How web directories have evolved since the late 2000s

Ask any SEO consultant under 35 what they think about web directories and you will get the same look. A small grimace. A polite cough. Then some version of: “Oh, those died around 2012, right?”

I have been hearing this for over a decade. It is wrong. Not partially wrong, not “well, it depends” wrong, but factually contradicted by the data anyone bothering to look can pull from BrightLocal, Birdeye, or their own Google Business Profile insights. The directory did not die. It changed shape, lost weight, got picky about who it lets in, and quietly became infrastructure for things SEO blogs do not write about, including the AI systems those same blogs now obsess over.

This article is my attempt to drag the corpse out of the morgue and prove it has been jogging the whole time.

The prevailing narrative about directories being dead

Open any SEO publication from the last decade and run a search for “web directories. You will find the same three or four arguments recycled by writers who, in many cases, were in primary school when DMOZ shut down. The narrative has calcified into received wisdom.

sankey-beta
  Business listing,Curated directories,60
  Business listing,Auto-approval farms,40
  Curated directories,Counted as citations,55
  Curated directories,Editorial rejection,5
  Auto-approval farms,Devalued by Google,38
  Auto-approval farms,Toxic backlinks,2
  Counted as citations,Local pack visibility,55
  Devalued by Google,Deindexed,38

What SEO blogs have claimed since 2011

The pattern starts roughly with the launch of Google Panda in February 2011, accelerates after Penguin in 2012, and hardens into orthodoxy by about 2015. The claims, in rough order of frequency: directories are link farms; directories are spam; directories pass no PageRank; Google ignores directory citations; nobody actually browses directories; directories were a 1998 idea that Google made obsolete.

I worked at a content agency in 2013 where the standard onboarding deck literally had a slide titled “Things we do not do anymore” with “directory submissions” listed between “article spinning” and “comment spam”. That deck, as far as I know, is still in circulation. I have seen variants of it at three different agencies since.

The trouble is that the slide bundles three completely different activities under one label. Submitting to 500 auto-approval link farms is not the same activity as getting your accountancy firm into the ICAEW’s member directory, which is not the same activity as appearing in a curated regional business index. Treating them identically is like saying “restaurants are bad for you” after one bad night at a petrol station kebab shop.

The Google Panda misinterpretation

Panda targeted thin content. Penguin targeted manipulative link patterns. Neither algorithm, in any documentation Google has ever published, targeted “directories” as a category. What got hammered were sites with low editorial standards, duplicate content, and reciprocal-link schemes. Plenty of directories fit that description; plenty did not.

The misreading happened because the casualties were visible and the survivors were quiet. When a directory of 4 million auto-imported listings lost 90% of its traffic overnight, that was a story. When a 12,000-listing curated industry index kept ticking along, nobody wrote a blog post about it. Absence of news became evidence of death.

Why this story stuck around

Three reasons, none of them flattering to my industry.

First, SEO blogs run on novelty. “Directories still work in some contexts” is not a headline. “Directories are dead” is. The death narrative gets clicks; the nuance does not.

Second, the people writing SEO blogs mostly do not work on the kinds of clients where directories matter. If your portfolio is SaaS startups chasing organic blog traffic, directories are genuinely irrelevant to you. If your portfolio is plumbers, solicitors, accountants, machine shops, B2B suppliers, or anything geographically anchored, directories are part of your monthly checklist whether you admit it publicly or not.

Third, the death narrative is convenient. If directories are dead, you do not have to do the boring work of researching which ones matter in your client’s vertical, writing 40 slightly different descriptions, and chasing approvals for six weeks. You can just write another blog post about topic clusters.

Evidence that contradicts the obituary

So what does the actual data show? I have spent the last few months pulling together numbers from citation tracking tools, Google Business Profile insights from a handful of clients who let me share aggregated figures, and the published research from a few of the firms still studying this.

journey
  title Auditing a regional accountancy firm's directory presence
  section Audit
    Pull citation report: 3: Operator
    Find 31 inconsistent citations: 2: Operator
    Spot 4 dead directory listings: 2: Operator
  section Cleanup
    Correct inconsistencies: 4: Operator
    Remove dead listings: 3: Operator
    Join two ICAEW directories: 5: Operator
  section Outcome
    Directory referrals reach 9 percent: 5: Operator
    Discovery searches rise 30 percent: 5: Operator

Traffic patterns from niche directories 2015-2024

The headline finding, from Birdeye’s directory research, is that businesses with structured, optimized directory listings see a 23% average increase in discovery rates compared with those without (Business Web Directory). That is not a vanity metric. Discovery rate is how often your business shows up in searches where you are not the target query, which is most searches.

BrightLocal’s citation work shows a similar pattern: businesses in multiple high-quality directories see a 23% increase in local search visibility (see the breakdown for new businesses). The fact that two independent studies landed on roughly the same number is, frankly, a bit suspicious to me, but the directional finding is consistent across every dataset I have looked at.

What I find more interesting than the headline numbers is the trajectory. From 2015 to about 2019, directory traffic was flat or declining in aggregate. From 2020 onward, niche and vertical directories started growing again, while generalist directories continued their slide. I have a client in industrial equipment whose top three referral sources after Google are all vertical directories that did not exist in 2015.

Local pack dependency on structured listings

This one is not arguable. The Google local pack, those three results with a map at the top of any locally-intended query, is built on citation data. Google cross-references your Business Profile against directory listings to confirm name, address, and phone consistency. Inconsistent citations correlate with lower local pack appearances. Anyone who has done local SEO seriously for more than 18 months knows this; the disagreement is only about which citations matter and how much.

Some businesses see a 40% increase in calls after proper directory listings (according to practitioner data collected here). I personally saw one mortgage broker client go from 4 to 11 weekly calls inside three months after we cleaned up 14 citation inconsistencies across the usual suspects, plus three vertical finance directories. Was it only the citations? No, we also rewrote the GBP description. But the citations were the only structured change in the first six weeks and the call jump started in week four.

B2B procurement research behaviour

Here is the part that almost nobody writes about because the data is messy and the buyers are private. B2B procurement still uses directories heavily. Not Yelp. Not Yellow Pages. Industry-specific directories: ThomasNet for industrial sourcing, Capterra and G2 for software, Clutch for agencies, Avvo for legal, Healthgrades for medical.

87% of consumers research businesses online before making purchasing decisions (recent consumer behaviour data), and for B2B buyers that number is effectively 100%. The question is not whether they research, but where. Gartner’s buyer journey work has shown for years that B2B buyers spend more time in independent third-party sources than on vendor websites. A meaningful slice of those third-party sources are, by any reasonable definition, directories.

The actual transformation nobody mapped correctly

If directories are not dead, what happened to them? The honest answer is that the category fragmented. There is no single thing called “a web directory” in 2024. There are at least four distinct categories, and lumping them together is the source of most of the confusion.

classDiagram
  class WebDirectory {
    +String editorialPolicy
    +int listingCount
    +verifyListing()
  }
  class GeneralistCatalog {
    +String browseByCategory
    +autoImportListings()
  }
  class VerticalDirectory {
    +String industryFocus
    +curateSubmissions()
  }
  class ReviewPlatform {
    +String ratingSystem
    +collectReviews()
  }
  class ApiDataNetwork {
    +String licensedFeed
    +distributeViaApi()
  }
  WebDirectory <|-- GeneralistCatalog
  WebDirectory <|-- VerticalDirectory
  WebDirectory <|-- ReviewPlatform
  WebDirectory <|-- ApiDataNetwork

From generic catalogs to vertical specialisation

The Yahoo Directory model, the alphabetical browse-by-category warehouse, is genuinely dead. What replaced it is vertical specialisation. Instead of one directory listing 2 million businesses across every conceivable category, you have hundreds of directories each covering a specific industry, profession, or geography with much higher editorial standards.

Examples I work with regularly: Houzz for interior design, BarkBox-adjacent platforms for pet services, FindLaw and Justia for legal, GoodFirms for tech services. These are not “directories” in the 2005 sense; they are vertical marketplaces with directory bones. They have reviews, profiles, lead-routing, sometimes booking. But underneath all that, they are still structured business listings, browsable by category.

The generalist directories that survived did it by getting selective. Jasmine Directory is a useful reference here: it kept the curated-submission model, refused the auto-approval route, and focused on editorial vetting rather than volume. That is the path that worked. Directories that tried to compete with Google on breadth got crushed; ones that competed on curation are still standing.

Curation replacing automation

The 2008 directory ran on auto-submission tools. You paid 47 dollars, software pushed your site to 500 directories overnight, and most of those listings were never reviewed by a human. By 2014, Google had successfully devalued essentially all of that activity.

The directories that adapted moved in the opposite direction. They added editorial review, required unique descriptions, charged submission fees that priced out spam, and started enforcing category accuracy. The trade-off was obvious: fewer listings, higher quality, slower growth, but listings that Google actually counted as citations.

API-first directories and embedded discovery

This is the part that most SEO writing misses entirely. The most important directories in 2024 are often the ones you have never visited as a human, because their interface is an API.

Yext, Foursquare’s Pilgrim SDK, Factual (now part of Foursquare), SafeGraph, and a dozen smaller players run what amount to directory databases that get embedded into other products. When you ask Siri for “coffee near me”, the result is not Apple’s own directory; it is licensed data from a network of directory providers. When a CRM auto-fills a company record, the source is often a directory API.

This is a fundamental category shift. Directories used to be destinations. Now many of them are data layers. The user never sees the directory, but the directory still gets paid, still verifies listings, still provides the structured data that the visible product depends on.

The AI angle deserves its own paragraph, because it is where the directory story gets genuinely interesting in 2024. Large language models hallucinate business information confidently and incorrectly. The fix, which several AI search products now use, is retrieval from structured sources. Directories with verified, current data are some of the cleanest structured sources available. I have seen Perplexity, ChatGPT search, and Google’s AI Overviews all surface information that traces back to directory entries. The “directories are obsolete because of AI” argument gets it exactly backwards.

Steelmanning the directories-are-obsolete camp

I have been one-sided so far, so let me try to make the opposing case as strongly as I can. There are real arguments on the other side, and pretending otherwise would be dishonest.

graph LR
  A[Search engines understand intent] --> B{Disintermediates directories?}
  B -->|Pizza near me| C[Yes: directory layer collapses]
  B -->|Six-figure software choice| D[No: buyers compare a curated set]
  A --> E[But engines need data to be good]
  E --> F[Directories supply that data]
  F --> G[Verified structured listings]
  D --> G

Where Yahoo and DMOZ genuinely failed

Yahoo’s directory was, for a brief period, the front door to the web. Then Google made it obsolete in about 18 months. DMOZ limped on with volunteer editors until 2017 when AOL finally shut it down. These failures were not because directories as a concept were wrong; they were because manually curating “the whole web” was always going to lose to algorithmic search at scale.

The lesson the obsolete-camp draws is correct as far as it goes: any directory that tries to compete with Google on coverage will lose. The error is generalising that to “all directories lose to Google”, which does not follow. A vertical directory of UK chartered surveyors is not competing with Google; it is competing for a slice of the trust and discovery layer that Google does not fill well.

The link spam era and its lasting damage

From roughly 2006 to 2012, the directory category was genuinely flooded with spam. Hundreds of “directories” existed purely to sell links. Their listings had no editorial value, their traffic was effectively zero, and their only purpose was passing PageRank.

The damage from that era is real and lingering. Many businesses still have toxic directory backlinks from 2009 that show up in audits. The directory category as a whole carries reputational baggage that newer entrants did not earn. When a marketing director hears “directory listing”, their pattern-match goes to the bad version, and that is a fair instinct given what they probably saw five years ago.

Search engine sophistication arguments

The strongest version of the obsolete-camp’s case is this: search engines have gotten so good at understanding intent, entities, and authority that the intermediary role directories used to play is collapsing. Why would I browse a directory of accountants when Google can answer “best accountant in Leeds for small ecommerce businesses” with a personalized pack of results?

This is a serious argument and I want to take it seriously. It is also, I think, partly right. For some queries and some user segments, the directory layer has genuinely been disintermediated. The 19-year-old looking for a pizza place does not use a directory, and never will again.

But “search engines are better at intent” runs into two limits. First, search engines need data to be that good, and directories supply a lot of that data. Second, search-engine answers are increasingly opinionated and opaque; users who want to compare options across a curated set, rather than accept Google’s top three, still benefit from directories. The B2B procurement case is the clearest version of this. No serious procurement team makes a six-figure software decision based on the Google local pack.

A decision framework for operators in 2024

If you are running marketing for a real business and trying to decide whether and how to invest in directory presence, here is how I would think about it. This is not “directories are great, do them all”. It is a triage framework.

gitGraph
  commit id: "Yahoo/DMOZ era"
  commit id: "Panda 2011"
  commit id: "Penguin 2012"
  branch curated
  checkout curated
  commit id: "Editorial review"
  commit id: "Unique descs"
  commit id: "Submission fees"
  checkout main
  commit id: "Link farms gone"
  merge curated id: "Now citations"
  branch api-first
  checkout api-first
  commit id: "Yext/Foursquare"
  commit id: "AI grounding 24"
  checkout main
  merge api-first id: "Data infra"

When directory submission still moves metrics

Directories move metrics when three conditions are true: your business has a geographic or vertical focus that maps to existing directories; your buyers conduct independent research before contacting you; and your competitors are listed in those directories.

If all three are true, directory investment is not optional, it is hygiene. You will lose market share to competitors who maintain better listings, in roughly the same way you would lose market share by having a worse website.

If only one or two are true, the calculation is messier. A pure-play SaaS company selling to global enterprises has weak geographic anchors and may not benefit from local directories, but probably does benefit from G2 and Capterra. A local restaurant has strong geographic anchors but limited vertical specialisation, so the local pack matters more than industry directories.

Audience contexts where directories outperform search

There are specific contexts where directories beat search engines for user intent. These are worth understanding because they tell you where to invest.

First, comparison shopping for considered purchases. Buyers who want to compare 8 to 15 vendors side by side, with structured attributes, find directories more useful than search. Google gives you “best in class”, which is fine when you trust the algorithm; directories give you the full set with filters, which is better when you do not.

Second, trust verification for high-stakes services. When the cost of choosing the wrong supplier is high (legal, medical, financial, structural), buyers want a third-party validator. Directory listings with verified credentials, accreditations, and review systems provide that validation in a way that ranking on Google does not.

Third, geographic discovery in unfamiliar locations. Travellers and people relocating use directories more than locals do, because they need browseable structure rather than name-based search.

You can see in Figure 3 why directory investment pays off even when the directory itself is not the conversion source. The directory shapes the consideration set; the vendor website closes the deal. Attribution models that credit only the last click systematically undervalue directories, which is part of why marketing teams underinvest in them.

Red flags that signal a dying directory

Not every directory is worth a submission. Here is how I quickly assess whether a directory is alive or fossilised.

I check the publication dates on listings. If the most recent additions are from 2019, the directory is not being maintained. I look at the editorial process: is there a review queue, a category structure, a contact for corrections? If submission is one form and an auto-approval, that is a link farm in disguise. I check the outbound link patterns. Directories that link to obvious spam alongside legitimate businesses have no editorial filter. I run the directory URL through a backlink tool to see whether quality sites still cite it. A directory with no inbound mentions from real publications is dying, regardless of what its homepage looks like.

A worked example of this in practice: I worked with a regional accountancy firm in 2023 that had decent Google rankings but flat lead growth. We audited their directory presence and found 31 inconsistent citations, listings on 4 dead directories, and no presence on the two main UK accountancy directories despite the firm having ICAEW members on staff. Over four months, we corrected the inconsistencies, removed the dead listings (some required emailing webmasters whose addresses bounced, which tells you everything), and got the firm into the two vertical directories with full member profiles. Direct directory referrals went from negligible to 9% of new client enquiries. More interestingly, GBP “discovery” searches, where the firm appeared for non-branded queries, increased by about 30%. Some of that was probably other factors. But the directory work was the only structural change in the first three months.

Was this a miracle? No. It was hygiene work that the firm’s previous agency had not bothered with because directories were “dead”. The opportunity in directory work in 2024 is partly that so many of your competitors believe the obituary and are not doing the boring maintenance.

Where this leaves us

The directory category in 2024 is smaller, more specialised, more selective, and more technically integrated than it was in 2008. It is also, depending on your industry, more important to your discovery footprint than it was at any point in the last 12 years. Those two facts are not in tension; they are the same fact stated from different angles.

If you are an operator deciding where to put marketing time next quarter, my honest recommendation is to spend 90 minutes auditing your current directory presence before you commission another piece of content. Pull a citation report. Check the top three vertical directories in your industry. Look at the consistency of your name, address, and phone across listings. You will probably find at least a dozen things wrong, and fixing them is faster and cheaper than almost anything else on your marketing roadmap.

The people writing directory obituaries can keep writing them. I would rather be the operator quietly collecting the leads they are leaving on the table.