Business Directories in the Age of Zero-Click Search: Adaptation or Obsolescence?

Last spring, I got a call from a CEO who ran a regional services directory covering three counties in the Midlands. He’d seen organic traffic drop 40% over fourteen months, and his board was asking whether they should wind the business down or pivot hard. I’d worked with him briefly in 2019 on his local SEO stack, which is probably why he phoned me instead of one of the London agencies that would have charged him triple what I did and given him worse advice.

What follows is roughly how that engagement went. I’ve changed some numbers and details, but the decision-making is real. If you run a directory, work for one, or you’re a business owner trying to figure out whether to still care about them — this walkthrough is for you.

Meeting the Regional Directory Client

The 40% traffic drop context

The directory had been profitable for nine years. Not glamorous, but profitable — the sort of business that pays three salaries, keeps the owner’s mortgage covered, and generates enough surplus to reinvest in a modest content team. Then, starting roughly at the beginning of 2024, traffic began bleeding out. Not a cliff edge; more like a slow puncture. Down 8% one quarter, then 12%, then another 14%.

When I pulled their Search Console data, the pattern was obvious within about ten minutes. Impressions were actually up by 22%. Clicks were down 40%. That’s the zero-click fingerprint — Google was surfacing their content more than ever, but users were getting what they needed from the SERP itself and never visiting the site.

The CEO thought he’d been penalised. He hadn’t. He’d been absorbed.

Their existing revenue model

Here’s where it got uncomfortable. Their monetisation was roughly:

• 68% pay-per-listing (tiered: £9/month basic, £29/month featured, £89/month premium)
• 22% display advertising (programmatic plus direct)
• 10% sponsored content and lead forms

Every single one of those revenue streams depended on pageviews. Listings were sold on the promise of exposure, ads paid per impression, and lead forms only fired if someone actually clicked through to a detail page. When traffic dropped 40%, ad revenue dropped 44% (small pricing lag), and — this was the painful one — listing renewals dropped from an 81% rate to 58% within two quarters. Businesses were looking at their analytics dashboards and seeing the referrals dry up.

Why panic wasn’t the answer

The CEO wanted to fire his content editor and double down on publishing volume. Twelve new articles a week instead of five. Classic panic response — if the machine is producing less output, feed it more input.

I told him no. Not because content volume never works, but because the problem wasn’t content deficit. The problem was that Google had decided his directory’s value proposition could be served on the SERP itself. Pumping out more articles for Google to cannibalise would just accelerate the bleed and burn cash he needed for the actual fix.

Diagnosing What Zero-Click Actually Broke

Pulling the SERP feature audit

First thing I did — before we touched schema, before we drafted a single strategy doc — was audit the top 500 queries the directory ranked for. I wanted to know exactly which SERP features were eating their clicks. We used Semrush for the feature breakdown and cross-referenced with manual spot-checks because automated tools lie about “People Also Ask” frequency.

The breakdown came back roughly:

• Local Pack / Map Pack eating 31% of their top queries
• Featured snippets pulling from their own content on 18%
• “People Also Ask” expansions providing answers on 44%
• Knowledge Panels for competitor businesses on 23%
• “Things to do / Places” carousels on 9%

(Percentages exceed 100 because queries often triggered multiple features.)

The featured snippets finding was the one that stopped me. Google was scraping their own listing descriptions, displaying them in the snippet box, and users were reading the answer and moving on. The directory had essentially trained Google to replace them.

Queries we lost vs. queries we still owned

Not every query was a loss. When I segmented by intent, a clear pattern emerged:

This table basically wrote our strategy. We stopped trying to compete for “electricians near me” — that war was lost the day Google launched the local pack — and started investing everything in the two categories still showing meaningful CTR: comparison content and problem-solving long tail.

The listing-detail pages still converting

Here’s the thing that saved the business. When I dug into the detail pages — the individual business profile URLs — their conversion metrics were actually better than in 2023. Fewer visitors, but those who arrived converted at 7.2% into lead-form submissions, up from 4.1%.

Why? Because the users who still clicked through were the ones with real commercial intent. Zero-click had filtered out the tyre-kickers. What arrived at the detail page was pre-qualified traffic.

This changed everything. If fewer-but-better was the new reality, the monetisation model needed to charge for quality of lead, not quantity of impression.

The Structured Data Pivot Decision

Why we bet on schema over content volume

Here’s the unfashionable opinion I gave the CEO over our third coffee: stop writing articles. Or rather, write half as many and spend the saved budget on structured data engineering.

My logic: if Google is going to answer queries on the SERP, then the question becomes whose information does Google use? The answer is: whoever feeds it the cleanest, most machine-readable, entity-linked data. Content volume was yesterday’s war. Structured data was today’s.

We reallocated their £14,000 monthly content budget like this:

• £5,000 to a schema engineer on contract (three days a week for four months)
• £4,000 to reduced but deeper content output (three pieces a week, each 2,000+ words, heavily fact-checked)
• £3,000 to review acquisition and verification tooling
• £2,000 retained for outreach and link-building on the comparison content

Choosing LocalBusiness over Organization markup

This sounds like a technicality; it’s not. It’s one of the most important decisions in the project.

The directory’s existing markup was generic Organization schema on every listing. Technically valid, but Google treats Organization as a weak signal for local intent. Switching each of their 11,400 active listings to LocalBusiness (with proper subtype — Plumber, Electrician, Dentist, etc.) unlocked a completely different set of SERP features.

The migration took six weeks. Not because it was technically hard — it was mostly a template change plus data enrichment — but because we had to manually classify listings that had ambiguous categories. A company listed as “property services” could be a letting agent, a surveyor, a maintenance firm, or a solicitor. Each maps to a different schema subtype, and getting it wrong gives you worse results than generic markup.

The FAQ schema experiment that backfired

I’ll own this one. I told the CEO we should deploy FAQ schema on every category page. Seemed obvious — FAQ-rich results were still showing in SERPs, click-through rates were reasonable, and it was quick to implement.

Three months in, Google announced it was restricting FAQ rich results to government and health sites. Overnight, we lost the SERP real estate we’d invested in. Worse, the FAQ blocks we’d written were now cluttering pages without earning visibility.

We kept the content (it genuinely helped users) but stripped the markup. Total wasted effort: roughly £4,800 in dev time and about a month of content production. Not catastrophic, but a reminder that betting on specific SERP features is riskier than betting on underlying entity recognition, which Google has shown no signs of deprecating.

Rebuilding for Entity Recognition

Feeding Google’s Knowledge Graph directly

If you accept that zero-click is permanent, then the goal shifts. You’re no longer trying to drag users to your website. You’re trying to become the source Google cites when it answers their question on the SERP. That’s a different game entirely, and it starts with making your entities legible.

We spent two months systematically ensuring every listing had:

• Consistent NAP data (Name, Address, Phone) matching Companies House records where applicable
• sameAs properties linking to the business’s social profiles, Companies House page, and — critically — any existing Wikidata entry
• Precise geo-coordinates rather than address-derived ones
• Proper opening hours in schema, not just as display text
• Verified founding dates and ownership where discoverable

Tedious. Utterly worth it.

Wikidata and Wikipedia citation strategy

This is where we did something most directories don’t bother with. We identified the 340 largest, most notable businesses in the directory and checked their Wikidata presence. About 60 had entries. Another 90 were notable enough to warrant one but didn’t have entries yet. For those 90, we created properly sourced Wikidata entries over three months — not as spam, but as genuine reference entries citing multiple sources including trade publications, Companies House filings, and local news coverage.

Why? Because Wikidata is one of the primary inputs to Google’s Knowledge Graph. When your directory becomes a frequently cited source on Wikidata entries, Google starts treating you as an authority on those entities. This is slower and less direct than link-building, but the effect compounds.

We also added our directory as a citation source on 23 existing Wikipedia articles about notable local businesses, being careful to only do so where we had genuinely unique, verifiable information not cited elsewhere. Fourteen of those citations stuck. Nine were reverted by editors within a week, which is honestly a better hit rate than I expected given Wikipedia’s allergy to directory-type sources.

Measuring brand query lift at 90 days

Ninety days after the entity work concluded, branded search volume for the directory was up 31%. Branded queries are a reasonable proxy for “Google sees you as a recognisable entity” — they’re the queries where someone types your directory name directly, often after encountering it somewhere Google surfaced.

More importantly, we started appearing as a source citation inside Google’s AI Overview results for about 12% of the category queries we tracked. That wasn’t a traffic win (most of those impressions didn’t convert to clicks) but it was a visibility win that mattered for the new business model we were about to build.

The Monetization Rewrite

Killing pay-per-listing for lead routing

This was the scary one. The CEO had to phone his biggest accounts — businesses paying £89/month for premium placement — and tell them the pricing model was changing. Some would leave. The question was whether the new model would earn more from the remaining customers than the old model earned from everyone.

The new structure:

• Free tier: Basic listing with NAP, hours, and one category. Indexed, schema’d, citation-grade.
• Verified tier: £19/month. Verified badge, photo galleries, review responses, richer schema markup, priority in internal search.
• Lead-routed tier: Pay per qualified lead. £12-£45 per lead depending on category and lead type (form submission, phone call tracked via CallRail, or booking).

We killed the middle tiers entirely. Simpler to sell, easier to justify, and aligned incentives properly — the directory now only made serious money when its partners made money.

Introducing verified-badge subscriptions

The verified badge wasn’t cosmetic. To earn it, businesses had to submit proof of trading address, a companies registration reference (for limited companies), and undergo a phone verification call. Fifteen minutes of work for them, about four minutes of work for us using a part-time VA.

Uptake was slower than we’d hoped — 18% of eligible listings subscribed in the first four months versus our 30% target. But here’s what surprised me: the verified businesses generated 3.4x more leads per listing than unverified ones, because users clearly trusted them more. That shifted the conversation from “why should I pay £19?” to “why are you leaving money on the table by not being verified?”

Lead attribution when clicks disappear

This was the thorniest technical problem. If users were getting business phone numbers from Google’s Knowledge Panel without ever visiting our directory, how could we attribute leads to our influence?

Two mechanisms, neither perfect:

First, we issued every verified business a unique trackable phone number (via CallRail) that they had to use as their primary listing number across all their citations — not just ours. Most agreed, because they wanted the call analytics anyway. When Google pulled their phone number into a Knowledge Panel, it pulled our tracked number. Calls got logged, we could attribute, and we could charge per call on the lead-routed tier.

Second, we required verified businesses to tag our directory as a source in their Google Business Profile’s “from the business” section. Not all did. Enough did.

Results After Eight Months

Revenue recovery at 127% of baseline

Eight months after we started — ten from the initial phone call — monthly recurring revenue hit 127% of the pre-decline baseline. Not a typo. Revenue was higher than it had been before the traffic collapsed.

The breakdown:

• Verified subscriptions: £24,800/month (new)
• Lead-routed fees: £41,200/month (new, highly variable)
• Remaining legacy listings: £8,600/month (declining, being phased out)
• Display/sponsored: £6,400/month (down from £19,000, we didn’t fix this)

We lost 34% of the pre-change advertiser base. We gained a revenue model that didn’t depend on traffic volume.

Traffic stayed down, that was fine

Organic traffic at month eight was actually still 36% below the 2023 peak. We barely clawed any of it back. And for the first time in my career, I didn’t care, because the correlation between traffic and revenue had been deliberately broken.

This is the bit that took the CEO six months to internalise. He kept looking at the traffic dashboard and wincing. I kept redirecting him to the revenue dashboard. Eventually he had his data team build a custom view that just… didn’t show organic traffic anymore. Honest to god, that was one of the most useful interventions we made. Stop measuring what no longer matters.

What the CEO didn’t expect

Two things caught him off-guard.

First, the businesses on the verified tier became an informal sales force. They told other businesses about the directory because they were actually getting leads. Referral signups accounted for 28% of new verified subscriptions by month seven.

Second, Google’s AI Overview results started citing the directory’s comparison content surprisingly often — about 19% of comparison queries in their vertical. That’s visibility we hadn’t planned for, but the entity recognition work had positioned us well. We now get mentioned in AI summaries that competitors don’t, which matters because those summaries are the first touchpoint for users doing local business discovery.

Adapting This Playbook to Your Constraints

The Midlands case had a particular shape: regional generalist directory, nine years of domain authority, budget north of ten grand a month, a CEO willing to make hard decisions. Most directories don’t have all of that. Let’s talk about what changes.

If you’re running a niche B2B directory

Good news: you’re probably more insulated from zero-click than the regional generalist was. B2B queries tend to be longer-tail, lower-volume, and less likely to trigger Local Pack or Knowledge Panel results. Your users are researchers, not consumers with immediate intent.

Bad news: your audience is also smaller, so the margin for error is thinner and you probably can’t support a full schema engineering project on contract.

What to prioritise instead:

• Deep, specification-heavy listing pages. B2B buyers want feature comparisons, integration details, compliance certifications. Content volume matters more here than in B2C directories because AI Overviews struggle to summarise technical specifications accurately.
• Gated content that captures email addresses. Lead generation to your directory’s own database is the real monetisation play. Charge listed businesses for access to that database.
• Industry-specific schema where available (there’s decent schema for software, for example, and growing coverage for industrial equipment).
• Integration with trade associations and professional bodies. Their citations carry disproportionate weight in B2B vertical searches.

I did a smaller piece of work with a manufacturing components directory last year where we basically skipped Google entirely and built their monetisation around email newsletters and a closed buyer network. Worked better than the SEO-first version. Sometimes the answer is to stop competing on the SERP.

The $5K budget version

If you’ve got five grand total — not monthly, total — to pivot an ailing directory, here’s what I’d do in that order:

Week one: Audit your top 100 queries for SERP features. Free tools (Search Console plus manual SERP checks) get you 80% of the insight. Identify which queries are lost, which are defendable. Kill content plans targeting lost queries.

Weeks two to four: Migrate your listing pages to proper LocalBusiness schema with correct subtypes. If you’re on WordPress, plugins like Schema Pro or RankMath Pro can do this for under £200/year. If you’re on a custom stack, budget £1,500-£2,500 for a freelance developer to do it once, properly.

Month two: Implement call tracking on at least your top 50 listings. CallRail, CallTrackingMetrics, or WhatConverts all work. £60-£200/month depending on volume.

Month three: Introduce a single paid tier that includes verification + call tracking. Price it at £15-£25/month. Phone your top 20 existing customers personally. Don’t email. Phone.

Months four onward: Use remaining budget on review acquisition and NAP consistency checks. Tools like BrightLocal’s citation resources are useful here for identifying where your data is inconsistent across the wider ecosystem.

Total cash outlay: roughly £3,800-£4,800 depending on your stack. The rest goes into buffer for mistakes, because you will make them.

When obsolescence is actually the right call

Not every directory should survive. I’ll say that plainly because too many consultants won’t.

If you’re running a thin, general directory with no real differentiation, no local knowledge, no verification process, and your only value proposition was “we aggregate listings” — your business model is genuinely obsolete. Google aggregates listings better than you ever will. So do Yelp, Yell, Bing Places, and half a dozen platforms with more capital than you’ll raise in a decade. Competing on aggregation alone is like running a video rental shop in 2015. You’re not going to adapt your way out of that.

Signs it’s time to wind down rather than pivot:

• Your domain has fewer than 200 referring domains after five years of operation
• You have no category or geographic specialisation that couldn’t be replicated in six months
• Your listings are 90%+ unverified and you have no mechanism to verify them at scale
• Your CEO’s mental model is still “we need more traffic” after repeated conversations
• Revenue is below £6,000/month and has been declining for over a year

In those cases, the kindest advice is: sell the domain to someone consolidating the space (and industry consolidation is very much happening, which creates acquisition opportunities), harvest the customer list, and put your energy into something with a fighting chance. Pivoting a weak directory is almost always more expensive than starting fresh.

What This Means Going Forward

The directory I worked with is still operating, still profitable, still growing. The CEO has stopped watching the traffic dashboard. We recently had a conversation about whether to launch a second directory in an adjacent vertical, and the whole conversation happened in revenue terms — cost per verified listing acquired, lifetime value per lead-routed customer, expected payback period on schema engineering. Not a single mention of organic sessions.

That’s the mental shift the zero-click era demands. If your directory business is still priced, sold, and measured on traffic, you’re operating a business model that the search engines have quietly deprecated. The ones surviving — and, quietly, thriving — are the ones that rebuilt around verified trust, lead attribution, and entity-level presence in the Knowledge Graph. The SERP became the marketplace. Your directory’s job is to be the most credible supplier on it, not to drag users off it.

If you’re starting that rebuild now, you’re not late. The directories that haven’t figured this out yet are the ones whose listings you’ll be acquiring in 2027. Start with the schema audit this week, the call tracking next month, and the monetisation rewrite the quarter after that. In that order, and with patience for the eight-month tail before the numbers turn.

Phone the CEO before you fire the content team, though. I’ve seen that mistake cost people more than the zero-click problem ever did.