This article was sponsored by Expert callers. The opinions expressed in this article are those of the sponsor.
Why did my inbound traffic suddenly drop?
Are B2B leads decreasing everywhere?
Are AI Insights Hurting B2B Leads?
For two decades, the logic was simple: more traffic meant more leads, and more leads meant more revenue. In 2026, this logic collapses.
In this article, we’ll look at how this change could be a solution in disguise.
Why Inbound Traffic Volume Dropped (But Your Offers Got Bigger)
Generative AI has effectively absorbed the first research phase of the buying journey. This still happens, just not on your site.
Where your top-of-funnel organic traffic actually went
AI insights and other LLM-based answer engines now synthesize information from across the web to answer top-of-funnel (TOFU) questions directly on the search engine results page (SERP).
When a procurement manager searches for “best CX outsourcing providers for mid-market SaaS,” they increasingly come across summaries, recommendations, and AI-curated results before reviewing traditional search lists. They receive a synthesized shortlist, with vendor summaries from around the web: case studies, reviews, analyst mentions, editorial coverage. The buyer forms an opinion, often almost definitive, before even visiting a seller’s website.
Why some B2B brands still get clicks
Seer Interactive’s AIO 2026 study, covering 5.47 million queries and 2.43 billion organic impressions across 53 brands, found that brands appearing on AIO SERPs but not mentioned in the AI overview saw organic CTR drop 67% from 2025 (Seer Interactive, 2026). Brands cited in the AIO generated +120% more organic clicks per impression than their competitors not cited on the same SERP. It is the gap between cited and unmentioned brands, not a universal collapse in traffic, that constitutes the operational dynamic.
Seer’s 2026 update also revealed early signs of CTR stabilizing in Q1 2026 after 18 months of decline. The recovery goes to the brands mentioned. The structural pressure on unmentioned brands has not been reversed; the situation simply stopped getting worse at the same rate.
Most B2B research now happens before you see a lead
Generative AI has become the primary method for finding buyers, returning a shortlist of suppliers before they visit a website (Forrester, 18,000 Buyers, 2026). 80% of the B2B purchasing journey now takes place without the involvement of the supplier. By the time contact is made, the shortlist is largely established.
Why a smaller pipeline is probably better
AI models synthesize vendor credibility signals such as case studies, third-party citations, verified reviews, and editorial coverage.
Through this process, AI models surface suppliers with the strongest corroborated presence. Suppliers with low credibility are not ranked lower in this environment. They are bypassed entirely at the search stage, before a buyer intends to click.
The funnel has not disappeared. The top has. What remains is a filtered pipeline: buyers who arrive having already completed their supplier search, carry a purchasing decision rather than a discovery question.
How to get quoted by AI so the right buyers find you
The AI decides which suppliers appear in a buyer’s search before that buyer clicks. These five steps ensure you are one of them.
Step 1: Check where you are showing up (and where you are not showing up) (weeks 1-2)
Extract data from your landing page.
Pull your top 50 organic landing pages from Google Search Console over the last 90 days. Record the query cluster, query type (informational/navigational/transactional), and CTR for each. High impressions with low CTR on transactional queries are a credibility problem, not a viewability problem, and require a different solution.
Create an inventory of third-party mentions.
Use both Ahrefs and SEMrush; they return different data sets, so you need both. Export, deduplicate and classify each external mention: editorial, directory, review, analyst quote, forum, social. Calculate your ratio of earned (editorial, analyst, verified reviews) and unearned mentions. For most B2B service companies, this ratio is much worse than expected.
Benchmark against 3 competitors.
Build a chart of gaps: publications that cite them but not you, examine platforms where they are established and you are absent, analyst reports that name them. This list of gaps is your list of outreach targets for steps 3 and 4.
Manually audit your AI surface.
Open ChatGPT, Claude and Perplexity. Run six to eight queries like a buyer would: “best (service category) for (customer type),” “compare (service) providers.” Screenshot of each response. Note whether you appear, how you are described, which competitors appear regularly, and which sources seem to shape the response. Repeat this audit every quarter.
Step 2: Fix your case studies so the AI can actually read them (weeks 2-8)
A credibility-level case study requires:
- A named or specifically described customer
- A quantified benchmark (“the average processing time was 8:42 and the CSAT was 61%”, not “they had difficulty”)
- A specific description of the work performed, including key decisions made
- A defined schedule
- Results in absolute terms, not just percentages
- A customer quote on the specific outcome, not a generic approval
- A named author with a related professional profile
Most businesses fail on components 1, 2, 3, and 7. Anonymous case studies with vague results carry minimal weight with search algorithms or AI models.
Production process.
Identify your five to ten most significant results from the last 24 months. Schedule 45-minute structured interviews with the customer contact and your internal delivery manager. Use a fixed model that forces the numbers: metrics before, metrics after, what changed and when. Assign a named lead author to write each one, a real person with an existing professional presence, not a generic corporate byline. Obtain written approval from the client on public quoting of metrics. Publish with Schema.org markup and immediately submit each URL for indexing through Google Search Console. Well done: three to four weeks per case study, from interview to publication.
Step 3: Obtain signatures in AI Trusts publications (60 to 90 days)
Build your target list from real signatures.
Create a media target list from actual bylines posted in the last 90 days, not a PR database. For CX outsourcing, this means Customer Think, ICMI, Contact Center Pipeline, CX Today. Create a tracking spreadsheet with name, publication, recent topics covered, and a note on which angle is most relevant to their specific beat.
Write an argument in three paragraphs.
Each pitch includes three paragraphs: why this story currently fits this publisher’s beat; what is the story in one sentence; what you offer (data, interview, exclusivity). Send individually. Follow up once every seven days. Expect 10-15% positive responses. For five courses per quarter, plan for 35 to 50 individual awareness-raising contacts.
Link each location to your site.
After each placement, add it to a press page on your site and link to the original. Cross-references strengthen the credibility signal in both directions.
Step 4: Get reviews on AI Cites platforms (weeks 4-6)
Prioritize platforms that the AI actually cites.
Prioritize review platforms that appeared in ChatGPT, AI Mode, Claude, or Perplexity quotes for your category during your Stage 1 audit. Assign review delivery to account managers, not marketing; the request carries more weight on the part of the relationship owner. Send a personalized email with a direct link to the submission form, not a landing page. No incentives: platform policies prohibit them and flagged reviews are deleted.
Integrate review requests into your delivery process.
Expect 30-40% conversion through warm personal outreach. Integrate the request into your delivery process 90 days after engagement and at the end of the project.
Respond to each review within 72 hours.
Including the most critical. A specific, thoughtful response to a negative review is itself a signal of credibility; it demonstrates that a real person is responsible for the results.
Step 5: Make Your Authors Verifiable on the Web (Weeks 1-4)
Establish the identity trail of each author.
For each team member who will be writing content: update their LinkedIn profile with specific areas of expertise and verifiable career history; create an author bio page on your website that links to their LinkedIn and describes their major in concrete terms; ensure that all content they produce links to this bio page; and when external internships come in, include a link to their company’s author page in the byline.
Why it matters for AI and research.
This creates a verifiable identity trail across multiple web properties. A search engine or AI model encountering content from a named individual can cross-reference that identity across your website, LinkedIn, and external posts, and interpret the consistent pattern as true subject matter expertise. Without this infrastructure, even strong content returns only a fraction of its potential credibility signal.
How to coordinate this AI research strategy
Managing these workflows in parallel requires at a minimum: a content strategist capable of conducting structured interviews and writing external publications; an account management resource for disseminating reviews; a senior subject matter expert available for media interviews and author attribution; and a project coordinator managing client approvals on multiple case studies simultaneously. In a well-resourced company with existing content and PR capacity: four to six months for measurable movement. Building from scratch: six to nine.
SEO rewarded visibility. It now rewards credibility. And only one of these compounds.
In the age of AI search, businesses with the strongest digital credibility may attract fewer visitors, but they will increasingly attract the right ones.
Image credits
Featured image: Image from Shutterstock. Used with permission.
