Directory and citation links

By Bill Hartzer · Hartzer Consulting · 25+ years in search, domains and expert witness work

What this tactic is, and why it is really two tactics

People say directory links and mean one of two completely different activities. Separating them is most of the work of understanding this page.

Local business citations are structured listings of a business on platforms that hold business data: Google Business Profile, Apple Business Connect, Bing Places, Yelp, the major data aggregators that feed smaller sites, industry association registers, chamber of commerce member lists, and vertical platforms like a legal directory or a restaurant guide. The core content of a citation is NAP data — name, address, phone number — usually with hours, categories and a website URL. Many citations are nofollowed, and some do not link at all. They are still worth having.

General web directories are the descendants of the 1990s web catalogue: a taxonomy of categories with a list of sites under each one, accepting submissions from anywhere, on any topic, usually for a fee. The two that mattered are both gone. The Yahoo Directory closed to submissions and then shut down. DMOZ, the Open Directory Project, went offline in 2017. What is left is a long tail of sites that exist only to sell listings.

The grade for this tactic is situational, and the situation is which of those two you are doing. Citations are ordinary business infrastructure. General directory submission is a paid link program with a lower price and a worse product.

How it works, step by step

The citation workflow is a data exercise, not an outreach exercise.

1. Fix your canonical NAP first. Decide the exact string for your business name, address and phone, down to whether it is Suite 4 or Ste 4, and write it down. Every listing you create or correct from this point uses that string. Doing this after you have built forty listings means auditing forty listings.
2. Claim the primary platforms. Google Business Profile first, then Apple, Bing, and the one or two vertical platforms that dominate your industry. These are the listings that appear in map results and voice assistants; they are the reason the exercise matters.
3. Handle the aggregators. A handful of data companies supply business records to hundreds of smaller sites. Correcting a record at the source fixes downstream copies you would never find manually.
4. Audit what already exists. Search your phone number in quotes, your business name in quotes, and old addresses. Most established businesses have duplicate and stale listings created by scrapers, former staff or previous agencies. Merging and correcting these is worth more than creating new ones.
5. Add the genuinely local and genuinely vertical. Chamber of commerce, business improvement district, trade body register, supplier locator, licensing board. These are the ones that carry actual links and actual referral traffic.
6. Stop. There is a finite list. When you have covered the platforms with users, the aggregators, and the bodies you actually belong to, the tactic is complete. Everything past that point is submission for its own sake.

The general directory workflow, for completeness, is: pay a fee, fill a form, receive a listing on a page nobody visits. There is no step five.

What it costs in time and effort

Citations cost a day or two of careful data work up front, and an hour or two a year afterwards. The up-front cost is dominated by auditing what already exists rather than creating anything new, because duplicate and conflicting records are the normal state for any business older than a few years.

You can pay for this. Citation management services charge a recurring fee to push and maintain your data across a network, and for a multi-location business that is usually money well spent, because the cost of doing it manually rises with the square of your locations and your staff turnover. For a single-location business, the manual version is a weekend.

General directory submission is priced as a volume product, and the price per listing falls as the quality falls to zero. The relevant cost is not the fee. It is that you have created a batch of near-identical listings with matched anchors across a set of sites that share hosting, templates and owners, and that pattern is visible to anyone who looks at your profile — including you, later, when you are trying to work out what went wrong.

When it works and when it does not

Citations work when the business has a physical location or a defined service area and competes in local results. For a plumber, a dental practice, a restaurant, a law firm with an office, or a franchise network, consistent citation data is a prerequisite rather than a tactic. It is not that the links are strong; most of them are nofollowed or weak. It is that the corroborating data helps a search engine be confident that your business is real, is where you say it is, and is the same entity across every mention of it.

They do not work as a link tactic in the ordinary sense. If you sell software to customers in forty countries, citations do almost nothing for you, and no amount of them will move a competitive national ranking.

General directories do not work at all, and I want to be specific about why rather than just asserting it. A directory link is unedited, sits on a page with dozens or hundreds of unrelated outbound links, is bought rather than earned, comes from a site with no readership, and follows a template shared with a hundred other directories run by the same operators. Every one of those properties is something search engines learned to discount many years ago. The listing is not usually harmful on its own. It is worthless, and a hundred worthless things bought together look like a purchase.

The one honest exception is a small number of curated vertical directories with genuine editorial standards and genuine users — a professional register that verifies licences, a niche marketplace buyers actually search. These are directories in name only; they are membership organizations. Judge them by whether a human being checks your application.

Common mistakes

• Treating citations and directories as one line item. The commonest error, and the reason people either dismiss both or buy both. They are unrelated activities with unrelated returns.
• Chasing volume. A published count of listings is a vanity metric. Fifteen accurate records on platforms with users beat three hundred on sites without them.
• Inconsistent NAP. Two phone numbers, an old suite number, a trading name on some listings and a legal name on others. This is the actual failure mode of local SEO, and it is caused by nobody writing the canonical string down.
• Ignoring duplicates. A second, unclaimed Google listing at an old address will quietly compete with the one you maintain.
• Buying packages of a thousand submissions. The sales pitch has not changed since 2006 and neither has the product.
• Assuming a directory is good because its domain metric is high. Directory sites accumulate links from everyone they list, which inflates domain-level scores without saying anything about the value of a listing page buried six levels deep.
• Never re-checking. Platforms close, get acquired, or drop their outbound links wholesale. Listings decay like every other link.

A worked example

A three-branch veterinary practice asks why it does not appear in map results for two of its three towns.

The audit comes first. Searching each phone number in quotes surfaces 34 existing listings. Of those, nine are correct, eleven carry a phone number retired two years ago, six use the old name from before a merger, five are for a branch that has moved, and three are duplicates of the main location created by scrapers. Nothing new has been built yet and there is already a fortnight of corrections available.

The work runs in order: claim and correct all three Google Business Profiles, fix Apple and Bing, correct the record at two data aggregators, then update the veterinary regulator's public register, the two professional association member lists the practice actually belongs to, and both local chambers. That is the finite list. It comes to roughly two dozen listings.

Of those, perhaps six carry a followed link. The regulator's register and the chambers are the ones with any independent standing at all. The rest contribute consistency rather than links.

Three months later the two missing branches appear in local results. Nothing in that outcome came from a link in the ordinary sense; it came from the same business data appearing in the same form everywhere a search engine looked. Had the same budget gone into five hundred general directory submissions, the branches would still be missing.

How to measure it

Citations and directories need different measures, because they claim to do different things.

• Consistency rate. The share of discoverable listings carrying your exact canonical NAP. This is the primary number for citations, and it is the one that moves local visibility.
• Duplicate count. Track it to zero. A duplicate is worse than a missing listing.
• Local pack and map visibility by location. Measured per branch, not as a site-wide average, because the whole point of the exercise is per-location.
• Referral traffic and calls per platform. Some verticals send real customers. Most send none. After six months you will know which listings are businesses and which are filing cabinets.
• For general directories: Trust Flow and referring traffic. Run the check honestly. On my own profile, 69.3% of 22,260 referring domains sit at Trust Flow 0 despite no link ever being bought — weak domains accumulate naturally, which is exactly why deliberately buying more of them adds nothing you were not going to get for free.
• Anchor distribution. If a batch of directory listings all use the same commercial anchor, that shows up in your anchor profile as a spike. Check for it before someone else does.

The verdict

Do the citations. Do them once, carefully, with a written canonical NAP and a real audit of what already exists. For any business with a location or a service area, this is infrastructure rather than link building, and skipping it is a self-inflicted wound.

Do not buy general directory submissions. The category has been dead since the Open Directory Project closed, the surviving sites exist to sell listings rather than to catalogue the web, and a bulk purchase produces the one thing you least want: a cluster of identical, unedited, paid links appearing at the same time with the same anchor.

The honest summary is that half of this tactic is necessary and unexciting, and the other half is the oldest product in the industry, still selling because it is cheap and because the invoice has the word links on it.