim | Evidence URL |
|---|---|---|---|---|---|---|---|---|
| Illustrative | P01 | Record actual mode | 1 | Record exact answer | Yes/No | Yes/No | Quote or none | Saved answer/source |