Research · Case Studies
What GEO Implementation Actually Produces — Client Results Across Five Disciplines
These case studies document Generative Engine Optimisation engagements across AI Visibility, Local GEO, Content Intelligence, Agent Readiness, and Hallucination Detection. Every result is measured by running the agreed buyer prompts in ChatGPT and Perplexity before and after the engagement. No dashboard scores — verifiable results only.
At a Glance
Common Threads Across These Engagements
Across the ten engagements, the same handful of failure patterns keep recurring. Three are pure technical inaccessibility: a robots.txt file silently blocking every AI crawler (Cloud Consulting, Bengaluru), or job listings and pricing rendered client-side in JavaScript that AI agents simply can't read (Fintech, Mumbai; IT Services, Chennai). Two involve AI actively misdescribing the company rather than ignoring it — an outdated product category or factual errors about founding year, client count, or certifications (HR Tech, Pune; EdTech, Hyderabad). Three are Local GEO engagements where a single well-reviewed competitor was being cited for every neighbourhood query regardless of how established the client actually was (Cosmetic Dentist, Hyderabad; Interior Design Studio, Bengaluru; CA Firm, Bengaluru). And two are citation-gap engagements where competitors were simply out-publishing the client with content structured for AI extraction (Cybersecurity, Bengaluru; HealthTech SaaS, Pune). None of the ten started from a position of moderate visibility — every engagement began at zero citations, or with an actively incorrect AI description, for the buyer queries that mattered most to that business.
How These Results Are Verified
Every engagement starts from the same baseline: 10 buyer queries — Money Prompts — agreed with the client before any work begins. Each prompt is run in ChatGPT and Perplexity and the result is recorded: cited or not cited, and which sources were cited instead. That first run is Day 1. The same prompts are re-run weekly, with full cross-platform audits at Day 30, 60, and 90, so the case studies below report a measured change over a fixed window, not a snapshot chosen after the fact.
This is also why the case studies read as “0 of 9 to 6 of 9” rather than a percentage or a score out of 100: the underlying number is a literal count of buyer prompts, or neighbourhood searches, where the brand went from absent to cited. Anyone can reproduce the check by opening ChatGPT or Perplexity and typing the same kind of query for their own category.
Industries Represented
| Industry | City | Discipline |
|---|---|---|
| AWS Cloud Consulting | Bengaluru | AI Visibility |
| HR Tech / Talent Matching SaaS | Pune | Hallucination Detection |
| Cosmetic & Dental Implants | Hyderabad | Local GEO |
| Interior Design | Bengaluru | Local GEO |
| Cybersecurity (VAPT / SOC / Cloud Security) | Bengaluru | Content Intelligence |
| HealthTech / Hospital Management SaaS | Pune | Content Intelligence |
| Fintech / GST Reconciliation SaaS | Mumbai | Agent Readiness |
| IT Services | Chennai | Agent Readiness |
| Corporate Training / EdTech | Hyderabad | Hallucination Detection |
| Chartered Accountancy | Bengaluru | Local GEO |
By Service
AI Visibility
Whether ChatGPT, Perplexity, Gemini, and Claude cite the brand at all for category and comparison queries.
Local GEO
Whether AI recommends the business by name when a nearby buyer asks for a local vendor.
Content Intelligence
Closing specific citation gaps with content structured for AI extraction rather than search ranking.
Agent Readiness
Whether AI agents can actually navigate and use the site — forms, buttons, and structure an agent can act on.
Hallucination Detection
Finding and correcting inaccurate descriptions AI platforms are already generating about the brand.
By City
Case studies grouped by the city where the engagement ran — useful for a visitor scanning for local proof.
Bengaluru (4)
Pune (2)
Hyderabad (2)
Mumbai (1)
Chennai (1)
All Case Studies
Full summaries — this is the canonical list. Each entry uses the real challenge/fix/result copy already published on that case study's own page.
1. How a Cloud Consulting Firm in Bengaluru Went From Zero AI Citations to Appearing in 4 of 10 Buyer Prompts in 11 Weeks
A mid-sized AWS cloud consulting firm in Bengaluru was invisible across ChatGPT, Perplexity, Gemini, and Claude despite holding an AWS Premier tier partnership and a domain authority of 61. The root cause was a robots.txt file written in 2022 that blocked every major AI crawler. Eleven weeks after the fix, the firm appeared in 4 of their 10 agreed buyer prompts.
2. The HR Tech Company That Was Being Described as “Basic ATS Software” by ChatGPT — And How That Changed
A Pune-based HR technology company had built an AI-powered talent matching platform. When enterprise prospects asked ChatGPT about them, the response described them as a basic applicant tracking system — accurate for their 2022 product, not their current one. Correcting the AI's description and building their citation surface took 10 weeks and produced citations in 3 of 8 target prompts.
3. Cosmetic Dentist in Banjara Hills Went From Invisible in AI to Recommended in 6 of 9 Hyderabad Neighbourhoods
A cosmetic and dental implants practice in Banjara Hills, Hyderabad, was not appearing when prospective patients asked ChatGPT or Perplexity for recommended cosmetic dentists nearby. A dominant competitor with 1,750+ Google reviews was being cited in every query. After eight weeks of Local GEO work, the practice appeared in 6 of 9 tracked Hyderabad neighbourhoods.
4. Interior Design Studio in Bengaluru: From Zero AI Mentions to Recommended Across 5 Neighbourhoods in 7 Weeks
A boutique interior design studio operating from Indiranagar, Bengaluru had strong word-of-mouth referrals but zero presence in AI responses when prospective clients asked ChatGPT or Perplexity for interior designers nearby. Seven weeks of Local GEO work produced a Recommendation Share of 5 of 8 target Bengaluru neighbourhoods.
5. The Cybersecurity Firm That Was Losing Procurement Shortlists Before the First Call
A Bengaluru-based cybersecurity firm offering VAPT, SOC services, and cloud security was being shortlisted by only one in three prospects who found them through AI research, with two competitors consistently cited ahead of them. After deploying citation-native content and closing three specific citation gaps, the firm moved from appearing in 1 of 8 prompts to 5 of 8 in 9 weeks.
6. HealthTech SaaS Platform: How Structured Content Moved the Needle From 0 to 3 of 10 AI Prompts in 45 Days
A Pune-based HealthTech SaaS company providing hospital management software was invisible in AI responses for every buyer query their sales team tracked, while competitors with smaller product footprints were recommended instead because their content was structured for AI extraction. Forty-five days of content intelligence work produced citations in 3 of 10 buyer prompts.
7. When AI Agents Could Not Use a Fintech Website — And How Fixing the Accessibility Tree Unlocked New Leads
A Mumbai-based B2B fintech company offering automated GST reconciliation software was structurally unreadable by AI agents — 62 empty buttons, 48 unlabelled form fields, and all pricing rendered client-side in JavaScript meant Perplexity's agent couldn't complete any action on the site. After an agent readiness overhaul, the company appeared in AI recommendations for three previously uncaptured buyer queries.
8. IT Services Company in Chennai: Making the Careers Page Readable by AI Agents
A Chennai-based IT services company's careers page was completely invisible to AI agents because every job listing loaded dynamically via JavaScript after a user interaction — AI platforms crawling for queries like “companies hiring cloud engineers Chennai” received an empty page. After converting to server-rendered listings with structured schema, the company began appearing in talent-discovery queries.
9. EdTech Platform in Hyderabad: Correcting Three Factual Errors That Were Costing Enterprise Sales Calls
A Hyderabad-based corporate training and EdTech platform was losing enterprise sales calls because ChatGPT and Gemini stated three factual errors about the company — wrong founding year, wrong client count, and a certification it had never claimed. Correcting the record through entity management and structured content took six weeks and fixed the AI responses across four platforms, with one external source update still pending.
10. Chartered Accountancy Firm in Bengaluru: Local GEO for Professional Services in a High-Trust Buying Category
A Bengaluru-based chartered accountancy firm specialising in startup taxation, international transfer pricing, and GST advisory wasn't appearing when prospective clients asked AI for CA firm recommendations nearby — despite 18 years in operation and a strong referral network, they had zero AI recommendation share. Eight weeks of Local GEO work produced citations in 4 of 7 target Bengaluru neighbourhoods.
Related Reading
Cross-links from the index page into the research library — ties the case studies back to the methodology and technical concepts behind them.
Frequently Asked Questions
Are these case studies from real clients?
Yes. Every case study documents a real Dattva engagement. Company names are withheld and industries are described generically, consistent with the confidentiality terms every client signs — but the city, discipline, timeline, and before-and-after result shown are accurate to that engagement.
How is each result measured?
Every case study is measured against the same Money Prompt framework used across all Dattva engagements: a fixed set of buyer queries is agreed with the client, run in ChatGPT and Perplexity before the engagement starts, and re-run on a schedule through the engagement. The before-and-after counts shown in each case study come directly from that comparison, not a proprietary score.
Why are the case studies anonymised?
Client confidentiality is a standard term of every Dattva engagement. Anonymising the company name doesn't anonymise the result — industry, city, discipline, and the quantitative outcome are always disclosed, so each case study stays specific enough to be useful and checkable in kind.
Which discipline is right for my business?
As a rough guide: category-level AI visibility problems point to AI Visibility Monitoring or Citation Gap Intelligence; a location-dependent business points to Local GEO; an AI platform describing the company inaccurately points to Hallucination Detection; unstructured web content points to Content Intelligence or the GEO Content Engine; and a website AI agents can't navigate points to Agent Readiness. The free AI Visibility Diagnostic identifies which gap applies before any engagement starts.
How long before I see results like these?
Engagement windows shown here range from 45 days to 11 weeks. Local GEO and hallucination-correction results tend to surface fastest, typically within 6 to 8 weeks. Broader AI Visibility and Content Intelligence gains, which depend on AI platforms re-crawling and re-indexing content, more often take the full 90-day cycle to compound.
Can I verify these results myself?
Not for the anonymised case studies specifically, since the company names aren't published. But the underlying method is fully transparent: any current client can open ChatGPT or Perplexity and run their own agreed Money Prompts at any time to see the same kind of before-and-after comparison for their own brand.
See what GEO implementation could produce for your brand
Get your free AI Visibility diagnostic across ChatGPT, Perplexity, Gemini, and Claude — with prioritised, copy-paste fixes at no cost.