AI Overviews Hit ~93% of Brand Queries: Mid-SERP Audit for B2B SaaS

Brand search used to feel like the last quiet room in SEO.

That assumption cracked on Sept 30. Chris Long’s post and Barry Schwartz’s write-up put numbers and screenshots on what operators were starting to feel: Google is putting AI Overviews on a large share of branded queries, including household and B2B names.

For pipeline teams, the risk is not only a missing blue-link click. It is a third-party story appearing next to your official site on the query that used to be yours by default.

---

What the Sept 30 evidence actually shows

Chris Long’s brand panel

Long’s claim is specific and limited:

• Sample: about 100 brands
• Result: 93% showed an AI Overview
• Named examples: Reddit, Salesforce, Amazon, Adobe, and more

That is a practitioner panel, not a peer-reviewed census. It is still useful because it names B2B-relevant brands (Salesforce) and matches what Barry could reproduce the same day.

Where the Overview sits on the SERP

Search Engine Roundtable is the placement detail most teams will miss if they only read the 93% headline:

• Many brand AIOs sit lower on the page, under the main company result or further down
• Some brands (Adobe in Barry’s checks) still get a top-of-page AIO
• Google’s own brand and many news publishers were exceptions in his spot checks

Mid-page does not mean low stakes. Buyers who scroll still read a summary assembled from sources Google chose. Your homepage can sit in position one while a listicle or review domain narrates the product in the Overview.

DemandSphere’s September curve (with COI)

DemandSphere’s Sept 30 post frames the same story as a month-long chart:

• Early September baseline: about 26.9% average AIO presence on branded keywords (Sept 1 to 5)
• Peak day: 90.5% on Sept 27
• Late month: 82.1% on Sept 29, after holding above 80% post-spike
• Sustained climb begins Sept 26; earlier spikes on Sept 18, 20, 23, and 24 fell back toward ~30%

Conflict of interest: DemandSphere sells AI Overview and AI Mode tracking. Use the direction and timing; do not treat their panel as industry truth without your own brand-query checks.

Definition to keep straight: they count AIO presence on a branded keyword whether or not your domain is cited inside the Overview. That is the right metric for whether Google is answering on your name, and the wrong metric for whether Google recommends you.

---

What this measures that classic brand SEO misses

1. Brand SERP feature rate: does an AIO fire on brand, brand pricing, brand vs rival, and brand alternatives queries?
2. AIO position band: top vs mid vs buried (Barry’s mid-page pattern)
3. Citation set on brand intents: which third-party URLs ground the summary (listicle layer logic still applies when Google needs consensus)
4. Narrative fidelity: does the Overview match your packaging, ICP, and security claims, or a stale competitor frame?

Search Console’s generative AI report can show impressions when your pages appear in AI features. It still will not tell you what the Overview said about you on a navigational query, or which rival review powered it.

---

Why B2B SaaS should care this week

1. Navigational monopoly is gone (or going). If Salesforce-class brands get Overviews, mid-market SaaS should not assume immunity.
2. Brand queries are reputation queries. DerivateX’s brand grounding and hallucination fix work was built for chat engines. Branded AIOs bring the same problem onto Google Search.
3. Third parties still write the shortlist. On commercial prompts, third-party listicles dominate citations. On brand prompts, those same pages can supply the facts Google reuses next to your official result.
4. Clicks and narrative are separate scoreboards. Pair this with the citation is not a click loop and zero-click strategy: even a mid-page AIO can satisfy intent without a session, or send the click to a source that is not you.

---

Operator playbook this week

1. Build a 20-query brand panel

Run these in an incognito or representative locale, screenshot presence + position + cited domains:

• Brand-only query
• Brand + software / platform
• Brand + pricing
• Brand vs each of your top 3 rivals
• Brand + alternatives
• Brand + reviews
• Brand + SOC 2 / security / integrations (pick what buyers actually type)

Log four columns: AIO yes/no, position band (top / mid / none), cited domains, and whether the prose matches current packaging.

2. Separate three outcomes

1. AIO present on the brand query
2. Brand cited or named inside that AIO
3. Brand recommended when the query is comparative

Collapsing those three into one AI-traffic vanity number hides the failure mode where Google answers on your name using someone else’s page.

3. Fix the grounders Google is already reading

When the Overview cites a stale G2 blurb, an old alternatives post, or a partner page with the wrong plan names, treat that as a content ops ticket, not a vanity mention. Update first-party pricing, security, and integration pages so extractable facts are current. Then chase corrections or replacements on the third-party URLs in the citation set.

4. Watch brand CTR and assisted pipeline, not only AIO rate

If branded AIO rate jumps and brand CTR falls, do not only celebrate visibility. Reconcile Search Console brand queries, generative AI impressions (when available), and CRM opportunities that started with a brand search or AI-influenced session.

---

Worked example: mid-market RevOps SaaS

Assume your brand is Acme, rivals are RivalA and RivalB.

Monday panel:

• Acme: AIO mid-page; cites a generic explainer + one G2 roundup; packaging year is wrong
• Acme pricing: AIO top; cites your pricing page + a 2024 affiliate comparison
• Acme vs RivalA: AIO top; recommends RivalA; cites two listicles you are missing from

Weak reaction: Brand still ranks number one, ignore.

Better reaction this week:

1. Correct plan names and last-updated dates on pricing and trust pages the same day
2. Open tickets for the two listicles that decide Acme vs RivalA
3. Report to the CMO: brand AIO presence, mid vs top rate, citation set quality, and whether brand CTR moved
4. Keep ChatGPT / Perplexity brand prompts on a separate sheet; Google branded AIO is one surface, not the whole program

---

How DerivateX thinks about branded AI Overviews

1. Brand SERPs are now an answer surface, not only a navigational SERP.
2. Measure presence, position, citation set, and narrative fidelity as four lines.
3. Prefer fixing extractable first-party facts and the third-party pages Google already trusts over prompt theater.
4. Keep Google branded AIOs separate from ChatGPT brand mentions; engines do not share one blended score (citation overlap is still low across systems).

If you need a prompt-level baseline across Google AI Overviews, AI Mode, ChatGPT, Perplexity, Gemini, and Claude, start with a free AI visibility audit. Engagement options are on pricing. For proof patterns, see the Gumlet and REsimpli case studies.

---

FAQ

Is 93% a universal rate?

No. It is Chris Long’s ~100-brand panel on Sept 30. DemandSphere’s tracked branded keywords show a late-September jump into the 80% to 90% range. Run your own brand panel before you quote either number to a board.

Does a mid-page AI Overview matter if we still rank number one?

Yes, if the Overview misstates pricing, ICP, or security, or if it satisfies the query without a click. Rank number one and narrative control are different jobs.

Does AIO presence mean Google cited our site?

Not necessarily. DemandSphere’s presence metric explicitly ignores whether the brand is cited inside the Overview. Always open the citation cards.

Should we panic about brand CTR?

Measure first. Some mid-page AIOs may barely move CTR; top-of-page ones may. Track brand query CTR alongside AIO presence for two to four weeks before changing budget.

What should we change this week?

Ship the 20-query brand panel, fix stale first-party facts the Overviews already touch, and open third-party correction or placement tickets for the URLs Google is citing on brand and brand-vs queries.