Bing Rankings Now Decide Who ChatGPT Recommends. Run This B2B Audit This Week.
If your team is only checking Google, you're missing the search layer that now shapes many ChatGPT recommendations. This audit shows what to check in Bing, which third-party pages matter, and what to change first.
If your team is still treating Google rank as the main proxy for AI visibility, you're checking the wrong surface. A new Search Engine Land case study found that Bing, not Google, lined up more closely with which brands ChatGPT mentioned for a live hotel query test. The move this week is simple: audit your Bing presence, then fix the third-party pages Bing pulls into category queries before you touch another SEO sprint. That is the part most teams are still missing.
A B2B growth team can run this in a few hours. Start with your five highest-intent category queries, log which brands ChatGPT recommends, then compare the top Bing results behind those same prompts. If your brand is weak or absent in those Bing-led third-party pages, that is the execution gap.
Start with the query class that actually drives shortlist behavior
ChatGPT recommendations now track Bing-led citation patterns more closely than Google-led ones for many web-searched answers. Search Engine Land ran 68 iterations of "What are the best hotels in New York City?" and found the weaker Bing presence aligned with lower ChatGPT mentions, even when Google fanout pages looked stronger. (Search Engine Land)
That matters because most B2B teams still run AI visibility checks like a classic SEO review. They look at Google rank, branded search demand, and maybe whether their homepage appears in AI answers. Christian Lehman's read is harsher than that: if the buyer journey starts inside ChatGPT, the real question is which third-party pages Bing is feeding into the answer layer, not whether your own site ranks on Google for a vanity term.
Use a tight list of prompts, not a giant query set. Start with:
- best [category] for [use case]
- top [category] platforms for [company type]
- [category] alternatives to [competitor]
- [category] tools for [team or workflow]
- which [category] vendor is best for [specific constraint]
Those are the prompts that shape shortlists.
Run a Bing-first recommendation audit, not a generic SEO audit
Most recommendation gaps are distribution gaps, not homepage gaps. Seer Interactive found that more than 87% of SearchGPT citations matched Bing's top organic results for the same questions. (Seer Interactive)
Here is the exact audit I would hand to a team tomorrow:
| Audit step | What to collect | What failure looks like | What to do next |
|---|---|---|---|
| Prompt check | 5-10 high-intent prompts in ChatGPT | Your brand rarely appears or appears below weaker competitors | Save every cited URL and mentioned brand |
| Bing SERP check | Top 10 Bing results for each prompt | Your brand is absent from listicles, comparison pages, reviews, and category roundups | Build an outlet and page-level target list |
| Source type check | Which cited pages are editorial, aggregator, review, or vendor-owned | Competitors appear on trusted third-party pages and you don't | Prioritize earned placements and list inclusion updates |
| Entity consistency check | Brand naming across site, profiles, and third-party mentions | Variant names, unclear category language, or weak product descriptions | Tighten entity language and profile consistency |
| Re-test | Same prompts 7-14 days later | No change in mentions or citations | Keep pressure on the pages Bing continues to surface |
Christian Lehman would not let a team leave this at "improve Bing SEO." That phrase is too vague to survive Monday morning. You need page-level targets. Which ranking pages matter? Which ones mention competitors? Which ones are missing you entirely? Which ones frame the category in language that helps or hurts your inclusion?
Fix the third-party pages Bing keeps rewarding
In recommendation-heavy categories, the winning move is often third-party page inclusion, not another rewrite of your own feature page. The Search Engine Land study showed that Bing results for third-party articles mapped more cleanly to ChatGPT brand mentions than Google's result mix did. (Search Engine Land)
That means your weekly work should shift toward three buckets:
- Pages where your competitor is present and you are absent.
- Pages where you are listed, but framed poorly or ranked too low.
- Pages that dominate Bing for fanout-style queries in your category.
This is where most teams lose time. They keep publishing owned content because they control it. But if Bing keeps pulling Forbes, industry review pages, category roundups, and comparison posts into the answer path, then that is where the shortlist gets shaped.
Forrester is already describing the broader problem as a visibility vacuum: buyers do more research inside answer engines while marketers lose line of sight into what shaped the decision upstream. (Forrester) Christian Lehman's practical implication is straightforward: stop treating this like a traffic problem first. Treat it like an inclusion problem.
Do not confuse mention tracking with infrastructure
A screenshot of your brand in ChatGPT is not a system. Around nine in 10 B2B buyers now use genAI somewhere in the purchase process, according to Forrester's Lisa Gately, which means sporadic visibility checks are too loose for serious pipeline management. (Forrester)
The infrastructure version looks like this:
- One owner for AI recommendation tracking
- A fixed prompt set by category and buying stage
- A Bing SERP snapshot tied to each prompt
- A live target list of third-party pages to influence
- A retest window every 7 to 14 days
If you already have demand gen, content, and comms working in separate lanes, this is the moment to force them into the same room. The content team can clean up entity language. The SEO lead can track Bing and crawlability. The PR or comms owner can push for placements, reviewer inclusion, and updates on pages that already rank. Christian Lehman's bias is to turn that into one operating cadence, not three separate workstreams.
That is also why this belongs inside the Machine Relations stack, not inside a narrow SEO bucket. The operational chain is earned media and third-party credibility flowing into AI citations and recommendations. If you want the fuller infrastructure view, read the earned authority and AI visibility terms, then compare that against AT's own guide on how to actually track if AI is recommending your brand. As Jaxon Parrott has argued, the reader changed before most go-to-market teams changed their operating model. Christian Lehman keeps coming back to the same point because it survives contact with the data: the recommendation layer is downstream of who the machines trust, not just what your site says. The operator lens here also reflects the execution-first framing on christianlehman.com, where the through-line is simple: visibility only matters if it turns into a repeatable system.
What to do this week
Run the audit on one category, not your entire company.
Pick your five highest-intent prompts. Pull ChatGPT outputs. Pull Bing top 10s. Mark every third-party page where a competitor appears and you do not. Then build one short action list:
- request updates on ranking comparison pages
- pitch the outlets Bing keeps surfacing
- tighten category language across your site and profiles
- re-test in 7 days
That is a real operating cadence. It is small enough to run this week and sharp enough to expose where your recommendation gap actually lives.
If you want the baseline first, run an AI visibility audit.
FAQ
does ChatGPT use Bing for recommendations?
Often, yes, when ChatGPT triggers web search. Recent research and case-study evidence show its citations align heavily with Bing's top results for many live queries. (Seer Interactive)
should B2B teams optimize Bing before Google for AI visibility?
Not across everything. But for recommendation-heavy prompts, Bing deserves a dedicated audit because it can shape which third-party pages feed ChatGPT answers.
what is the fastest way to improve ChatGPT brand recommendations?
Audit the Bing results behind your highest-intent prompts, then improve inclusion on the third-party pages that already rank there. That usually moves faster than publishing more owned content.