Is Google Fixing B2B Marketing?

Mosegas 4 hours ago

0 0 6 minutes read

This post is sponsored by Calling Experts. The opinions expressed in this article are those of the sponsors.

Why is my incoming traffic suddenly dropping?

Are B2B leads dropping everywhere?

Are AI Overviews Hurting B2B Leads?

For two decades, the concept was simple: more traffic meant more leads, more leads meant more income. In 2026, that logic crumbles.

In this article, we will consider how this change can be a subtle adjustment.

Why Your Inbound Traffic Volume Went Down (But Your Offers Got Bigger)

Generative AI has successfully taken over the first phase of shopping journey research. It’s still happening, just not on your site.

Where Organic Top-Of-Funnel Traffic Actually Goes

Overview AI and other LLM-based answer engines now aggregate information from across the web to answer top search queries (TOFU) right on the search engine results page (SERP).

When a procurement manager searches for ‘leading CX vendors for mid-market SaaS,’ they increasingly encounter AI-generated summaries, recommendations, and curated results before reviewing traditional search lists. They get an integrated shortlist, with vendor summaries taken from all over the web: studies, reviews, analyst comments, editorial coverage. A buyer makes an observation, usually a near-final one, before visiting a seller’s website.

Why Some B2B Brands Are Still Getting Clicks

Seer Interactive’s 2026 AIO study, which included 5.47M queries and 2.43 billion organic impressions for 53 brands, found that brands appearing in AIO-present SERPs but not cited in the AI Overview saw their organic CTR drop by 67% in 2025 (Seer Interactive, 2026). Brands cited in AIO earned +120% more organic clicks per impression than competitors not cited in the same SERP. The gap between branded and unbranded brands, not global traffic breakdowns, is the variable.

Seer’s 2026 update also found the first signs of CTR stabilization in Q1 2026 after 18 months of decline. Rebates are available from the brands listed. Structural pressure on non-brands has not abated; it just stops being bad to the same extent.

Most B2B Research Now Happens Before You See a Lead

Generative AI has become the primary research method for buyers, returning a short list of sellers before they visit any website (Forrester, 18,000 buyers, 2026). 80% of the B2B buying journey now happens without the involvement of a salesperson. By the time the connection is made, the short list has been largely resolved.

Why A Smaller Pipeline Is Probably Better

AI models include signals of seller credibility such as case studies, third-party quotes, verified reviews, editorial coverage.

Through this process, AI models present marketers with a proven strong presence. Sellers with low credibility do not place lower in this area. They are completely bypassed in the research phase, before the consumer forms the intention to click.

The funnel has not disappeared. On top of it it has. What’s left is a filtered pipeline: incoming buyers have completed their seller research, driving the acquisition decision rather than the acquisition question.

How To Get Targeted With AI So The Right Buyers Can Find You

AI determines which sellers will appear in a buyer’s search before that buyer clicks. These five steps make sure you are one of them.

Step 1: Check Where You’re Coming (And Where You’re Not) (Weeks 1–2)

Pull data for your landing page.

Pull your top 50 landing pages into Google Search Console within the next 90 days. Record the group of questions, the type of question (informational/navigational/transactional), and the CTR for each. High impressions with low CTR on transactional queries is a credibility problem, not a visibility problem, and requires a different solution.

Create a list of third-party claims.

Use both Ahrefs and SEMrush; they return different datasets, so you need both. Extract, extract, and categorize each external mention: editorial, directory, review, analyst quote, forum, public. Calculate your average earned mentions (editorial, reviewer, verified review) to find out what you didn’t earn. For many B2B service companies, this ratio is worse than expected.

Benchmark against 3 competitors.

Create a gap table: publishing that mentions you but not you, review the forums where they are established and you are not there, the analyst reports that he mentions them. That gap list is your target list for steps 3 and 4.

Test your AI yourself.

Open ChatGPT, Claude, and Perplexity. Answer six to eight questions as a consumer would: “the best [service category] for [client type],” “compare [service] sellers.” Screenshot all the answer. Note whether it appears, how it is described, which competitors appear consistently, and which sources seem to shape the response. Repeat this test every quarter.

Step 2: Prepare Your Lessons So AI Can Really Learn (Weeks 2–8)

A reliability range test requires:

A named or specified client
Calculated baseline (“average hold time was 8:42 and CSAT was 61%,” not “they were struggling”)
A specific description of the work done includes key decisions made
Defined timeline
Results in absolute terms, not just percentages
A client quote for a specific outcome, not a general agreement
A named author with a professional profile linked

Most companies fail in factors 1, 2, 3, and 7. Unknown news studies with unclear results carry little weight with search algorithms or AI models.

Production process.

Find your five to ten strongest results from the last 24 months. Schedule 45-minute scheduled interviews with both your client contact and your internal delivery facilitator. Use a fixed template that forces the numbers: metrics before, metrics after, what changed and when. Assign a top rated writer to write each one, a real person with existing expertise, not a generic company name line. Get the client’s written output from the public citation metric. Publish with Schema.org markup and submit each URL to be indexed through Google Search Console instantly. Done right: three to four weeks per case study from interview to publication.

Step 3: Get Byline In Publications AI Trusts (60–90 Days)

Build your target list with real lines.

Build a media target list from real lines published in the last 90 days, not a PR database. By CX release, that means Customer Mindset, ICMI, Contact Center Pipeline, CX Today. Create a tracking spreadsheet with name, publications, recent topics covered, and angle notes that best fit their specific rhythm.

Write a three-part speech.

Each pitch has three sections: why this story fits this editor’s beat right now; what is the story in one sentence; what you offer (data, interviews, specials). Send individually. Follow up once in seven days. Expect a good 10–15% response. For five placements per quarter, plan to contact about 35–50 people.

Link all placements back to your site.

After every placement, add it to the media page on your site and link to the original. Cross-referencing strengthens the credibility signal in both directions.

Step 4: Get Updated on AI Forums Quotes (Weeks 4–6)

Prioritize AI-suggested platforms.

Prioritize the review forums that appeared in ChatGPT, AI Mode, Claude, or Perplexity quotes for your category during your Step 1 research. Assign review access to account managers, not sales; the request has more weight from the owner of the relationship. Send a personal email with a direct link to the submission form, not the homepage. There are no incentives: forum policies prohibit them and flagged reviews are removed.

Build review requests into your delivery system.

Expect 30–40% conversion on warm personal outreach. Build an application into your delivery plan within 90 days of participating and completing the project.

Respond to all reviews within 72 hours.

Including serious ones. A specific, thoughtful response to a negative review is itself a signal of credibility; it shows that the real person is responsible for the results.

Step 5: Get Your Writers Web-Convinced (Weeks 1–4)

Set up an ID track for each author.

For each team member who will be writing the content: update their LinkedIn profile with specific expertise and work history to be verified; create an author bio page on your website that links to their LinkedIn and describes their expertise in concrete terms; make sure all the content you produce links back to that bio page; and when the external placement is in the world, include a link to their company author page in the written line.

Why this is important for AI and search.

This creates an identity trail that can be verified across multiple web properties. A search engine or AI model that encounters a named person’s content can show who you are across your website, LinkedIn, and external publications, and interpret a consistent pattern as a true subject matter expert. Without this infrastructure, even solid content returns part of the potential credibility signal.

How to Link This AI Search Strategy

Running these workflows in parallel requires at least: a content strategist who knows how to conduct structured interviews and draft external publications; account management service to review access; senior subject matter expert available for media interviews and author briefings; and a project coordinator who manages client approvals across multiple studies simultaneously. For a company with existing content and PR skills: four to six months to a measurable move. Building from scratch: six to nine.