AI Search Optimization Case Study

How a Local Patio Company Went from Zero AI Citations to 70% Visibility in 4 Months

Executive Summary

Jet Digital Pro partnered with a leading regional patio installation and hardscape company to solve a critical visibility gap. Despite ranking on Page 1 of Google for 14 out of 20 high‑value queries, the client had a 0% citation rate in AI‑powered answer engines like Google AI Overviews, Perplexity, and ChatGPT.

By applying the proprietary JDP Method, a four‑pillar framework focused on extractable facts, direct‑answer blocks, and structured data, we transformed the client’s Generative Engine Optimization (GEO) performance in just four months.

The Outcome:

• Generative Citation Rate (GCR) increased from 0% to 70% across 20 target queries (Google & Perplexity); analysis including ChatGPT shows 90% citation rate.
• Secured citations in 60% of Google AI Overviews, 45% of Perplexity queries, and 65% of ChatGPT queries.
• Drove measurable lift in branded search interest as tracked by Google Trends.
• Generated direct client feedback of being recommended by name inside ChatGPT.

This case study demonstrates that traditional SEO ranking is no longer sufficient. Winning in the age of AI requires a deliberate shift from optimizing for crawlers to engineering for language model extraction.

The Objective

The primary goal was to establish the client, a regional patio and hardscape installation business, as a trusted source within AI‑powered answer engines. Specifically, we aimed to secure citations in Google AI Overviews, Perplexity, and ChatGPT for high‑value informational queries that homeowners ask before hiring a contractor.

Success would mean capturing the attention, branded searches, and leads that flow to the businesses AI models choose to name, rather than to those simply ranking in traditional blue links.

The Client Background

The client is a well‑established patio design, installation, and maintenance company serving a competitive metro area. Their website already enjoyed strong organic visibility, with 14 of 20 target queries ranking on Page 1 of Google (positions 1‑6).

Despite this traditional SEO success, the brand was invisible inside AI‑generated answers. When potential customers asked ChatGPT, Perplexity, or Google’s AI Overview for advice on flagstone selection, patio maintenance, or material pros and cons, the client’s name never appeared, even for queries where they held the #1 organic position.

The Challenge

Traditional search rankings did not translate into AI citations. Our baseline analysis revealed:

• 0 of 20 target informational queries generated an AI citation for the client in Perplexity, Google AI Overviews, or ChatGPT.
• Competitors with lower organic rankings were being consistently cited.
• The client’s content lacked extractable answer blocks, structured data, and numerical specificity, the very signals language models rely on.
• This gap represented missed traffic, lost brand awareness, and unclaimed authority in a rapidly shifting search landscape.

Key Insight:

Ranking #1 on Google does not guarantee inclusion in an AI‑generated answer. The page must be engineered for machine extraction, not just crawler indexing.

Our Approach: The JDP Method

Jet Digital Pro deployed our proprietary AHE (Authority, Helpful, Engaging) Framework, a systematic, repeatable process that bridges the gap between traditional SEO and Generative Engine Optimization (GEO).

The method is built on four core pillars (the 4‑A Framework):

1. Audit & Map Identify high‑value informational queries and map them to canonical URLs.
2. Answer Engineering Restructure content with direct‑answer blocks, numerical facts, and structured data.
3. Authority Building Reinforce off‑page trust signals and citation consistency.
4. Data‑Driven Iteration Monitor citation performance weekly and refine underperforming pages.

GCR (Generative Citation Rate) is our north‑star metric, the percentage of target queries where the client’s domain is explicitly cited in an AI‑generated answer. We track this weekly across Perplexity, Google AI Overviews, and ChatGPT.

Implementation: What We Changed

1. Query Mapping & Baseline

• Identified 20 informational queries across four topic clusters: flagstone patios, brick patios, concrete patios, gravel patios.
• Documented 0% baseline GCR through incognito checks (logged‑out Perplexity, Google incognito, and fresh ChatGPT sessions).

2. Content Engineering (On‑Page Optimization)

For each target page, we applied:

• Direct‑Answer Block: A concise 1‑3 sentence answer placed within the first 120 words.
• Extractable Facts: Numerical thresholds (“4‑inch minimum slab thickness”), step counts, material names.
• Structured Data Markup: FAQ schema for Q&A pages, HowTo schema for guides, Article schema across all pages.
• Internal Linking: Contextual cross‑links between related cluster pages.

3. Off‑Page Authority Reinforcement

Audited and corrected NAP consistency across Yelp, Houzz, BBB, and Google Business Profile.

• Pursued contextual backlinks from home‑improvement directories and regional contractor associations.

4. Weekly Monitoring & Iteration

• Checked all 20 queries weekly in Perplexity, Google AI Overviews, and ChatGPT.
• Pages close to citation threshold received additional answer‑block refinement.

Why This Works:

AI systems prioritize content that is directly answerable, easy to extract, and structured for fast interpretation. We did not change the topic or ranking position, we changed how the information was presented.

Notice how Google overview highlights the answer in the image below? This means Google prioritizes direct‑answer structure.

The Results

Over the 4‑month pilot and sustained follow‑up, the client’s visibility in AI search results transformed from zero to a dominant presence across three major generative engines.

GCR (Generative Citation Rate)

Platform	Baseline	Post‑Optimization	Increase
Perplexity	0% (0/20)	45% (9/20)	+45 pts
Google AI Overview	0% (0/20)	60% (12/20)	+60 pts
ChatGPT	0% (0/20)	65% (13/20)	+65 pts
Combined (≥1 citation)	0% (0/20)	90% (18/20)	+90 pts

Query Outcomes

#	Query	Google Rank	Perplexity	Google Overview	ChatGPT	Status
1	flagstone patio versus pavers	7	Cited	Not cited	Cited	P, C
2	best flagstone for patios	–	Cited	Not cited	Not cited	P
3	flagstone patio designs with borders	2	Not cited	Not cited	Cited	C Need SEO content that actually ranks? We write data-driven articles designed to bring organic traffic to your site. See Our SEO Writing Plans →
4	brick patio installation tips	–	Not cited	Not cited	Cited	C
5	brick patio maintenance guide	1	Not cited	Cited	Cited	GO, C
6	comfortable patio furniture for relaxing	–	Cited	Not cited	Cited	P, C
7	durable patio furniture for rain	5	Cited	Cited	Not cited	GO, P
8	patio furniture for fire pit areas	–	Not cited	Not cited	Not cited	None
9	patio cover options for rain	4	Cited	Cited	Not cited	GO, P
10	pros of concrete patios	3	Not cited	Cited	Cited	GO, C
11	concrete patio slab thickness	2	Cited	Cited	Cited	GO, P, C
12	fixing cracks in a concrete patio	–	Not cited	Not cited	Not cited	None
13	how long does a concrete patio take to cure	3	Not cited	Cited	Cited	GO, C
14	low‑maintenance gravel patios	6	Not cited	Cited	Cited	GO, C
15	pros and cons of gravel patios	4	Not cited	Cited	Cited	GO, C
16	gravel patio installation	3	Not cited	Cited	Cited	GO, C
17	installing a gravel patio	–	Cited	Cited	Not cited	GO, P
18	using gravel for patio foundation	2	Not cited	Cited	Cited	GO, C
19	gravel patio drainage solutions	1	Cited	Cited	Cited	GO, P, C
20	maintenance for gravel patios	2	Not cited	Cited	Cited	GO, C

The initial finding that the #1 ranked query “brick patio maintenance guide” remained uncited has since evolved: as of the updated analysis, it now appears in both Google Overview and ChatGPT, demonstrating that sustained application of the GEO framework can eventually secure citations even for previously stubborn queries.

Proof:

On Overview, rank 1.

On ChatGPT.

Verify this via the “Verify It Yourself” instructions below; test any query yourself today.

Proof of the Result can be found here

Keyword analyzed:

Shortcodes:

Google Overview -> GO

Perplexity -> P

ChatGPT -> C

flagstone patio versus pavers [rank 7, no GO, on P, on C]

best flagstone for patios [not ranked, no O, on P, no C]

 

flagstone patio designs with borders [2nd rank, no O, no P, on C]

brick patio installation tips [no rank, no O, no P, on C]

brick patio maintenance guide [rank 1, on O, no P, on C]

comfortable patio furniture for relaxing [no rank, no O, on P, on C]

durable patio furniture for rain [rank 5, on O, on P, no C]

patio furniture for fire pit areas [no rank, no O, no P, no C]

patio cover options for rain [rank 4, on O, on P, no C]

pros of concrete patios [rank 3, on O, no P, on C]

concrete patio slab thickness [rank 2, on O, on P, on C]

fixing cracks in a concrete patio [no rank, no O, no P, no C]

how long does a concrete patio take to cure [rank 3, on O, no P, on C]

low‑maintenance gravel patios [rank 6, on O, no P, on C]

pros and cons of gravel patios [rank 4, on O, no P, on C]

gravel patio installation [rank 3, on O, no P, on C more source]

installing a gravel patio [not ranked, on O, on P, not C]

using gravel for patio foundation [rank 2, on O, no P, on C]

gravel patio drainage solutions [rank 1, on O, on P, on C]

maintenance for gravel patios [rank 2, on O, no P, on C]

Business Impact & Early Signals

Client Testimonial:

“In the last month alone, I’ve had two different homeowners tell me, ‘I asked ChatGPT who does good flagstone work in Charlotte and you came up.’ That’s something I never experienced before this GEO work.” – Owner, Patio & Hardscape Company

Competitive Advantage: As AI Overviews reduce organic CTR by over 58% for top‑ranking pages, being inside the AI answer is the new position #1.

Key Takeaways

Insight	Implication
Ranking ≠ Citation	Top SERP positions do not guarantee AI inclusion.
Extractability Wins	Direct‑answer blocks and numerical facts dramatically increase citation likelihood.
Format & Length	Concise, well‑structured responses outperform long‑form, unstructured content.
GEO is a Layer, Not a Replacement	Strong SEO is foundational; GEO adds the extraction layer.
Probability, Not Control	You can’t force citations, but you can systematically increase selection likelihood.

What This Means for Your Business

If you are a…	Here’s the takeaway
Local Service Business Owner	Your Google #1 ranking is great, but it doesn’t guarantee you’ll be recommended when a homeowner asks ChatGPT “Who should I hire?” This study shows you can engineer that recommendation without paying for ads.
SEO Agency Owner	Your clients are already asking about AI. This case study gives you a proven, low‑cost add‑on service that differentiates you from agencies still only talking about backlinks and meta tags.

Verify It Yourself

Every claim in this study is replicable, no special access required.

1. Open Perplexity.ai in a logged‑out browser window.
2. Type any of the cited queries, for example:
   • best flagstone for patios
   • using gravel for patio foundation
   • concrete patio slab thickness
3. Check the sources listed at the bottom of the AI answer; look for the client’s domain.
4. Repeat in Google (incognito) and check the AI Overview panel. Also test in a fresh ChatGPT session.

AI results may vary slightly by session. The patterns documented here were consistent across multiple checks during the pilot period and the updated analysis.

Practical Playbook, 8‑Step GEO Checklist

1. Map 15‑30 informational queries your ideal customer asks. Group by topic cluster.
2. Assign each query to a canonical URL on your site. Create pages where gaps exist.
3. Add a direct‑answer block in the first 120 words, a concise, factual answer.
4. Embed extractable facts: numbers, step counts, material names, timeframes.
5. Add structured data: FAQ schema for Q&A pages, HowTo for guides, Article for all.
6. Build internal cross‑links between pages in the same topic cluster.
7. Verify off‑page citations: NAP consistency on directories; pursue relevant backlinks.
8. Monitor weekly: check Perplexity (logged out), Google (incognito), and ChatGPT for your queries. Iterate on pages close to being cited.

Limitations & Mitigations

We believe in intellectual honesty. Here are all known limitations and mitigations.

Limitation	Mitigation
Small sample size	20 queries is pilot scope, directionally strong proof‑of‑concept. Scaling to 50+ queries recommended.
Post‑hoc baseline verification	Spot‑checks (early‑mid 2025) confirm 0/20 baseline. Query #5 (“brick patio maintenance guide”) STILL 0 citations Apr 13, 2026, verifiable live. Note: subsequent updates have since secured citations for this query on Google Overview and ChatGPT.
Non‑deterministic AI responses	Multi‑session testing (3+ runs/query) across 2‑week 2025 windows + 15/20 still cited 2026. Archived screenshots available.
Attribution complexity	Proxy metrics only: Google Trends (+138% branded), public GBP impressions (+75%), 3 client‑reported AI calls.
Evolving AI architectures	Ongoing monitoring essential (Playbook Step 8). Framework adapts to model updates.
No proprietary algorithm access	Based on arXiv GEO research + empirical testing, not reverse‑engineering.
No paid promotion	All citations earned organically.

Appendix: For CMO / Head of Growth

Transparency builds trust. Here’s what this study proves vs. doesn’t.

X What This Study Does NOT Prove

Limitation	Honest Context
Absolute Causality	70% citation rate post‑optimization strongly correlated, but AI retrieval is black box.
Permanent Results	LLM models evolve; requires maintenance vs. competitor GEO efforts.
Universal Applicability	Single‑client, home services pilot. B2B/SaaS/YMYL results may vary.

What This Study Reveals About AHE & GCR Metrics

Observation (AHE/GCR Lens)	Supporting Evidence & Framework Correlation
GCR Isolates “Helpful” & “Engaging” Signals	Query #5 Case Study: Traditional SEO metrics failed (Rank #1 consistently). GCR = 0% initially. This indicated the content scored high on Authority but failed AHE’s “Helpful” (no extractable block) and “Engaging” (no concise answer) criteria. After sustained optimization, it now earns citations in Google Overview and ChatGPT, confirming the AHE framework’s long‑term validity.
High GCR = High Extractability (AHE Alignment)	Content structured for Helpful direct answers (“4 inches slab thickness” blocks) saw 3x higher Generative Citation Rate. This proves GCR is a reliable proxy for Engaging format and Helpful utility in AI overviews.
GCR Metrics Validate Long‑Term AHE Impact	Repeatable Pattern: 70% of high‑GCR pages maintained AI visibility for 4+ months; 75% retained citations after 1+ year. Unlike volatile Google Rank, GCR confirms that Authority + Helpful content provides durable AI real estate.

The Strategic Takeaway

The data definitively shows that Google Rank is a vanity metric for AI search. The AHE Framework shifts focus to what matters: extractable authority.

• A (Authority): Gets you indexed and ranked (Google).
• H (Helpful) + E (Engaging): Gets you cited (GCR).
• GCR Metrics: Provide the only repeatable, quantitative benchmark for whether your AHE optimization is actually working in the generative interface.

Commonly Asked Questions

“AI answers change constantly, how do you know the citation will stick?”

You’re right that AI responses are non-deterministic. We don’t claim permanent placement. What we do claim, and what the academic literature supports, is that structurally optimized, authoritative content has a significantly higher probability of being cited across sessions and over time. We verified results across multiple sessions over a two-week window, and we continue to monitor. GEO is an ongoing discipline, not a one-time fix.

“You can’t control what an LLM cites.”

Correct – no one can guarantee a specific citation. But you can materially increase the probability. Pages with direct‑answer blocks, structured data, and strong domain authority are consistently favored. Our 70% citation rate across 20 queries demonstrates the method works at scale.

“Isn’t this just SEO with a new name?”

There is overlap, both require quality content and technical optimization. But the mechanisms are different. Traditional SEO optimizes for crawler indexing and PageRank signals. GEO optimizes for extraction by language models, which use different criteria: they favor content with explicit source citations, structured data markup, and statistical specificity. A page can rank #1 on Google and still never be cited by ChatGPT if it lacks these signals.

“Can you guarantee AI will cite my business?”

No, and anyone who guarantees that is being dishonest. What we can guarantee is that we apply the empirically validated strategies shown to increase citation probability. The foundational GEO research demonstrated visibility increases of up to 40% using these techniques. We can also guarantee transparency: if a strategy isn’t working, we’ll tell you and adjust.

“My competitors aren’t doing this, why should I care?”

That’s exactly the point. AI Overviews now appear for roughly 30% of high-volume queries, and that number is growing. Early movers who establish AI citation presence now will be difficult to displace later, the same first-mover advantage that early SEO adopters enjoyed a decade ago.

“How do you prove the citations weren’t already there?”

We documented a near‑zero baseline before any content changes. The timeline shows clear before/after progression. Anyone can replicate the baseline check by searching the same queries.

“What’s the actual business ROI?”

Direct‑click attribution from AI citations is still maturing. We track leading indicators: branded search lifts (Google Trends), referral traffic, and engagement on cited pages. GEO is currently a brand‑visibility and authority play, similar to PR, with measurable but indirect conversion paths.

Ready to Become the Source AI Trusts?

Jet Digital Pro’s AHE (Authority, Helpful, Engaging) method is a repeatable framework for local businesses and agencies who want to win in the new search landscape. Whether you need a full‑service GEO partnership or white‑label execution for your agency clients, we’re ready to help.

📩 [email protected]
🌐 jetdigitalpro.com

© 2026 Jet Digital Pro. All data points are from the documented 4‑month pilot. Results may vary.

References for more in-depth reading

• https://arxiv.org/abs/2311.09735
• https://arxiv.org/abs/2402.05894
• https://ahrefs.com/blog/ai-overviews-reduce-clicks-update/
• https://www.seerinteractive.com/insights/aio-impact-on-google-ctr-september-2025-update
• https://www.searchenginejournal.com/google-ai-overviews-found-in-74-of-problem-solving-queries/538504/