Verified against ChatGPT · 2026-08-11
Research competitors around what they can actually do, not what their marketing copy claims
Builds a competitor research brief that separates verifiable product capability from marketing language, and flags claims that couldn't be confirmed rather than repeating them as fact.
The prompt
Ready to copy — highlighted parts are example details you can swap.
I need competitor research that's useful for a real product decision, which means it can't just repeat what each competitor's website says about itself. Marketing copy describes what a company wants you to believe, not necessarily what the product does. MY PRODUCT AND THE DECISION THIS INFORMS We're a mid-market project management tool deciding whether to build native time tracking or keep recommending an integration partner. COMPETITORS TO RESEARCH Asana, Monday.com, and ClickUp. SPECIFIC CAPABILITIES I NEED VERIFIED Native time tracking, whether it supports billable-vs-non-billable hour tagging, and whether it exports to QuickBooks without a third-party connector. WHERE TO LOOK BEYOND THEIR OWN SITE G2 and Capterra reviews, their public changelog pages, and their help-center documentation rather than their marketing pages. For each competitor, separate two things clearly: what they claim on their own marketing pages, and what can be independently verified through G2 and Capterra reviews, their public changelog pages, and their help-center documentation rather than their marketing pages. — review sites, changelog/release notes, documentation, user forums, or third-party comparisons. Where a marketing claim can't be verified independently, say so explicitly rather than repeating it as if it were a confirmed fact; a lot of competitor research fails exactly here by transcribing homepage copy as if it were neutral information. For each capability in Native time tracking, whether it supports billable-vs-non-billable hour tagging, and whether it exports to QuickBooks without a third-party connector., give a direct verdict: confirmed working (with source), claimed but unverified, or confirmed absent/limited. Flag anything that looks like a common SaaS overstatement pattern — "AI-powered" with no specifics, "enterprise-grade" with no named certification, "99.9% uptime" with no public status page to check it against. WHAT NOT TO DO Do not present the research as a feature-comparison table with checkmarks pulled straight from each competitor's own pricing page — that's just relaying their sales pitch, not researching them. OUTPUT FORMAT Per competitor: 1) one-line summary of their actual positioning, 2) a table of the requested capabilities with a confirmed/claimed/absent verdict and source for each, 3) the single most overstated marketing claim you found on their site, named specifically.
Customize
Optional — swap in your own details for the highlighted parts above.
Why this works
A model asked to research a competitor will, absent other instruction, gravitate toward that competitor's own homepage and pricing page as the primary source, simply because that content is written to be the clearest, most quotable summary of the product and is usually the most heavily indexed page about it — which means the research ends up repeating the subject's own sales pitch back as if it were neutral fact. Explicitly separating "what they claim" from "what's independently verifiable" forces a second retrieval pass against a different kind of source, and naming specific independent source types (review sites, changelogs, documentation) gives the model concrete places to check a claim against rather than leaving "verify" as an abstract instruction it has no obvious way to act on. Requiring a three-way verdict per capability — confirmed, claimed-but-unverified, or absent — matters because it removes the middle-ground option of just describing the feature neutrally without committing to whether it's real; a plain description reads as tacit endorsement even when the model never actually confirmed the claim. Asking specifically for the single most overstated marketing claim trains attention on the genre of vague SaaS superlative language ("AI-powered," "enterprise-grade") that's specifically designed to sound substantive while being unfalsifiable, which is exactly the kind of claim that would otherwise slip through a feature-comparison table unchallenged because it isn't a concrete enough statement to obviously flag as false.
What you get back
Competitor: Monday.com. Positioning: broad work-OS platform targeting cross-functional teams. Native time tracking: confirmed present (documented in their help center), but billable/non-billable tagging requires a paid add-on per their own docs — claimed as "built-in" on the marketing page, actually gated behind a higher tier. Most overstated claim: "AI-powered resource management" with no specifics on what the AI actually does differently from rule-based automation.
Verified against
ChatGPT GPT-5.1 · 2026-08-11
Changelog
- 2026-08-11 — Initial publish, verified against ChatGPT GPT-5.1.
Need this built into your business?
If a prompt isn't enough — what Scult builds, built and maintained for you — that's Scult's day job.
EXPLORE WHAT SCULT BUILDS
