47 prompts every B2B brand should monitor weekly

Most B2B brands monitor 5 to 8 prompts. Their competitors monitor zero. Both are wrong — the right number, after auditing 200+ pilots, is between 40 and 80, segmented by funnel stage. Below: the exact 47 we hand to every new customer on day one.

Replace [CATEGORY] with your product category, [BRAND] with your brand, [COMPETITOR] with your top rival. Don't worry about quality at first — quantity gives you the trend. Refine in week 3.

Top of funnel — discovery (10 prompts)

1. What's the best [CATEGORY] for B2B SaaS?
2. What [CATEGORY] should I use in 2026?
3. Recommend a [CATEGORY] tool.
4. Top 5 [CATEGORY] platforms.
5. What [CATEGORY] do startups use?
6. Best [CATEGORY] for a small team.
7. What [CATEGORY] do enterprise companies use?
8. Cheapest [CATEGORY] that doesn't suck.
9. Most popular [CATEGORY] right now.
10. What [CATEGORY] is everyone switching to?

Comparison — middle of funnel (12 prompts)

1. [BRAND] vs [COMPETITOR_1] — which is better?
2. [BRAND] vs [COMPETITOR_1] vs [COMPETITOR_2].
3. What's the difference between [BRAND] and [COMPETITOR_1]?
4. Is [BRAND] better than [COMPETITOR_1]?
5. Why use [BRAND] over [COMPETITOR_1]?
6. Pros and cons of [BRAND] vs [COMPETITOR_1].
7. Which is more affordable: [BRAND] or [COMPETITOR_1]?
8. Which has better support: [BRAND] or [COMPETITOR_1]?
9. [BRAND] alternatives.
10. [COMPETITOR_1] alternatives.
11. Tools like [BRAND].
12. What's a cheaper [BRAND]?

Alternative — defensive (8 prompts)

1. Alternative to [COMPETITOR_1].
2. Best [COMPETITOR_1] alternatives in 2026.
3. Open-source alternative to [COMPETITOR_1].
4. Free alternative to [COMPETITOR_1].
5. Why are people leaving [COMPETITOR_1]?
6. What's better than [COMPETITOR_1]?
7. [COMPETITOR_1] is too expensive — alternatives?
8. Switching from [COMPETITOR_1] — what should I try?

Use case — late funnel (10 prompts)

1. Best [CATEGORY] for [SPECIFIC USE CASE].
2. How do I [JOB-TO-BE-DONE] — what tool?
3. What does [TARGET PERSONA] use for [JOB]?
4. Best [CATEGORY] for a 50-person engineering team.
5. Best [CATEGORY] for a remote team.
6. Best [CATEGORY] for a regulated industry.
7. Best [CATEGORY] that integrates with [POPULAR TOOL].
8. Best [CATEGORY] with a free tier.
9. Best [CATEGORY] that doesn't require a sales call.
10. Best [CATEGORY] for non-technical users.

Authority and trust (7 prompts)

1. Is [BRAND] legit?
2. Is [BRAND] safe?
3. Is [BRAND] trustworthy?
4. Is [BRAND] still around?
5. Is [BRAND] funded?
6. Who owns [BRAND]?
7. Reviews of [BRAND].

How to use this list

Week 1. Pick 25 prompts from the list. Run them daily across ChatGPT, Claude, and Perplexity at minimum. Just record presence/absence and which brands are named.

Week 2. Refine. Drop prompts where every answer is the same regardless of brand (these are noise — too generic). Add specific prompts your sales team hears.

Week 3+. Trend. Plot citation rate per prompt over time. Look for prompts where you're absent — those are content opportunities. Look for prompts where competitors are gaining — those are PR opportunities.

What we found running these 47 across 200+ B2B brands

Four patterns repeat with enough consistency to be planning assumptions:

1. The "alternative" prompts are the highest-leverage and the most ignored. Average pilot citation rate on category prompts: 31%. On comparison prompts: 24%. On alternative prompts: 9%. That 9% is also where buyers with the highest purchase intent live — they've named your competitor and are actively shopping. If you only own three prompt types, own the alternative ones.
2. "Authority and trust" prompts are early warning systems, not acquisition channels. Citation rate barely moves with marketing effort. But when "Is [BRAND] still around?" suddenly returns a hedged answer, you have a PR problem brewing — typically a churned customer Reddit post or a Substack takedown. Two pilots caught reputation incidents 11 and 19 days before they would have hit their CEOs' inboxes.
3. Persona drift is bigger than engine drift. Same prompt run as "founder of a 12-person startup" vs "VP IT at a 5,000-person bank" returned completely different top-3 leaderboards in 73% of comparison prompts. If you only run prompts from one persona, you're seeing 33% of your reality.
4. The decay rate is faster than people expect. Inclusion rates drop 14–22% on average within 6 weeks of a major model update (e.g. GPT-5 → 5.1, Sonnet 4.5 → 4.6). Treat citation share like ad spend: it's not a banked asset, it depreciates.

The shortcut

If you'd rather not maintain this manually, BrandMirror runs 340 prompts daily across all four engines, with statistical sampling, drift alerts, and weekly reports. Pilots from $1,200/month — see the report layout first, or check what your category looks like on the public index.