Build organic authority

You rank for the easy terms and cannot get near the ones that matter

A site can be technically sound, well written and completely unable to compete for the searches that pay. That ceiling is not fixed by more pages. It is fixed by depth on a subject, coverage from people who are not you, and facts about your organisation that agree with each other — over a timeframe nobody enjoys hearing.

Book a Strategy Call See how we work

The reframe

Authority is not a score you raise, it is a subject you own

The plateau almost every business hits looks the same. Specific, long searches work. Broad, valuable ones do not. The instinct is to conclude that something is missing from the pages, and to commission more of them. That instinct is what keeps businesses on the plateau, because it treats a subject-level problem as a page-level one.

It helps to think about what a search engine is being asked to do on a competitive commercial term. Among a hundred pages that all describe the same service competently, it has to decide which organisation is most likely to be a good answer. Very little of that decision comes from the page in front of it. Most of it comes from what else exists: whether this site has covered the surrounding subject repeatedly and connectedly, whether other people have referred to it for that subject without being asked, and whether the facts about the organisation are consistent enough to be confident it is who it says it is.

That is what authority means in practice, and it explains a pattern that otherwise looks unfair. A competitor with a worse page can outrank you because the page is not the unit being judged. The site is, on that subject, and they have been building it for longer.

None of this is a metric you can address directly. There is no field to fill in, no score to raise, and the third-party numbers sold as authority are estimates produced by companies that do not run a search engine. What can be addressed is the three things underneath: depth on a subject, coverage from elsewhere, and consistency about who you are.

Why the ceiling is there

Three reasons a competent site cannot break through

These are not technical faults. Sites in this position are usually well built, and that is precisely why the plateau is confusing to the people running them.

The site covers many subjects and owns none.: Content has been produced against a calendar rather than against a question. The result is a wide, shallow footprint: enough to be found for unusual phrasings, never enough to be treated as a source on anything. The tell is a content list where no two consecutive pieces belong together, and an internal link structure that exists mainly through a navigation menu. Adding another unrelated piece extends the problem.
Nobody outside the business has ever referred to it.: Every link pointing at the site was placed by the business itself: directories, partner pages, a supplier list, an old press release, some guest posts bought in a bundle. There is no reference from anyone who read something and decided it was worth pointing at. That absence is legible, and it caps what any amount of on-site work can achieve on a contested term.
The organisation is hard to pin down.: Three versions of the company name in circulation, a trading name that differs from the registered one, an old address on half the web, no consistent description of what the business does, and profiles under a founder who left in 2019. This is not fatal on its own, but it makes every other signal harder to attribute confidently to one organisation, and it is the cheapest of the three to fix.

How the depth gets built

Pick one subject and exhaust it before starting another

This is deliberately narrow. The discipline is in what you decline to publish, which is harder to sustain than the writing itself.

Choose the subject the business must own
Not the biggest search term. The subject where your commercial future sits, where you have genuine expertise, and where the demand is real. One subject. Writing this down as a sentence, and having the leadership agree to it, is most of the work and the step most often skipped.
You get: One written subject commitment, signed off
Map the whole question, not the keywords
List everything a buyer needs to resolve on the way to a decision in that subject: the definitions, the comparisons, the objections, the costs, the mistakes, the edge cases. Keyword tools show the phrasings; they do not show the gaps between them, which is where the pages that get referenced tend to live.
You get: A complete question map with the gaps marked
Fix what already exists before adding to it
Most sites in this position already have three thin pieces on the subject, published years apart, competing with each other. Consolidate them into one, redirect the rest, and make the survivor genuinely better than anything currently ranking. This usually moves things before a single new page is written.
You get: A consolidated, corrected existing set
Connect the set so it reads as one body of work
Internal links written as sentences, in the direction a reader would actually travel, from the supporting pieces toward the commercially important ones. A group of unlinked pages on one subject is not depth; it is a collection. This step costs almost nothing and is left out of most content programmes.
You get: An internal link structure written by hand
Publish the one thing only you could publish
Data you hold, a survey you ran, a method you use, a position nobody else in the category will take publicly. This is what gives other people a reason to refer to you, and it is the only reliable route to earned coverage that does not depend on asking strangers for favours.
You get: One genuinely original asset per subject
Keep it current and resist widening
Revisit the set on a schedule and update it properly, with the date and the change recorded. Then do not start a second subject until the first is unambiguously working. The pressure to widen will come from inside the business, and giving in to it is how the plateau gets rebuilt one level higher.
You get: A maintenance schedule and a stated stopping rule

The routes, honestly compared

Five ways people try to build authority

Two of these work slowly, one works and is expensive, and two carry a liability whose timing you do not control. All five are sold with equal confidence.

Routes to organic authority compared by what they actually produce and what they risk
Dimension	What it produces	What it costs	What can go wrong
Depth on one subject	Recognition as a source, compounding across the whole subject area	Time, editorial discipline, and refusing to publish other things	Nothing, except that it is slow enough for a business to abandon it early
Original research or data	A reason for other people to reference you unprompted	Real cost and real effort, and it must be genuinely new	A weak study gets ignored, or worse, gets picked apart in public
Earned editorial coverage	References from sources with their own standing and readership	Sustained outreach, a story worth telling, and a high failure rate	Nothing structural, but the hit rate is low and the work is relentless
Guest placements bought at scale	A large number of links from sites that exist to sell links	Cheap per unit, which is the warning rather than the attraction	Against search engine policy, and the sites themselves tend to be devalued together
Paid link networks and expired domains	A short-lived rise that looks like progress on a chart	Ongoing payment to keep something you never owned	A liability of unknown size, triggered at a time nobody at your company chooses

What we work on

Four inputs, and the honest timeline for each

Topical depth
A connected body of work that exhausts one subject, with the existing thin pages consolidated rather than left to compete. Early movement on the specific edge searches within a few months; movement on the head terms considerably later, and only if the subject holds.
Depth as an architecture decision
Earned coverage
References from people with no obligation to give them, obtained by publishing something worth referencing rather than by asking. Individual results are unpredictable; the accumulation is what matters, and it is read over years rather than quarters.
Coverage from publications you do not control
Entity consistency
One name, one description, one address, one set of profiles, agreeing with each other everywhere the organisation appears. The fastest of the four to complete and the least visible in its effect, which is why it is almost always still outstanding.
A timeline nobody enjoys
Six to eighteen months before movement on contested commercial searches is unambiguous, longer against entrenched incumbents. The correct response is to hold those terms with paid media in the meantime and to stop treating the organic programme as something that should report monthly wins.

Questions

Questions from people who have already done the easy work

Is domain authority a real thing we should be trying to raise?

The scores sold under that name are third-party estimates produced by companies that do not operate a search engine. They are useful for comparing two sites roughly and useless as a target, because optimising toward an estimate optimises toward whatever that vendor happens to measure.

The underlying idea is real: some sites are trusted more than others for some subjects. But it is subject-specific rather than site-wide, which is why a site can be authoritative on one topic and invisible on another it has never covered properly.

Our competitors obviously buy links. Why should we not?

Because you are comparing their present state with your present state and ignoring the liability sitting underneath theirs. Purchased links are against the policies of the search engines you depend on, and the consequence, when it arrives, is not scheduled by you.

The practical argument is simpler than the ethical one. A business that has to disclose its marketing practices during due diligence, or that suddenly has to rebuild its entire acquisition channel in a quarter, has bought a problem more expensive than the visibility it rented.

How long does this actually take?

On genuinely competitive commercial searches, somewhere between six and eighteen months before the movement is unambiguous, and longer in categories where the incumbents have a decade of accumulated coverage. Anyone giving you a tighter number is guessing.

What can be seen sooner is direction: the long, specific searches around a subject start moving well before the head term does, and that early movement on the edges is the honest leading indicator that the depth is being recognised.

We publish constantly and nothing happens. What are we doing wrong?

Almost certainly breadth instead of depth. Forty articles across forty unrelated topics leave a site with no subject, and each one competes for attention with the others rather than reinforcing them.

The alternative is to pick a question your buyers genuinely have, cover it completely across a connected set of pages, link them properly, and keep them current. That is a smaller amount of writing and a larger amount of deciding.

Can we shorten this with paid media in the meantime?

Yes, and we usually recommend it. Paid search can hold the commercial terms while the organic depth accumulates, and it produces demand data that makes the content decisions better informed than they would otherwise be.

Just be clear with yourself about which budget is doing which job. The paid spend is renting presence; it stops when the payments do. The authority work is the part that still exists in three years, and confusing the two is how organic programmes get cancelled six months before they would have paid.

Find out what is actually capping your organic ceiling

Send us the five searches you cannot reach and the content you have already published. We will tell you whether the constraint is depth, coverage or consistency, what the realistic timeline looks like, and whether any of the links pointing at you are a liability.

Book a Strategy Call

If we don't deliver the work we agreed to deliver for reasons within our control, you don't pay for the undelivered work. Read our guarantee

Where to go next

Last updated September 7, 2026 · Reviewed by Zubair Afzal

Choose the subject the business must own

Map the whole question, not the keywords

Fix what already exists before adding to it

Connect the set so it reads as one body of work

Publish the one thing only you could publish

Keep it current and resist widening

Topical depth

Earned coverage

Entity consistency

A timeline nobody enjoys