From Invisible to Cited: Building AI Search Visibility for a Direct-to-Consumer Brand
A 6-month SEO & AI-visibility case study — January to June 2026, with figures and screenshots sourced directly from the monthly client reports
The client, at a glance
The client is a boutique e-commerce brand selling premium gift products online. EbizON was engaged to grow organic and AI-driven discovery — making sure the brand shows up not just in Google's ten blue links, but inside the answers ChatGPT, Google AI Overviews, AI Mode, and Gemini give shoppers researching relevant gifts.
22 → 26
Domain Rating
Jan 2026 → Jun 2026
848 → 4,100
total backlinks
Dec 2025 → Jun 2026
13 → 130
AI-referral sessions/mo.
Jan 2026 → Jun 2026
332 / 454
tracked AI prompts citing the client (Jun)
ENGAGEMENT WINDOW:
Jan – Jun 2026
6 monthly report cycles across GA4, Search Console, Ahrefs/SE Ranking, and a dedicated AI-visibility tracker.
1. THE PROBLEM
Ranking on Google, invisible in AI answers
JANUARY 2026 BASELINE — before EbizON's AI-visibility program began, sessions were down 16.9% year-over-year (4,134 in Jan'26 vs. 4,978 in Jan'25), and the site had plenty of topical relevance but almost none of it converted into AI or search-answer visibility.
0.15% CTR on the questions people actually ask
Jan 2026: 9,381 impressions on "question" queries like "what is truffle" and "what is aphrodisiac chocolate" produced just 14 clicks — a 0.15% site CTR. The content ranked; it was never surfaced as an answer.
"Medium," then "Low" AI visibility
Feb 2026: AI Visibility score opened at 40/100 ("occasionally mentioned in LLM outputs"). By Apr 2026 it read 30/100 ("rarely mentioned... compared to competitors") as the AI-answer landscape got more competitive.
Thin domain authority
Dec 2025–Feb 2026: Domain Rating stuck at 22 for three straight months, with only 848–947 backlinks across ~278 referring domains — too little third-party authority for AI engines to trust and cite.
On-page issues blocking crawlers
Jan 2026: 14+ pages missing an H1 tag. Feb 2026: 11 more pages found with duplicate H1s — structural problems that confuse both search crawlers and the content-extraction pipelines LLMs rely on.
Proof of problem- Straight from the January 2026 dashboard
Why it matters: AI answer engines reuse the same crawlers and content-extraction logic as search — pages with weak structure or zero click-through are just as invisible to an LLM as they are to a snippet box.
GA4 Analytics Overview, January 2026 report — Sessions 4,134 (Jan'26) vs. 4,978 (Jan'25); AI Traffic 76 vs. 23; Item revenue $7,473 vs. $6,863.
Search Console "Question Queries" segment, January 2026 report — 14 clicks / 9,381 impressions / 0.15% Site CTR / avg. position 10.87.
2. OUR ANALYSIS
A single dashboard, four data sources
GA4
Sessions, AI-referral traffic, content engagement, and the add-to-cart → checkout → purchase funnel.
Search Console
Impressions, clicks, and position — isolating a "Question Queries" segment to flag high-impression, zero-click informational searches as a proxy for AI/snippet invisibility.
Ahrefs / SE Ranking
Domain Rating, backlinks, referring domains, keyword-rank tracking, plus a monthly on-page technical audit.
AI Visibility Tracker
AI Visibility score, mentions & citations by platform (ChatGPT, AI Overview, AI Mode, Gemini), Top Cited Pages, Top Performing Topics, and a head-to-head benchmark against three named local competitors.
AI Competitor Analysis, Feb 2026 report — the client scored 40 vs. 24–27 for three tracked rivals, with an explicit flag: "In the Personalized and Promotional Gifts topic, competitor brands are mentioned but yours isn't."
3. WHAT WE DID
a. Making the site legible to crawlers and LLMs alike
TECHNICAL FOUNDATION
Fixed missing / duplicate H1 tags across dozens of pages
Resolved crawl errors, verified robots.txt, repaired broken internal links
Corrected URL structure and cleaned up domain redirects
Added descriptive alt tags across product and CDN images
Connected GSC ↔ GA4 for clean AI-referral attribution
Re-submitted every new or updated URL to Search Console on a monthly cadence to keep indexing (and AI crawling) fresh
ANSWER-FIRST CONTENT
Mined Search Console's Question Queries to find exactly what people ask ("why does chocolate bloom," "what is parve chocolate," "kosher vs. non-kosher chocolate")
Published and refreshed FAQ-style, directly answerable explainer posts around those real questions
Ran a rolling seasonal content calendar — Valentine's, Passover, Purim, Mother's Day, Rosh Hashanah, Sukkot, Tu B'Shvat — to keep giving AI engines fresh, citable material
Prioritized topics flagged by the AI tracker as competitor-mentioned / brand-absent gaps
b. Earning the authority AI engines cite
AI answer engines lean heavily on third-party corroboration. EbizON ran a continuous digital-PR and link-building program alongside monthly AI-performance monitoring to keep closing the gap.
DIGITAL PR & LINK BUILDING
30+ hand-placed backlinks with branded, locally relevant anchor text ("local chocolate shop," "kosher chocolate covered strawberries")
10 earned listicle placements on high-authority publishers (DR 61–75, incl. MSN.com, NextLuxury, Kalkine Media)
Result: Domain Rating 22 → 26, referring domains 280 → 388, total backlinks 848 → 4,100 over 7 months
CONTINUOUS AI MONITORING
Tracked AI Visibility score and mentions/citations by platform (ChatGPT, Google AI Overview, AI Mode, Gemini) every month
Benchmarked mention share against three named local competitors to spot and close gaps Benchmarked mention share against three named local competitors to spot and close gaps
Monitored Top Cited Pages and Top Performing Topics to steer the content calendar toward the highest opportunity prompts each cycle
AI Competitor Analysis, Feb 2026 report — the client scored 40 vs. 24–27 for three tracked rivals, with an explicit flag: "In the Personalized and Promotional Gifts topic, competitor brands are mentioned but yours isn't."
c. The work behind the numbers, one month at a time
| Month | Technical / on-page | Content published | Off-page / authority |
|---|---|---|---|
| Jan 2026 | Audited & flagged 14+ pages missing H1 tags; 1 GSC resubmission | Yes | 5 backlinks (DR 34–59) – local / kosher / strawberries anchors |
| Feb 2026 | Flagged 11 pages with duplicate H1 tags; 2 GSC resubmissions; 1 blog update | Yes | 5 backlinks (DR 36–61) + 10 listicle placements (DR 61–75) on MSN, NextLuxury, Kalkine Media |
| Mar 2026 | Continued H1 remediation across flagged pages | Yes | 5 backlinks (DR 33–63) |
| Apr 2026 | Fixed crawl errors, verified robots.txt, repaired broken links, corrected URL structure, domain redirects; 6 GSC resubmissions; alt-tags added (166 logged actions) | 4 seasonal blog updates | 5 backlinks (DR 30–63) + 5 content-syndication placements (Medium, Evernote, Wakelet, JustPaste.it, LiveJournal) |
| May 2026 | ~42 logged actions: GSC resubmissions & content updates | Mother's Day platters, sympathy/shiva collection, parve-chocolate & personalized-chocolate guides | 5 backlinks (DR 30–63) + 5 content-syndication placements |
| Jun 2026 | 20 logged actions: GSC resubmissions & content updates | Tu B'Shvat, Passover, Rosh Hashanah, Sukkot seasonal pages | 5 backlinks (DR 38–69) – gift baskets / Mother's Day / local chocolate anchors |
Every backlink count, listicle, and on-page fix in this table is logged in the client's own monthly report activity tables — nothing here is estimated.
Jan 2026 On-Page log: the H1 audit that kicked off the cleanup.
Jun 2026 On-Page log: same discipline, now on seasonal pages.
4. RESULTS
a. Domain authority climbed overall
Source: monthly Ahrefs / SE Ranking snapshots, EbizON SEO reports, Dec 2025 – Jun 2026.
b. Sessions: every reporting window, before vs. After
Jan'26 is the one window where the current period trails the prior year — the starting problem this program was built to reverse. Every window after it, current exceeds prior.
Source: GA4 Overview, each EbizON report, Jan–Jun 2026. Jan'26 vs. Jan'25 and Feb'26 vs. Feb'25 are both reported directly; all later "prior period" totals are back-calculated from each report's own stated % change (current ÷ (1 + %Δ)).
c. Revenue grew versus every prior period — no exceptions
Source: EbizON monthly / quarterly / semiannual SEO reports, Conversions module, Jan–Jun 2026. Each bar's absolute total revenue for that reporting window is printed below it.
d. Out-mentioned every tracked local rival
Live proof: AI Competitor Analysis, May 2026 report — the exact numbers behind the chart above.
GA4 "AI Assistant Traffic," June 2026 report — 36 users, 45 sessions, 57s avg. engagement.
332 / 454
Tracked AI prompts explicitly cite a client page as a source (Jun 2026)
45 sessions
Landed via a brand-new "AI Assistant" GA4 channel in June — a category that didn't exist as trackable traffic in January
5. THEN VS. NOW
AI visibility: where we started vs. where we are now
| Metric | Then (start of tracking) | Now |
|---|---|---|
| AI-referral sessions to priority pages | 13 sessions — Jan 2026 | 130 sessions — Jun 2026 (10×) |
| Question-query answer capture | 0.15% site CTR — Jan 2026 only (14 clicks / 9,381 impr.) | 0.22% site CTR — cumulative H1 2026 (149 clicks / 66,432 impr.) |
| Domain authority (Ahrefs) | DR 22 · 947 backlinks · 278 ref. domains — Jan 8, 2026 | DR 26 · 4,100 backlinks · 388 ref. domains — Jun 30, 2026 |
| Vendor AI Visibility score (0–100) | 40/100 "Medium" — first tracked, Feb 2026 | 30/100 "Low" — most recent reading, Apr 2026* |
| Mention share vs. 3 local competitors | Not yet tracked before May 2026 | 235 mentions vs. 40–79 for rivals — May 2026 (3–6× lead) |
| AI-answer citation footprint | Not yet tracked in Q1 2026 | Cited in 332 of 454 tracked prompts — Jun 2026 |
*The vendor's 0–100 score moved with a fast-changing benchmark as more brands entered AI-answer tracking — read alongside the rows below it, the underlying substance (referral traffic, authority, competitive mention share, citations) all grew over the same window.
"Then": AI Performance gauge, Apr 2026 report.
"Now": AI Citation Performance, Jun 2026 report — the client's domain appears among cited domains across multiple live AI prompts.
6. YEARLY COMPARISON
H1 2026 vs. H2 2025: every core metric up
GA4 Analytics Overview, June 2026 report — Sessions 26,350 (+7.4%), the raw figures behind the index chart on the left.
Feb 2026 vs Feb 2025: Sessions +17.7% · Item revenue +72.1% · AI-referral sessions +66.7% Jan 2026 vs Jan 2025: Sessions were down 16.9% — the baseline this program reversed.
7. KEY TAKEAWAYS
What moved AI visibility
- Technical hygiene is a prerequisite, not a nice-to-have — H1 tags, crawl errors, and broken links were quietly blocking both Google and AI crawlers — fixing them came first.
- Answer real questions, not just keywords — Content built directly around Search Console's zero-click "question queries" is what AI engines had material to cite.
- Authority still has to be earned off-site — Backlinks and listicle placements nearly quintupled (848 → 4,100) — the third-party corroboration AI models weight when choosing what to cite.
- Consistency compounds — Revenue grew against its comparable prior period in every single report — Feb, Q1, Q2, H2'25→H1'26 — not one isolated spike.
- A new channel emerged, not just better rankings — By June, GA4 recorded real sessions from an "AI Assistant" channel that simply didn't exist as trackable traffic in January.
- Competitive position, not just absolute score — The vendor AI-visibility score moved with a fast-changing landscape, but the client's mention volume stayed 3–6× ahead of its three closest tracked local competitors throughout.