SEO for B2B services firms

Search still pulls the highest-intent buyers, but the shape of the channel is changing. Ranking on Google now matters because it is the doorway to AI citations. Both feed the same pipeline, and they share most of the work underneath.

Written by Peter Korpak Chief Analyst at 100Signals Updated
89%

of software development agencies scanned position for 3+ verticals — only 4% get cited by AI in any of them.

Source: 100Signals scan of 1,700+ agencies across 30 verticals, Q1 2026.

What this is

SEO for B2B services firms is the practice of making your expertise discoverable to buyers who already know they have a problem.

That means ranking for commercial-intent queries in Google, getting cited by ChatGPT and Perplexity when prospects ask for recommendations, and building the structured content, technical foundations, and domain authority that feed both channels simultaneously. Done well it compounds for 12-24 months; done generic, it produces traffic that never books a meeting.

How to think about it
Target queries
Commercial-intent and comparison queries — "best X agency for Y", "X for Y companies", "Z alternatives" — not informational "what is" pages. Informational queries get answered by AI Overviews without a click; the ones that still drive pipeline are the queries a buyer runs when they already know what they need and are deciding who to hire. That distinction should govern the entire keyword strategy.
Primary channel
Google organic for direct clicks; AI answer engines (ChatGPT, Perplexity, Claude, Gemini) for cited recommendations. Same content, two retrieval systems. Pages optimised for Google's quality signals — structured data, author attribution, original data — also score better on AI retrieval. The work overlaps by roughly 70%; the remaining 30% is entity hygiene and third-party mentions that the models weight higher than Google does.
Time to result
Technical wins land in weeks. Content-led rankings compound over 3-6 months. AI citations can show up 4-8 weeks after publication once retrieval indexes refresh. Measuring at month one misses most of the curve — the realistic assessment window is month 6 for organic pipeline and month 3 for AI citation share. Anything flat past month 4 is usually a niche specificity problem, not an SEO execution problem.
Core dependency
Niche specificity. Broad positioning ("we do everything for everyone") ranks nowhere and gets cited nowhere — regardless of content volume. The query cluster a firm can realistically own is defined by how narrow its stated expertise is. A software agency claiming ten verticals fights ten times as many competitors for ten times as few queries. An agency claiming one vertical owns a much smaller query set and wins it completely.
Measurement
Pipeline attributed to organic and AI-assistant referrals — not rank reports. Keyword rankings are a leading indicator, not the outcome. The reporting that matters connects a closed deal back through the pages the buyer read before booking the first call. That tracing, plus AI citation share across 50-100 monitored queries, gives a picture rank reports can never provide.
Common failure
Treating SEO as a content factory instead of an authority system. Volume without expertise ranks briefly, then gets absorbed into AI Overviews. The pattern: a firm publishes 30 blog posts in a quarter, sees a brief traffic lift, watches AI Overviews claim the snippet, and calls SEO dead. The real diagnosis is that none of the 30 posts contained anything a buyer could not find on a dozen other sites. Expertise that cannot be replicated by asking an AI assistant is what earns the citation and the click.
The framework

The Authority Flywheel

  1. Define the niche

    One vertical, one buyer type. Broad positioning has no query cluster to rank for. The decision feels like restriction; it functions like concentration — every piece of content, every link, and every citation compounds toward the same authority signal rather than scattering across twelve disconnected topics.

  2. Build the topical core

    Pillar hub, 6-10 spokes, one unified entity — not disconnected posts competing for attention. The pillar earns the link equity; the spokes answer the surrounding questions that signal to the model that the firm understands the full problem, not just the headline keyword.

  3. Prove expertise on-page

    Case studies with real numbers, architectural decisions, post-mortems. E-E-A-T signals the AI models already read. A case study that names the client, the before state, the specific intervention, and the measured result carries more retrieval weight than five generic guides on the same topic.

  4. Fix the technical floor

    Core Web Vitals, structured data, crawl efficiency. Ranking ceiling is whatever the slowest page allows. LLM crawlability adds a second technical layer: an llms.txt file, consistent canonical URLs, and author schema that matches the linked LinkedIn and About page all matter for AI retrieval in ways that traditional SEO audits miss.

  5. Earn citations

    Digital PR and third-party mentions. AI citation systems weight what other sites say about you more than what you say about yourself. A mention in one relevant trade publication often outweighs twenty self-published pages on the same topic, because the model treats third-party mention as independent confirmation rather than self-promotion.

  6. Monitor AI discovery

    Track what ChatGPT and Perplexity cite for your queries. Refresh pages where the citation pool has shifted. The retrieval pool is not static — pages that owned citations in Q1 can drop out by Q3 if competitors publish more specific content. Quarterly probes replace quarterly rank reports as the primary signal.

SEO vs adjacent visibility services — when to use which
SEO AI Visibility Digital PR
Primary channel Google organic search — clicks from buyers who searched the query ChatGPT, Perplexity, Claude, Gemini — named in AI answers Third-party publications, podcasts, and analyst reports
Output Ranking pages that capture commercial-intent traffic and compound over 12+ months Named citations inside AI-generated answers when buyers ask for recommendations Backlinks, branded mentions, and off-site authority signals
Time to result Technical fixes in weeks; content compounds over 3-6 months 4-8 weeks after publication for Perplexity; 8-12 weeks for ChatGPT 2-4 months per earned placement once relationships exist
Depends on Niche specificity + original proof content + technical foundation SEO eligibility + entity-consistent schema + third-party mentions Genuine operator expertise and an insight editors want to publish
When to lead with it Buyers are already searching commercial-intent queries in your niche Buyers in your space ask AI assistants for agency recommendations You need off-site credibility signals to unlock rankings and citations
8 guides · 6 lists

SEO by firm type

Or browse by firm type
For Software Development Agencies Your homepage lists eight verticals. For IT Companies Your longest-tenured client put the renewal out to three bids. For Consulting Firms Your white paper got read. For Managed Service Providers Referrals got you to $3M. For AI Consultancies Buyers cannot tell you apart from the dev shop that retitled "custom software" as "AI services. For Design Agencies Your best work of the last 18 months is under NDA. For Web Development Agencies Claude Code, Lovable, and Wix Harmony are eating your $5k to $50k tier. For Cybersecurity Companies Every competitor is running the same fear-based messaging.
Related services
Content Marketing Most B2B content is written for search engines that no longer click through. Digital PR Trade publication coverage is no longer an ego exercise. AI Visibility Buyers now ask AI assistants for agency shortlists before they ever land on a website.
Written by
Peter Korpak, Founder of 100Signals

Peter Korpak

Founder, 100Signals

Ex-Head of Marketing at Brainhub, an FT 1000 Fastest-Growing Company in Europe in 2021 and 2022. Former analyst at Credit Suisse and Aviva Investors. Eight years building pipeline for B2B services firms, 300+ outbound campaigns across 15+ agencies, top programs landing 40%+ positive reply rate. Writes about positioning, lead generation, and AI visibility for agency operators.

FAQ
Is SEO still worth the investment for a B2B services firm in 2026?
Yes — the thesis just shifted. Informational traffic is declining because AI Overviews absorb it. Commercial-intent queries still drive pipeline, and the content that ranks for them now doubles as the retrieval source for ChatGPT and Perplexity. The same work, two compounding channels. The firms writing SEO off are usually the ones that ran it as a content calendar rather than an authority system. Volume was never the mechanism — expertise was, and expertise still earns both the click and the citation.
How long until SEO produces pipeline?
Technical fixes move rankings in weeks. New content-led pages typically take 3-6 months to settle. AI citations can appear within 4-8 weeks of publication once the indexes refresh. Treat anything under 90 days as setup; judge pipeline impact at month 6. The firms that abandon the channel at month three have consistently been the ones who would have seen compounding at month seven. Set the expectation before the program starts — it makes the flat months survivable.
What should we prioritise — SEO or AI visibility?
They are the same investment now. Pages that rank on Google get crawled by AI retrieval systems. Structured data that Google parses also helps LLMs extract your expertise. Fix SEO fundamentals first; AI eligibility follows. The one area where they diverge is off-site validation: third-party mentions and external citations carry disproportionate weight for AI retrieval relative to their SEO value, so a digital PR program that would be optional in a pure-SEO strategy becomes mandatory when AI citations are the target.
How much of the SEO stack can we automate?
Roughly 60-70% of tactical work — keyword clustering, content outlines, technical audits, internal linking, reporting. What cannot be automated is what differentiates you: original case studies, technical proof, and the strategic judgement about which niches are defensible. An AI-generated outline for a fintech dev agency case study produces the same outline as every competitor's. The case study that earns citations is the one where someone wrote down what actually happened, with numbers and named decisions. That part requires a human who was in the room.
Which keywords should we target?
Stop targeting "software development company". Target niche commercial intent: "fintech software development agency", "HIPAA-compliant healthcare API partner", "legacy Java migration consulting". Lower volume, dramatically higher conversion, far less competition. The buyer searching the niche phrase already knows what they want — the conversion rate from niche commercial query to booked meeting is typically 5-10x the rate from a broad generic query. Lower volume is not a problem when every visitor is a real prospect.
What is the difference between SEO for services firms versus SaaS companies?
Services firms sell trust and expertise across a long cycle; SaaS firms sell a product with a fast trial. The content strategies that follow are completely different. SaaS can rank on category education and volume; services firms need proof content — case studies, architectural decision records, named-client outcomes. SaaS can afford to optimise for traffic; services firms need to optimise for qualified intent, which means fewer pages, more depth, and a much tighter query cluster. Copying a SaaS content strategy onto a services firm consistently produces traffic that never books.

Stop buying seo in isolation.

SEO works when it's one coordinated channel inside a pipeline, not a standalone purchase. Book a call to see how it fits, or run the free scan to see where you stand first.

Book a Pipeline call

Or see where you stand first:

Free. Results in 24 hours.