Birdview PSA — 0.9% to 21.5% ChatGPT visibility in 8 weeks

• Started Feb 2026 at 0.9 percent ChatGPT mention rate in the PMS cluster.
• Mapped 100 LLM queries and weighted Decision-stage prompts 2x in scoring.
• Shipped 8 articles in 8 weeks — one on Birdview's site, one on Medium per topic.
• Reached 21.5 percent ChatGPT mention rate, 27.8 percent Gemini, and 103 unique queries with brand presence.
• 95 of 100 LLM queries overlap with Birdview's SEO keywords — one asset, two channels.

B2B software buyers increasingly start their research in ChatGPT, Gemini, or Perplexity rather than Google. They ask things like "how to choose PMS for a distributed team" — and get a direct answer with recommendations. If your brand is not in that answer, you are out of the buying process before the customer ever visits your website.

Birdview had strong AI authority in a narrow PSA cluster — but in the broader, more competitive PMS cluster nearly nothing. That is where the larger audience lives: consulting firms, architecture practices, professional services teams.

1. Mapped the query landscape. We collected 100 LLM queries in the PMS space and tagged each by funnel stage (awareness → evaluation → decision). We tracked which sites AI systems were already citing for each query.
2. Focused on Decision-stage queries only. We used a "Decision × 2 + Evaluation × 1" scoring model. The top 5 topics were all Decision-stage prompts ("how to choose PMS for a company", "how to pick PMS for distributed teams", "what indicates low implementation risk", etc.). No dilution across the funnel.
3. Published by protocol. Each topic shipped as two pieces — one on the client's site, one on Medium. Every article passed an LLM-citability check: answer in first 150 words, H2s as user questions, self-contained paragraphs, tables, and FAQs. Score under 80 = rewrite before publish.
4. Measured weekly from day one. We ran 16 control scans across ChatGPT, Gemini, Perplexity, and Google AIO over eight weeks. The data showed Gemini responded fastest to scenario-based content — so we doubled down on that format mid-pilot.

Three things actually drove the lift, and they all rewrite how Birdview's marketing team thinks about content distribution:

1. Platform choice over volume. Perplexity indexes Medium aggressively and cites it as a source. Publishing one piece on Birdview's site alone left significant AI reach unclaimed.
2. Decision-stage focus over funnel coverage. Awareness content matters, but at Decision stage an AI answer is directly adjacent to a purchase conversation. Eight weeks of focused effort produced measurable pipeline impact.
3. Weekly measurement over quarterly reviews. Without weekly data we could not see Gemini responding to scenario-based content. The ability to shift format mid-pilot is what made the result possible.

• ChatGPT mention rate: 0.9% → 21.5% in the PMS cluster (a 23-times lift).
• Unique LLM queries where Birdview appears: 8 → 103.
• Gemini mention rate: 27.8%.
• SEO bonus: 95 of the 100 target LLM queries overlap with Birdview's organic SEO keywords. The same content drives both channels.

The numbers above are the eight-week pilot. They were measured on the PMS cluster in early 2026. Since then the account moved to continuous measurement, on a much wider base.

Here is the latest full scan — 13 August 2026, against the previous one on 16 July:

Read the last row carefully. It is the one that misleads. Share fell a point. Citations did not — they went from 166 to 413.

The base grew. Five engines instead of two, and nearly three times the AI responses. A share is a fraction, and this denominator tripled. The system calls that a denominator effect, not lost visibility, and that reading is correct.

It is also why the two sets of numbers here do not contradict each other. 21.5% was a ChatGPT mention rate, inside one cluster of about a hundred prompts. 14% is the share of answers citing the site across 589 prompts and five engines. Different question. Different denominator.

In the same window 183 pages gained citations and one lost them. The strongest movers: calculate-project-profitability-real-time-psa (+17), psa-implementation-timeline (+15), from-spreadsheets-to-psa (+14).

What is still open

The scan names the gaps too. We publish them, because a case that lists only wins is not evidence of measurement:

• 360 of 589 prompts still never cite the site. For 309 a candidate page already exists and needs upgrading — cheaper than new content. 51 have no page at all.
• The site is cited 413 times. The brand name appears in 82 answers (Asana: 384). Engines learn names from third-party pages. Own-site content cannot close this one.
• Ranked shortlists name Monday 16 times against our 1. Those slots are won by comparison pages. That work is queued, not done.

Three things actually drove the lift:

1. Platform choice. Perplexity indexes Medium aggressively. Publishing one piece on the client's domain alone would have left significant AI reach unclaimed.
2. Decision-stage focus. Awareness content matters, but at Decision stage an AI answer is directly adjacent to a purchase conversation. Eight weeks of focused effort produced measurable pipeline impact.
3. Measurement from day one. Without weekly data we could not see Gemini responding to scenario-based content. The ability to shift format mid-pilot is what made the result possible.

Full case study: https://www.linkedin.com/pulse/we-grew-b2b-saas-brands-chatgpt-visibility-23x-8-weeks-shevchenko-kh6rf