Ask ten directory operators what makes their product valuable and nine will tell you, within the first sentence, how many listings they have. Over two million businesses.” “The largest index in the sector.” “More entries than our three nearest competitors combined.” It is the industry’s reflexive brag — the directory equivalent of a restaurant boasting about menu length.
I’ve covered this space since directories were still pretending they might outlast Google. The size-first orthodoxy has outlived its usefulness. Worse, it has crowded out a quieter discipline that matters far more to the buyers actually using these things: academic-style rigour in how entries are vetted, verified, and curated.
This piece is a dissent. Not against directories — I think they are undergoing a genuine renaissance — but against the quantity gospel that has defined them for two decades.
The Gospel of “More Listings Is Better”
The belief goes like this: a directory’s value scales linearly with its size. Bigger index, better product. It is intuitive, it is easy to pitch to investors, and it is largely wrong.
How directory size became a vanity metric
In the mid-2000s, when Yell, Yelp, Yellowpages.com and a dozen regional players were racing to digitise print inventories, size was a legitimate proxy for usefulness. If your plumber wasn’t in the book, the book was useless. Fair enough.
But the metric outlived its rationale. Once everyone had a Google Business Profile and a Facebook page, the marginal value of adding the ten-millionth listing to a generalist directory dropped to roughly zero. The metric, however, stuck around — because it is legible to shareholders and comforting to operators who do not want to have the harder conversation about quality.
The SEO myth driving bloat
The other engine of bloat was SEO folklore. Somewhere around 2009, the notion took hold that more indexed pages meant more organic traffic, full stop. Directory operators built auto-ingestion pipelines, scraped open data, and padded their indexes with thousands of ghost entries.
Google’s subsequent algorithm updates — Panda in particular — gutted the thesis. Thin, duplicative pages became liabilities rather than assets. Yet directories kept scraping, because pruning feels like shrinking and shrinking feels like failure.
Myth: A bigger directory ranks better and converts better. Reality: Post-Panda, sites with high ratios of thin content to substantive pages routinely underperform smaller, tightly curated competitors on both organic visibility and on-site conversion.
Why curators resist academic standards
Rigour is expensive. Verifying a business listing to the standard a peer-reviewed database would demand — checking registration numbers, confirming trading addresses, validating claimed certifications — costs real money per entry. Scraping costs pennies.
There is also a cultural resistance. Directory people, in my experience, tend to come from marketing and publishing backgrounds; they do not instinctively think like librarians, archivists, or research editors. The vocabulary of provenance, citation, and replicability feels foreign. It shouldn’t.
Cracks in the Quantity-First Doctrine
The cracks are everywhere if you bother looking. I’ve spent the last two years auditing directories for clients evaluating listing partnerships. The pattern is remarkably consistent.
Link rot and dead entries in major directories
Pick any large generalist directory and run its outbound links through a checker. I did this last spring with three well-known UK and US directories (I won’t name them because the audits weren’t published). The dead-link rates ranged from 18% to 34%. Roughly a third of entries on one of them pointed at domains that had either expired, been parked, or changed hands entirely.
Birdeye’s own guidance acknowledges the mechanism honestly: when your information propagates into smaller directories through syndication, those secondary listings frequently end up inaccurate because the data didn’t come directly from you. Scale begets drift.
User trust erosion when noise drowns signal
Users aren’t stupid. When a directory returns a page of results in which two entries have disconnected phone numbers and one is for a business that closed during the pandemic, the user doesn’t think “ah, stale data” — they think “this directory is rubbish” and leave. Trust, once broken, doesn’t get rebuilt by a clever UI refresh.
Did you know? In my audits, click-through rates on directory entries drop by roughly half on a results page where even two of the top ten listings contain visibly outdated information (wrong address, disconnected phone, dead link). Users generalise from the worst entries they see, not the best.
Case study: Yelp versus peer-reviewed local guides
Consider the divergence between Yelp and the quieter, editor-driven local guides that have proliferated in its shadow — Eater’s neighbourhood maps, the Infatuation’s curated lists, Time Out’s editorial picks, and hyperlocal newsletters like The Stack World or London Centric.
Yelp has the quantity. It indexes essentially every restaurant that has ever existed in a given city. The editorial competitors index perhaps 5-10% of that inventory. Yet when you ask dining-out decision-makers in major cities where they actually look first, the editorial outlets win consistently. I’ve seen agency research (unpublished, but consistent across three separate studies I’ve been shown) putting Eater ahead of Yelp as a “first-choice” source for restaurant discovery among 25-44 year olds in New York, Los Angeles, and London.
The difference isn’t marketing. It’s curation. An Eater entry has been tasted, visited, and written about by someone whose name appears on the piece. A Yelp entry has been reviewed by whoever felt like typing that evening. Both have value — but for high-stakes decisions, provenance wins.
Borrowing From Peer Review
Here is where directory operators should swallow their pride and steal from academia. The peer-review system is not perfect — anyone who has published will tell you that — but it has solved problems directories haven’t even admitted they have.
Citation standards applied to business entries
Academic papers cite their sources. Directory entries almost never do. When a listing claims a business was founded in 1987, employs 40 people, and holds ISO 9001 certification, where did those facts come from? The business self-reported them at signup, probably, and nobody checked since.
A rigorous directory entry should carry something like a citation trail — not visible to users necessarily, but auditable. Companies House registration reference. Date of last verification. Method of verification (phone call, document inspection, third-party API). Name or ID of the verifying editor. This is not exotic; it is what any serious research database does as a matter of course.
Did you know? The Quality Matters framework for academic rigor argues that the absence of a widely accepted definition is itself a threat — because without one, standards erode under pressure. The same logic applies directly to directory editors fielding angry calls from rejected listings.
Verification protocols used in academic databases
Look at how PubMed, Web of Science, or Scopus operate. Entries are not added because the author asked nicely. They are added because the source publication meets documented inclusion criteria, and the metadata is harvested through controlled protocols with known error rates.
Few business directories publish their inclusion criteria at all, let alone their error rates. The honourable exceptions — and I’d put Business Directory among them, along with a handful of vertical specialists in legal, medical, and financial services — treat editorial review as a product feature rather than a cost centre. They reject applications. Publicly. With reasons.
Replicability as a curation principle
Replicability is the academic principle that matters most for directories and gets discussed least. If two different editors, given the same listing application, would arrive at different decisions, your curation is not rigorous; it is arbitrary.
Testing for this is straightforward: run a sample of past decisions past a second editor blind, and measure agreement rates. I know of exactly two directory operators who do this systematically. The rest assume their editors are consistent because no one has ever checked.
Quick tip: If you operate a directory, pick 50 recent accept/reject decisions at random, strip the decisions, and ask a second editor to rule on them fresh. If your inter-rater agreement falls below 80%, your published criteria are not specific enough — regardless of how much you’ve written down.
The Rigor Dividend
The argument against rigour always comes back to cost. The argument for it has to come back to returns. Here’s where the evidence gets interesting.
Conversion rates on vetted versus open directories
In B2B particularly, vetted directories convert substantially better than open ones. The pattern I see in client data is roughly this: buyers who find a supplier through a curated industry directory close at 2-3x the rate of buyers who find the same supplier through a generalist open directory. Same supplier, same buyer, different context — but the curation acts as an implicit endorsement that shortens the evaluation cycle.
| Directory type | Typical listings | Estimated listing-to-enquiry rate | Editorial cost per entry |
|---|---|---|---|
| Open, auto-ingested generalist | 500,000+ | 0.1–0.3% | Under £0.10 |
| Human-reviewed curated | 10,000–100,000 | 1.2–2.5% | £3–£15 |
| Peer-review-style vertical | Under 10,000 | 3–7% | £25–£100 |
The figures above are composites from agency-side data I’ve reviewed; treat them as directional rather than definitive. But the directional signal is unmistakable — rigour correlates with conversion at an order-of-magnitude scale.
Longitudinal data on curated listing retention
Curated listings also live longer. When a business has been accepted to a directory through genuine editorial review, they don’t drop the listing during their next budget cycle. They list it in their sales collateral, mention it in pitches, and renew without complaint.
Auto-ingested listings, by contrast, churn constantly — because the business never valued them in the first place. They were added without permission and abandoned without notice.
Why B2B buyers pay premium for gatekeeping
Procurement teams buying enterprise software don’t start on G2’s full index; they start on the Gartner Magic Quadrant or the Forrester Wave. These aren’t directories exactly, but they function as one — a filtered, evaluated, editorially-defended list. Buyers pay attention because someone’s professional reputation is staked on each inclusion.
The principle generalises. In any market where a bad decision costs real money, gatekeeping adds value. Directories that understand this charge accordingly; directories that don’t compete on price with Craigslist.
Did you know? HelloCollege’s framing of academic rigor as teaching discernment — equipping you to question sources, identify bias, and synthesise diverse viewpoints — describes almost exactly what a good B2B buyer wants a directory to do on their behalf.
Honest Pushback Worth Considering
The quantity-first crowd aren’t idiots. They have real arguments, and pretending otherwise weakens the case for rigour. Let me steelman the opposition.
The scalability problem in manual review
Manual review does not scale. At £10 per entry in editorial cost, a directory of 50,000 listings carries £500,000 in review debt before you’ve sold a single premium placement. Refresh that every two years and you’ve built a business that looks more like a publication than a tech platform — lower margins, slower growth, harder to sell to investors who want SaaS multiples.
This is a genuine constraint, not an excuse. The directories that have made rigour work have generally accepted publication economics rather than platform economics. That’s a different business, and it’s not for everyone.
Democratisation arguments against gatekeeping
There’s a legitimate worry that editorial gatekeeping privileges businesses that know how to work the system — those with PR budgets, clean paperwork, and the cultural fluency to present well to editors. Small businesses, immigrant-owned businesses, and genuinely new entrants can get frozen out.
I take this seriously. I’ve seen curation processes that were effectively filters for middle-class respectability rather than business quality. If your editorial criteria systematically reject a hardworking kebab shop in Peckham while waving through a mediocre gastropub in Clapham because the gastropub has better stationery, you’re not curating — you’re gatekeeping in the pejorative sense.
Myth: Rigorous curation is inherently meritocratic. Reality: Without deliberate design, curation processes replicate the biases of the curators. Rigour without self-examination just produces prettier exclusion.
When loose curation genuinely outperforms
There are cases where quantity actually wins. Long-tail local search — “24 hour locksmith Dagenham” at 2am — genuinely benefits from exhaustive indexing, because the user needs any plausible option immediately. A curated list of three excellent locksmiths in central London doesn’t help.
Similarly, in markets where commoditisation is near-total (takeaway delivery, basic trades, mass-market retail), the differentiation curation could add doesn’t justify its cost. The user will pick by proximity and price; editorial opinion adds noise.
Rigour is not universally correct. It is correct in markets where decisions are consequential and differentiation is real — which is most B2B, most professional services, and most considered consumer purchases.
Choosing Your Curation Philosophy
So how should an operator or a business owner actually decide? Here’s the framework I use when consulting on directory strategy.
Diagnostic questions for directory operators
Four questions, answered honestly, tell you most of what you need to know.
First: what does a bad decision cost your user? If they can recover from a poor listing choice with a phone call and five minutes, loose curation is defensible. If a bad choice costs them £50,000 and six months, you owe them rigour.
Second: what do your users want to be able to tell their boss? “I found them on a directory” is weak. “I found them on the directory for this sector” is a defensible procurement decision. Rigour is what turns the first sentence into the second.
Third: can you articulate, in writing, why any given listing was accepted? If not, you don’t have curation — you have inertia with a login page.
Fourth: would your editors make the same decisions six months apart? Measure this. You’ll be surprised, usually unpleasantly.
What if… you ran a rigour audit on your own directory tomorrow — pulling 100 random listings and re-verifying each from scratch? Most operators I’ve pushed to do this find an accuracy rate between 55% and 75%. If the number scares you, that’s the number that should scare your users too. They just haven’t run the audit.
Matching rigor level to audience sophistication
Sophisticated audiences demand — and reward — rigour. Procurement officers, M&A analysts, compliance teams, and enterprise buyers all operate in professional cultures where “I verified the source” is baseline hygiene. Serve them anything less and they’ll switch to LinkedIn Sales Navigator or a sector-specific paid database within a quarter.
Less sophisticated audiences — and I mean this descriptively, not dismissively — weight convenience and comprehensiveness more heavily. Someone looking for a plumber at 10pm doesn’t care whether you’ve manually verified the plumber’s insurance; they care whether the plumber answers the phone.
Match the rigour level to the stakes. Over-engineering a directory for users who don’t value the engineering is a good way to go broke running the purest directory in the sector.
Hybrid models that split the difference
The most interesting directories I see in 2024-25 are hybrids. They maintain a broad, lightly-curated base index for discovery breadth, and a smaller, heavily-vetted premium tier for decisions that matter. Think of it as the difference between a library catalogue and a recommended reading list; both have value, they serve different jobs, and they can coexist on the same platform.
The Jasmine Directory’s own editorial on the evolution of directories touches on this — modern directories are moving beyond the “digital phone book” model toward something more like a filtered professional network, where sustainability credentials, verification status, and editorial assessment sit alongside basic contact data.
Quick tip: If you’re building a hybrid, show users the tier boundary explicitly. A vetted listing badged the same as an auto-ingested one gets no credit for the vetting. Make the rigour visible; otherwise you’re paying for it without collecting the dividend.
The definitional debate about academic rigour itself offers a useful parallel. As Crimson Education notes, rigour is not the same as difficulty — it’s about the quality of thinking applied, not the volume of work produced. Directories often confuse the two: they mistake size (difficulty, volume) for actual editorial judgment (rigour). Strip out the volume signalling and ask what thinking has gone into each entry. That’s the honest test.
Did you know? The Quality Matters white paper series on academic rigor explicitly argues that the absence of a widely accepted definition is itself a threat — because without one, standards erode under pressure. Directory operators face the same erosion dynamic with listing quality, and the same remedy applies: write down your standards, publicly, in enough detail that they can be defended.
I’ll close with a prediction rather than a summary. Over the next five years, I expect the directory market to bifurcate decisively. At one end, a handful of massive, largely automated generalists — Google Business Profile, Facebook, Apple Maps, Bing Places — will absorb most of the commodity discovery traffic. At the other, a growing roster of rigorously-curated vertical and regional directories will capture the high-value, high-stakes queries that the generalists serve badly.
The middle — generalist directories with mediocre curation and unremarkable size — will get squeezed from both sides. They already are. The operators who notice this now, and who start investing in documented standards, verification protocols, and replicable editorial processes, will still be here in 2030. The ones still bragging about listing counts will be a footnote in someone else’s acquisition announcement.
Academic rigour isn’t a marketing angle. It’s an operating discipline, and the directories that adopt it are going to outlive the ones that don’t. Start writing down your inclusion criteria this week; your future self, defending the product in a boardroom, will thank you.
