Author: Amritpal Singh
Profile: https://www.linkedin.com/in/amritpal-singh-marketing
A B2B SaaS company came to me with zero AI search presence.
Not low presence. Genuinely invisible.
I ran their top 5 buyer questions through ChatGPT, Perplexity, and Gemini.
Competitors cited in every answer. Their brand: not mentioned once across 15 queries.
8 weeks later: cited in 34% of the relevant AI search answers in their category.
Here is exactly what changed, and what didn't.
WHAT DIDN'T CHANGE
Their website. Their domain authority. Their existing blog content. Their Google rankings.
No technical SEO overhaul. No link-building campaign. No redesign.
WEEKS 1–2: AI QUESTION MAPPING
Ran 40 buyer questions through ChatGPT and Perplexity. Documented every cited source. Found 3 patterns: which domains got cited most, which content format got cited most, and which question types produced the richest AI answers.
Their competitors' advantage: structured, long-form content answering specific problem questions - published freely, no gates, no promotion.
WEEKS 3–4: REDDIT GEO FOUNDATION
Identified 4 subreddits where their exact buyer type was actively asking the problem questions their product solves.
Wrote 6 long-form answer posts - no promotion, no product mentions, genuine practitioner-level advice.
Posted during peak hours. Responded to every early comment within 2 hours.
By week 4: 3 posts above 100 upvotes. First Perplexity citation detected at day 19.
WEEKS 5–6: OWNED CONTENT LAYER
Built 4 long-form pieces on their site directly answering the buyer questions getting the most AI search volume.
Structured for AI extraction: headers, numbered steps, specific data points, first-person experience framing.
Cross-linked from the Reddit posts in follow-up comments once posts had traction.
WEEKS 7–8: CITATION TESTING AND OPTIMISATION
Tested 40 buyer queries across 3 AI engines weekly. Tracked citation rate from 0% to 34%.
Highest-citation format discovered: 800–1,200 word structured answer + one real data point + one named tool + first-person experience framing.
THE 8-WEEK RESULT
AI citation rate: 0% → 34% of relevant buyer queries.
AI referral traffic in GA4: appearing from week 5 onwards.
First inbound lead from AI search citation: day 47.
Pipeline from GEO in month 3: 6 qualified conversations, 2 in active proposal stage.
The founder: 'We were invisible to the AI search ecosystem. Now we're part of it. The leads feel different, they already understand why they need this.'
→ Debate I want to close the week with: when a B2B company has zero AI search presence right now, what do you think is the biggest barrier to building it, not knowing which buyer questions to target, not having content that AI engines want to cite, or not having a community distribution channel like Reddit to seed the citations from?
I've seen all three and they need completely different starting points.
Not low presence. Genuinely invisible.
I ran their top 5 buyer questions through ChatGPT, Perplexity, and Gemini.
Competitors cited in every answer. Their brand: not mentioned once across 15 queries.
8 weeks later: cited in 34% of the relevant AI search answers in their category.
Here is exactly what changed, and what didn't.
WHAT DIDN'T CHANGE
Their website. Their domain authority. Their existing blog content. Their Google rankings.
No technical SEO overhaul. No link-building campaign. No redesign.
WEEKS 1–2: AI QUESTION MAPPING
Ran 40 buyer questions through ChatGPT and Perplexity. Documented every cited source. Found 3 patterns: which domains got cited most, which content format got cited most, and which question types produced the richest AI answers.
Their competitors' advantage: structured, long-form content answering specific problem questions - published freely, no gates, no promotion.
WEEKS 3–4: REDDIT GEO FOUNDATION
Identified 4 subreddits where their exact buyer type was actively asking the problem questions their product solves.
Wrote 6 long-form answer posts - no promotion, no product mentions, genuine practitioner-level advice.
Posted during peak hours. Responded to every early comment within 2 hours.
By week 4: 3 posts above 100 upvotes. First Perplexity citation detected at day 19.
WEEKS 5–6: OWNED CONTENT LAYER
Built 4 long-form pieces on their site directly answering the buyer questions getting the most AI search volume.
Structured for AI extraction: headers, numbered steps, specific data points, first-person experience framing.
Cross-linked from the Reddit posts in follow-up comments once posts had traction.
WEEKS 7–8: CITATION TESTING AND OPTIMISATION
Tested 40 buyer queries across 3 AI engines weekly. Tracked citation rate from 0% to 34%.
Highest-citation format discovered: 800–1,200 word structured answer + one real data point + one named tool + first-person experience framing.
THE 8-WEEK RESULT
AI citation rate: 0% → 34% of relevant buyer queries.
AI referral traffic in GA4: appearing from week 5 onwards.
First inbound lead from AI search citation: day 47.
Pipeline from GEO in month 3: 6 qualified conversations, 2 in active proposal stage.
The founder: 'We were invisible to the AI search ecosystem. Now we're part of it. The leads feel different, they already understand why they need this.'
→ Debate I want to close the week with: when a B2B company has zero AI search presence right now, what do you think is the biggest barrier to building it, not knowing which buyer questions to target, not having content that AI engines want to cite, or not having a community distribution channel like Reddit to seed the citations from?
I've seen all three and they need completely different starting points.