B2B SaaS · Case Study

3.4x Visibility in AI Engines — How Was It Achieved in 4 Months?

A Turkey-based B2B SaaS company increased its brand mention rate in ChatGPT and Perplexity responses from 19% to 64% in four months. In this analysis, we examine every step of the process carried out with SEOYEN modules.

Client Profile: Who Is This Company?

The company we worked with is a B2B SaaS provider founded in one of Turkey’s leading industrial cities and grown over 11 years into a mid-sized software company since its founding. Because of name and industry confidentiality, we will refer to it as “Company X.”

Company X’s product is industry-specific business process management software for SMBs in manufacturing, logistics, and retail. More specialized in a specific vertical than general-purpose ERP solutions, this software offers inventory management, order tracking, field service integration, and real-time reporting modules. Its customer base consists of more than 340 active organizations across Turkey, ranging from 200 to 1,500 employees.

The marketing team consists of three people: a content lead, a digital advertiser, and a marketing manager leading the team. They already had traditional SEO efforts in place and ranked on the first page of Google for some target keywords, but the entire setup was built around Google. There was still no system in place to monitor the new search behaviors introduced by the AI era.

The Core Problem: Being Invisible in AI Engines

In late 2025, Company X’s marketing manager noticed the sales team mentioning a new pattern: about one-third of prospects requesting demos were starting conversations by saying, “I checked ChatGPT, and it recommended this.” The vast majority of those prospects had been shown not Company X, but two competitor brands as the recommended software.

The manager ran a test personally and asked industry-specific queries to ChatGPT, Google Gemini, and Perplexity.

• “software recommendations for manufacturing SMBs” → Company X was not mentioned at all.
• “ERP alternative comparison for a small business” → Two foreign brands and one Turkish competitor appeared.
• “Which inventory management software is best?” → The top three players in the industry were listed, and Company X was absent.

A company averaging position 4.2 in Google search rankings was almost completely invisible in AI responses. The problem was clearly defined: Google SEO and AI Visibility were completely different disciplines, and the company had done nothing for the second one.

When they contacted SEOYEN, the goal was single and measurable: significantly increase AI answer visibility for 28 priority queries within 4 months.

The 4-Month Process: What Did We Do Month by Month?

Month 1 — Discovery and Baseline Measurement

The first month was built entirely around measurement and strategy. Through SEOYEN’s AI Visibility Module, 28 different target queries were sent to ChatGPT (GPT-4o), Google Gemini (1.5 Pro), Perplexity AI, and Anthropic Claude. The responses from each engine were reviewed to document whether Company X was mentioned, in what context competitors stood out, and which sources the AIs referenced.

During the same period, competitors’ backlink profiles were mapped using SEOYEN’s Backlink Analysis Module. It was found that Competitor A had published a guest-post series on two major industry blogs, and those blogs were among the sites ChatGPT frequently cited as sources. Competitor B, meanwhile, used FAQPage schema much more extensively; the Structured Data on its product comparison pages was being quoted directly by Gemini.

Baseline table (end of month 1): Measurements were taken across 28 queries. Company X was mentioned in the response of at least one AI engine in only 5 of those queries (18.9%). The average visibility rate of 4 competitor companies was 47%.

Month 2 — Rebuilding the Content Architecture

Using SEOYEN’s Keyword Research Module, software comparison phrases frequently used by AI engines were identified. These phrases were divided into three categories: (1) direct product comparisons (“X software or Y software?”), (2) use-case-focused queries (“how do you choose inventory software for a manufacturing SMB?”), and (3) industry-wide guides (“how many months does an ERP implementation take?”).

Within this framework, 6 new long-form content pieces were prepared. Content production used not SEOYEN’s AI features, but the in-house marketing team and a specialist freelance writer — because AI engines tend to classify human expertise as a more trustworthy source.

The following schema types were added to each piece of content:

• SoftwareApplication: Product name, category, operating system support, price range, and review score were marked up structurally.
• FAQPage: A “Frequently Asked Questions” block was added below each piece of content; the questions were selected from real queries asked directly to AI engines.
• HowTo: Step-by-step guide content such as “How is inventory management software evaluated?” was marked up with HowTo schema.
• Review (aggregate): Existing customer reviews were compiled and integrated into the product page with AggregateRating.

During the same period, guest-content pitches were sent to 3 industry publications that had been identified earlier. 2 accepted; the publications went live at the beginning of month 3.

Month 3 — Citation Growth and Monitoring

Month 3 focused on making the content “citable by AI.” Two critical steps were taken for this:

First, Quick Answer Optimization: AI engines often copy or paraphrase specific paragraphs in their responses. Those paragraphs need to be concrete, short, factual, and focused on a single topic. Company X’s existing content was restructured, and each section was given 2-3 sentence “AI-ready summary” paragraphs.

Second, E-E-A-T Signals: Signals related to “Experience, Expertise, Authoritativeness, Trustworthiness” (E-E-A-T), which influence Google’s and therefore AI systems’ trust evaluations, were strengthened. The authors’ LinkedIn profiles were linked on the content pages, and the company’s founding date and certifications were moved into visible sections.

Weekly automated scans were set up with SEOYEN’s AI Visibility Module. Each of the 28 queries began being checked across all engines twice a week, and the results were tracked in the SEOYEN dashboard. By the end of month 3, the visibility rate had risen from 18.9% to 41%.

Month 4 — Growth and Conversion Impact

In month 4, the focus shifted to reflecting the visibility gains in sales channels. A “Solution Recommended in AI Engines” badge was added to the demo page, along with a real-time updatable summary showing which engines it appeared in. The A/B test ran for 3 weeks; the version with the badge and social proof block increased the demo conversion rate by 22%.

During the same period, 2 additional guest-content pieces were published, and 6 new backlinks were earned (month 4 total: 8 authority backlinks). A comparison article published on one of the industry’s leading technology portals became content that ChatGPT cited as a source in 4 different queries within one week.

By the end of month 4, Company X was mentioned in the response of at least one AI engine in 64% of target queries (18/28), usually in a positive context such as “recommended for consideration.” Demo requests increased by 78% compared with the previous period.

Visibility by AI Engine: Before and After

The table below shows how often Company X was mentioned in the responses generated by all four engines across 28 target queries. For each engine, it answers the question of how many of the 28 queries produced a brand mention.

* Each of the 28 target queries was reviewed separately for every engine. The values are the month-4 averages of the weekly crawl results from SEOYEN’s AI Visibility Module.

Which Query Types Produced the Fastest Results?

The 28 queries progressed at different speeds. The breakdown by intervention type was as follows:

Quick Wins (6-8 Weeks)

The fastest progress was observed in query categories where FAQPage schema was added. In direct definition queries such as “what does X software do?” and “what does feature Y mean?”, Gemini began directly quoting the FAQPage schema blocks on the product page. In this category, visibility rose from 14% at the start to 58% within 8 weeks.

Example query types: “how does business process management software work”, “what is inventory tracking software used for”, “what are the advantages of industry-specific software for SMBs”.

Mid-Term Wins (10-14 Weeks)

Comparison and decision-support queries fell into this category. For these types of queries, ChatGPT and Perplexity place more weight on trustworthy third-party sources (blog posts, independent portals). About 3-4 weeks after the guest content was published, visibility in this category began rising materially.

Example query types: “ERP comparison for a small business”, “which manufacturing software is suitable”, “industry-specific ERP alternatives in Turkey”.

Longer-Cycle Queries (16+ Weeks)

Progress was slower for direct comparison queries against specific competitor brands. In these queries, AI engines appeared to require more established, multi-source references. By the end of month 4, 34% visibility had been achieved in this segment; the process is expected to push this rate even higher in months 5-6.

How Did SEOYEN Modules Fit Into the Operation?

Competitor Comparison: Where Did Things Stand After 4 Months?

At the start of the project, Company X ranked last in visibility among 4 competitors across the 28 tracked queries. While the competitors’ starting average was 47%, Company X began at 19% — meaning it lagged the main competitors by roughly 2.5x.

By the end of month 4, Company X had passed two of those competitors with a 64% brand mention rate; the gap with the leading competitor, Competitor A, narrowed (71% vs 64%). Its in-market visibility ranking rose from 4th to 2nd.

One notable finding: Competitor A’s visibility advantage came entirely from a long-term accumulation of backlinks, and that accumulation could not be overtaken in a short period. However, in terms of content quality and schema usage, Company X had surpassed all competitors by the end of month 4. This indicates potential to take the lead in the medium term (6-8 months).

“Our customers now ask ChatGPT about software and come to us after choosing one. When I first saw SEOYEN’s AI Visibility report, I realized 80% of that choice was going to our competitors. Four months later, it had turned in our favor. Not only visibility changed, but the quality of incoming demos changed too — now people come in already knowing the product and having compared it.”

4 Critical Lessons From This Case