Section 1
The five challenges at a glance
The shift from generic to branded search advantage creates five distinct problems for service firms, summarized in the table below. A framing note first, because this argument is often made sloppily: 'brand search is the new SEO' does not mean technical SEO is dead or that category content is worthless. It means the economic returns of search have redistributed, away from ranking for generic informational queries, whose clicks AI Overviews have measurably absorbed, and toward owning the demand that arrives pre-sold, typing your name. The theoretical foundation comes from the Ehrenberg-Bass Institute's research program: brands grow by reaching all category buyers and building mental availability, the propensity to be noticed and thought of in buying situations, through distinctive assets and consistent presence, rather than through persuasion-led differentiation (Sharp, How Brands Grow, 2010). Attribution accuracy matters here: Ehrenberg-Bass describes mental availability as memory structures linked to buying situations, not mere awareness, and its researchers are themselves prominent critics of some popular budget heuristics, including Binet and Field's 60/40 guideline, a live methodological debate worth knowing before quoting either camp as settled law. The practical evidence that branded demand survives AI disintermediation, meanwhile, comes from click-pattern research on the sites still winning Google traffic.
Section 2
Challenge one: branded queries are the demand AI cannot intercept
The structural argument is simple: a generative system can answer 'how to fix agency utilization rates' from a hundred sources, but it cannot answer 'BizGrowthAxel pricing' without referencing the named firm. The empirical pattern matches. Seer Interactive's Search Console analysis across 25.1 million impressions found organic click-through on AI Overview queries collapsing 61%, with informational queries hit hardest (Seer Interactive, 2025). Yet the categories of search that still reliably send clicks, documented in Cyrus Shepard's analysis of sites that grew Google traffic through the decline, and echoed in SparkToro's 2026 study, are branded searches, local queries, and high-intent transactional or tactical queries (Zyppy, 2025; SparkToro/Similarweb, 2026). Pew's behavioral data supplies the mechanism: AI summaries satisfy informational intent in place, so the remaining click motivation is specific, the user wants a particular source, not a synthesis (Pew, 2025). Google's own product decisions corroborate the shift: in November 2025 it shipped a branded-queries filter in Search Console, formalizing the branded-versus-generic distinction in its core measurement tool (Google, 2025). For service firms the strategic arithmetic follows: a buyer who searches your name converts at rates generic traffic never approached, arrives pre-sold by whatever built the memory, and is invisible to competitors bidding on category terms. The scarce input is no longer rankings, it is being the name the buyer remembers when the need matures. That input is manufactured upstream of the search box.
Section 3
Challenge two: mental availability is the engine of branded demand
Why does one firm get name-searched while an equally competent rival waits for referrals? The best-evidenced answer comes from the Ehrenberg-Bass Institute's research program. In How Brands Grow, Byron Sharp synthesizes decades of buyer-behavior data into a two-factor model: brands grow by building mental availability, the propensity for the brand to be noticed or come to mind in buying situations, and physical availability, being easy to find and buy (Sharp, 2010). Three findings matter most for service firms. First, mental availability is not generic awareness; it is a network of memory associations linking the brand to the situations in which buying occurs, what the institute calls category entry points, such as 'our delivery is chaotic' or 'we need to raise prices.' The more entry points a brand attaches to, the more often it surfaces unprompted. Second, distinctiveness beats differentiation: Sharp's data shows brands seldom compete on meaningfully perceived difference; they compete on being recognized and recalled, which makes consistent distinctive assets, name, visual identity, recurring phrases, a founder's voice, the working capital of memory (Ehrenberg-Bass). Third, reach beats loyalty targeting: growth comes disproportionately from light and non-buyers, which in B2B terms means the 95% of buyers currently out-of-market (Edelman-LinkedIn, 2024). The synthesis with search data is direct: mental availability is the upstream variable, and branded search volume is its downstream, measurable trace. Firms that publish consistently against buyer category-entry points are, in Sharp's terms, building the memory structures that AI-era search converts into clicks.
Section 4
Challenge three: budgets and dashboards still point at the old game
The third challenge is allocative. Most 5-7 figure service firms spend nearly everything on activation, ads, outbound, conversion content, because those channels produce attributable, immediate results, while brand building pays on a lag through mechanisms last-click dashboards cannot see. The best-known evidence for rebalancing is Binet and Field's analysis of roughly 1,000 IPA effectiveness cases, which found campaigns weighted around 60% brand building to 40% activation delivered the strongest long-term effects, a guideline, its authors stress, that flexes by category and brand size, and one that Ehrenberg-Bass researchers have publicly contested on methodological grounds (Binet & Field, IPA; Ehrenberg-Bass). Operators need not adjudicate the academic dispute to extract the robust core: both camps agree that memory-building activity with broad reach drives long-run growth, and that firms allocating zero to it are mispriced for the AI-search era, where memory is the only demand that survives disintermediation. The measurement layer compounds the bias. Branded search volume, the cleanest available proxy for B2B mental availability, went largely untracked until Google's Search Console branded filter shipped in late 2025 (Google, 2025). Direct traffic, substantially memory-driven visitation, is routinely dismissed as unattributable noise rather than read as a brand indicator. And 'dark' demand sources like AI assistant recommendations leave no referrer at all. McKinsey's analysis of AI-era search reaches the same conclusion from the buyer side: discovery is dispersing across AI surfaces while conversion concentrates on brands buyers already know and name (McKinsey, 2025). The dashboard, not the market, is what tells most firms brand is unaffordable.
Section 5
Innovative solutions
Advanced operators are translating brand science into service-firm practice through five moves. First, category-entry-point mapping: interviewing clients to enumerate the fifteen to twenty situations that trigger purchase, 'we missed a quarter,' 'a key hire quit,' 'referrals dried up', then assigning content, talks, and founder posts to each, so memory is built against buying situations rather than topics (method per Ehrenberg-Bass research). Second, distinctive-asset codification: a one-page register of the firm's name conventions, frameworks with proprietary names, visual identity, and recurring phrases, enforced across every surface, because consistency is what compounds recognition into recall (Sharp, 2010). Third, founder-as-asset: in expertise categories the founder's face, voice, and named point of view are the most distinctive assets available, and zero-click platforms distribute them free; the newsletter and email list then capture the most-convinced fraction. Fourth, branded-demand instrumentation: Search Console's branded filter for query trends, direct-traffic trendlines, share of 'heard about you from' mentions, and AI-assistant spot checks, asking the major assistants who they recommend for your category, reviewed quarterly as the brand P&L (Google, 2025). Fifth, deliberate reweighting: moving toward a defensible brand-to-activation split on the Binet-Field evidence, flagged honestly as a guideline under debate, with the brand budget spent on reach among category buyers rather than retargeting the already-convinced. None of this requires consumer-brand budgets; it requires consistency that most competitors cannot sustain.
Section 6
Solution framework
The Memorability Operating System runs on four disciplines. Discipline one, Codify: define the memory you intend to own, one category, three named positions, a register of distinctive assets including the founder's voice and any proprietary frameworks. The test is whether a client could repeat your positioning to a peer after thirty days; if not, it is not yet codified. Discipline two, Repeat: publish against the mapped category entry points on a fixed cadence across the email list and two rented channels, holding assets ruthlessly consistent, the Ehrenberg-Bass evidence is unambiguous that memory structures build through consistent, repeated exposure, and that rebrands and message churn liquidate accumulated availability (Sharp, 2010). Discipline three, Distribute: place the codified message where category buyers already pay attention, podcasts, partner newsletters, communities, events, prioritizing reach across the buyer population over depth with existing fans, per the light-buyer growth evidence. Discipline four, Measure: the branded-demand panel, branded query impressions and clicks via Search Console's filter, direct traffic, self-reported attribution share, and quarterly AI-assistant recommendation checks, reviewed against publishing and distribution activity (Google, 2025). The framework's promise matches the evidence rather than exceeding it: mental availability builds slowly, compounds durably, and converts into the one search demand stream, your name, that no algorithm update, AI Overview, or competitor bid can intercept. For a service firm, that is the closest thing search strategy now offers to a moat.
Section 7
Evidence-based action plan
Days 1-15: baseline branded demand. Use Search Console's branded-queries filter to capture twelve months of branded impressions and clicks, pull direct-traffic trends, and tally 'how did you hear about us' responses; ask three major AI assistants who they recommend in your category and record the answers (Google, 2025). Days 16-30: run category-entry-point interviews with five to eight clients, producing a mapped list of buying triggers and the words clients actually use. Draft the distinctive-asset register: name conventions, framework names, visual identity, recurring phrases, founder positioning. Days 31-60: ship the consistency layer. Align every public surface, site, profiles, email signatures, proposal documents, to the register; begin publishing one entry-point-targeted piece weekly through the newsletter and the founder's primary rented channel, with identical positioning language everywhere. Days 61-90: extend reach. Book two appearances on podcasts or partner newsletters your buyers already consume; review the branded-demand panel against activity. Quarterly thereafter: track branded impressions, direct traffic, attribution share, and AI-recommendation presence, expecting visible movement in two to four quarters, the lag the brand-building evidence predicts (Binet & Field, IPA). The strategic summary for the AI-search era: rankings for generic queries are a depreciating asset, citations are a contested asset, but a buyer who types your name is demand you own. Build the memory; the searches follow. For adjacent evidence in this pillar, see [Being the Cited Source: The GEO Evidence Playbook for AI Citations](/blog/growth-geo-cited-source-ai-citations-playbook) and [Original Research as a Demand Engine: The Evidence on Data-Driven Content](/blog/growth-original-research-demand-engine).