Customer Story: How a B2B SaaS Series B Company Owns Its Category in Perplexity

Zero citations on Perplexity. Zero on ChatGPT. Zero on Gemini. For a Series B b2b saas company with 180 employees and a category they invented, that stung. Their organic traffic from Google was healthy, until it wasn't. The problem wasn't their SEO. The problem was that 22% of their target buyers now start research inside an AI answer engine, and the company simply didn't exist there. Competitors were getting cited for exactly the prompts that defined the space. The fix took four months, a remap of their content strategy, and a platform built for answer engine optimisation. The result: the company now appears in 84% of category-defining Perplexity prompts, often as the first or only citation, and AI-driven signups grew 27% quarter-over-quarter.

This is the step-by-step playbook the company followed (and you can replicate) to own a category in Perplexity using SearchFit's AEO workflow, from baseline measurement to revenue correlation. It's not theory. It's what worked.

The b2b Company at a Glance

• Stage: Series B, $38M total raised
• Headcount: ~180
• Product: Project management platform for remote engineering teams
• Category: "Engineering Collaboration OS" (the term they were trying to own)
• AI Visibility Problem: 0% citation rate across ChatGPT, Perplexity, Gemini, and Claude for 25 category-defining prompts
• Goal: Become the primary cited source when anyone asks an AI engine to compare, define, or recommend solutions in their space

The company’s founders knew category creation is the most durable moat in B2B SaaS. They had done the hard work of defining a new space, but they were absent from the channel where early adopters increasingly discover tools: AI-powered answer engines. That absence was measurable, and fixable.

Prerequisites

Before you start, you need three things:

1. Access to an AEO platform that tracks per-prompt, per-engine citations. SearchFit was purpose-built for this, it monitors visibility across ChatGPT, Perplexity, Gemini, Claude, and Grok, and correlates citations to traffic and signups. You can certainly piece together data from manual queries, but at any scale beyond five prompts, that's a time-wasting dead end.
2. A defined list of your category-defining prompts. These aren't your top Google keywords; they're the natural-language questions buyers ask when exploring the space: "What is the best engineering collaboration tool?" or "Compare Linear vs. Jira for remote teams."
3. Executive buy-in to invest in answer engine optimization as a new channel. If your leadership still equates SEO with Google rankings alone, start by sharing SearchFit's AEO KPI framework for executives to frame the opportunity in revenue terms.

Once these are in place, follow the seven steps below.

Step 1: Establish Your AI Visibility Baseline, No Guessing Allowed

The company's CMO assumed they'd be fine. They had strong domain authority, published thought leadership weekly, and ranked top 3 on Google for dozens of money terms. The assumption was wrong. Perplexity doesn't care about your DR. It cares whether your content is structured as a direct, authoritative answer to the specific query.

SearchFit's first scan across 25 category-defining prompts revealed:

• ChatGPT: cited in 0% of responses (even when source material was available via browsing)
• Perplexity: cited in 0%
• Gemini: cited in 0%
• Claude: cited in 0%

Baseline: 0/100 possible citations. Invisibility.

Meanwhile, three competitors, two direct rivals and an adjacent enterprise tool, were getting cited regularly. One smaller startup had zero Google rankings but appeared in 68% of Perplexity responses for the same prompts, purely through content architecture.

Pro Tip: Don't just count citations. Segment them by prompt type. SearchFit's Competitor Citation Gap Analysis workflow automatically buckets prompts into comparison, definition, recommendation, and how-to categories so you can see where the gaps widen. In this company's case, the biggest gap was on comparison prompts, the ones with highest purchase intent.

Takeaway: Your Google ranking is a lagging indicator for AI engine presence. Get a real baseline before you do anything else.

Step 2: Map Every Category-Defining Prompt, Then Prioritize by Commercial Intent

The company spent two days in a cross-functional session (SEO lead, content team, product marketing) and generated 78 natural-language prompts that a buyer in their space would type into Perplexity or ChatGPT. Examples:

• "What's the best tool for managing remote engineering teams?"
• "Jira alternative for software engineers that supports async work"
• "Engineering collaboration platform for Series A startups"
• "Linear vs. [Company Name] for remote team productivity"

They uploaded these 78 prompts into SearchFit and added 37 more automatically suggested by the platform based on competitor analysis. Total: 115 prompts. Then they applied three filters:

1. Monthly search volume equivalence (SearchFit estimates a proxy query volume from multiple sources)
2. Commercial intent strength (prompts with comparison or “best” language scored higher)
3. Current citation gap (competitors cited, company not)

The result: a tiered list of 40 high-priority prompts where AI visibility would directly influence pipeline.

Warning: Beware of chasing every mention of your category. The company wasted two weeks early on optimizing for a high-volume prompt, “history of project management tools”, that drives zero pipeline. Filter ruthlessly for buyer intent. Use SearchFit's revenue correlation playbook to connect prompt type to actual conversion events later.

Step 3: Reverse-Engineer the Content Structures That Already Win

With the priority prompts defined, the team used SearchFit to pull full citation data for each: which URLs were being cited, what snippet was pulled, and what content structures repeated. They didn't guess. They deconstructed.

Across the 40 high-priority prompts, three structural patterns dominated:

• Concise Definition + Table Comparison: A crisp 2-sentence definition followed by a 3-5 column comparison table. LLMs love this because it's easy to extract.
• "List of 5" With Differentiators: Headline like "5 Engineering Collaboration Platforms Compared" with h3 subheads and a unique differentiator for each.
• FAQ-Style Sections at the End: A run of h3 question headers with 40-60 word answers directly underneath.

SearchFit's reverse-engineering workflow allowed them to export a spreadsheet showing, per competitor, the exact page structure that earned the citation. They saw, for example, that one rival's comparison page had 2,100 words, an h2 titled "How to choose the right platform," and a simple HTML table. Their own comparison page was 4,800 words of narrative with no scannable table structure. The difference in AI citation rate was 78% vs. 0%.

Action: They rebuilt their top 12 pages with the winning structures, not by copying, but by applying the patterns to proprietary data and case studies only they had. This followed the exact principles in how to write content that Perplexity actually cites.

Step 4: Engineer Content Specifically for Perplexity (and Every Other Engine)

This is where most teams trip. They write one “optimized” page and hope it works everywhere. It doesn't. The company instead built a content engine using SearchFit’s AI-first content strategies framework. They produced:

• Pillar pages for the 15 highest-value prompts, structured for citation extraction (clear definitions, tables, h3 FAQ sections)
• Supporting pages answering long-tail variants (each targeting a specific prompt cluster)
• Entity-rich author biographies and About page updates to signal authorship (more on this in Step 5)

Key tactical decisions:

A. Short paragraphs, no fluff. LLMs prioritize content that gets to the point in 200-300 word chunks under descriptive subheadings. Long, meandering introductions did not get cited.

B. Use of comparison tables with structured data. For any prompt containing “vs” or “alternatives,” they embedded an HTML table with clear column headers. Perplexity frequently pulls directly from these, if the HTML is clean.

C. Prompt-specific metadata. SearchFit's content helper suggested schema markup refinements and entity linking that increased topic relevance scores by 34% in early tests.

D. Optimize for query-prompt alignment. For “What is Engineering Collaboration OS?”, the winning structure was a 60-word definition, a 1-sentence value prop, and a 3-item bullet list, then the rest of the page. Overly comprehensive Wikipedia-style entries got ignored. SearchFit’s 5 content strategies for AI search visibility outlines this precise approach.

Pro Tip: Run your draft content through SearchFit's pre-publication scanner. It scores your page against the top-performing cited pages for that prompt across all engines, showing you exactly what's missing. The company's content team used this pre-flight check for every piece before publishing. Average first-pass score: 42/100. After revisions: 86/100.

Step 5: Build Entity SEO So the AI Engines Recognize Your Brand as a Source

This step is non-negotiable. Perplexity, ChatGPT, and Gemini don't just weigh backlinks. They weigh entity signals, how well-connected your brand is to the topic across the web. The company's entity SEO offensive focused on three moves:

1. Claim and Optimize Every Knowledge Base Entry. They ensured consistent, detailed entries on Crunchbase, LinkedIn, G2, Product Hunt, and Capterra. Each included the same company description with the category phrase “Engineering Collaboration OS.” This reinforced the category-term-brand link.

2. Unlinked Brand Mentions on Authoritative Domains. Using SearchFit's brand mentions vs. backlinks tracker, they identified 90+ unlinked mentions across tech media, engineering blogs, and venture capital write-ups. They systematically reached out for link attribution, not for PageRank, but for entity connectivity.

3. Author Entity Pages That Signal E-E-A-T. Every writer had a bio page with structured data, external bylines, and links to their LinkedIn and open-source contributions. This aligns with E-E-A-T reimagined for the AI search era: LLMs evaluate content partly by who wrote it and whether that person has demonstrated expertise elsewhere. The company's Director of Engineering had a 15-year open-source track record. They used it.

Within six weeks, SearchFit's entity graph showed a 3.2x increase in brand-to-category connection strength. The first citation appeared on Perplexity: for the prompt “What is Engineering Collaboration OS?” they were now the primary source, beating Wikipedia and three media sites.

Step 6: Track Citation Growth and Connect It to Revenue, Not Just Rankings

The CEO didn't care about citation counts. She cared about pipeline. So the team set up a measurement stack using SearchFit’s attribution framework for AI search traffic.

Here’s how they tracked impact:

A. Prompt-level citation weekly. SearchFit automatically re-queried all 115 prompts weekly across ChatGPT, Perplexity, Gemini, and Claude, logging which source was cited and in what position.

B. UTM parameters for all content cited in AI responses. Every URL the company controlled carried a ref=aeo UTM. When Perplexity displayed a citation, users clicking from the AI interface landed on a tagged page. That let them isolate AI-driven sessions in their product analytics.

C. AI-driven signup cohort analysis. Using the method outlined in SearchFit's cohort analysis for AEO-driven signups, they compared activation and retention of users who came via AI engines vs. Google organic. The finding: AI-referred users had a 22% higher activation rate on day 7 and were 15% more likely to invite a team member. Higher intent.

D. Revenue attribution. SearchFit's revenue correlation model tied citation growth to pipeline influenced. They assigned a conservative attribution window (7-day first touch) and found that, by month 3, AI search accounted for 9.4% of all new trial signups, with a clear upward trajectory. Within four months, AI-driven signups hit 14% of the total.

They didn't just climb a leaderboard. They moved the business.

Step 7: Scale Across All Answer Engines, Not Just Perplexity

While Perplexity was the initial battleground (it currently drives the highest click-through rate from citations among researchers and early adopters), the company quickly expanded the playbook to ChatGPT, Gemini, and Claude.

The same content structures that earned Perplexity citations translated well, with minor adjustments. For ChatGPT, which often incorporates source browsing results, they ensured their comparison tables were included in the root HTML and not loaded via JavaScript. For Gemini, they added more detailed entity markup. For Claude, they focused on extremely concise answers (Claude favors 30-60 word definitions before linking).

SearchFit's dashboard allowed them to track AEO success with the 12 metrics that matter, across all engines in one place. That unified view prevented the team from over-rotating on one engine at the expense of another.

Results across engines at month 4:

• Perplexity: 84% citation rate on priority prompts, 1st citation in 62% of cases
• ChatGPT: 71% citation rate
• Gemini: 63% citation rate
• Claude: 58% citation rate

They replicated the pattern, not just the tactics.

What the Company Would Have Done Differently

In a retrospective, the VP of Growth highlighted three lessons:

1. Start sooner. They wasted six months attributing the traffic decline to seasonal factors and Google updates before testing AI engines. Once they ran the baseline in SearchFit and saw 0% presence, the urgency became real. You can't fix what you don't measure.
2. Don’t ask content to rank on AI engines without changing the content structure. Repurposed blog posts didn't work. Only pages purpose-built for citation did. That meant breaking the traditional content calendar to prioritize AEO content.
3. Treat AEO as a standalone channel. It's not SEO 2.0. It requires its own KPIs, its own content operations, and its own attribution model. They hired a part-time AEO strategist who used SearchFit 10 hours a week to manage the entire program.

Pro Tip: If you're an agency managing AI visibility for multiple clients, the SearchFit API allows white-label dashboards and bulk prompt tracking. One agency built a white-label AEO dashboard to service 14 SaaS clients, reducing manual checking from 20 hours to 45 minutes per week.

The Bigger Picture: Category Ownership in the Age of LLMs

Owning a category in a search engine that doesn't display a traditional SERP requires a mindset shift. It's not about position zero. It's about being the source the engine trusts enough to cite, by name, with a link. That trust comes from a combination of content structure, entity authority, and consistent visibility across the entire prompt landscape.

The Series B company didn't just improve rankings; they reshaped how their category is explained by AI. A year ago, Perplexity would describe “Engineering Collaboration OS” by referencing generic articles. Now, it cites their definition 84% of the time, often as the sole source. They own the definition because they systematically engineered for it.

This playbook works for any B2B SaaS company willing to commit to answer engine optimization as a core growth channel, and to use SearchFit to measure, monitor, and optimize continuously. The B2B SaaS model itself rewards first movers who define a category and then defend it across every discovery surface. AI engines are now the most important surface.

Key Takeaways

1. Zero AI visibility is a real, measurable risk for SaaS companies that rely on category leadership. You cannot assume Google rankings transfer.
2. Start with a baseline across all engines using a tool that tracks per-prompt citations. Data dispels assumptions.
3. Not all prompts are equal. Prioritize those with clear commercial intent and where competitors are already cited.
4. Citation-winning content follows predictable structures: concise definitions, comparison tables, and FAQ sections. Reverse-engineer what’s working.
5. Entity SEO builds the connective tissue that LLMs need to recognize your brand as a source. Unlinked mentions, knowledge bases, and author authority matter.
6. Measure what counts: citation rate, UTM-tracked sessions, AI-driven signups, and revenue. Use a KPI framework for AEO.
7. Scale across engines with minor structural tweaks, what works for Perplexity works for others when adjusted.

In a market where B2B SaaS growth strategies are shifting toward AI-first marketing, the companies that get cited by AI engines today will own the buying committee of tomorrow. SearchFit gives you the playbook and the instrumentation. The rest is execution.