What Is Holistic SEO?
The industry names a variant of this work for every context it meets. Enterprise, B2B, international, global, holistic. Most of those describe circumstances rather than disciplines. The useful version of this page explains what genuinely differs in each while saying plainly that the underlying work does not.
Most Of These Are Labels, Not Disciplines
There is no separate methodology for enterprise, for business selling, for international or for global. The work is largely identical. What changes is scale, governance and constraint. Those change what is difficult rather than what is done.
Why the labels exist anyway. They are useful to sell.
A named variant lets a supplier claim a speciality, justify a different price and imply that ordinary work would not suffice. None of that requires the variant to be real. Most buyers have no way to test the claim.
What actually differs between contexts. Three things, consistently.
How many people have to agree before anything happens. How outcomes are measured. And how many variables are in play at once. None of those is a method. All of them change what a project feels like.
Why that distinction is worth money to you. It changes what you ask for.
If the work is the same, the question stops being whether a supplier does your named variant and becomes whether they have handled your constraints before. Those are very different questions and only the second is answerable.
What this page does with the four terms. Takes each seriously.
Dismissing them all would be as unhelpful as the labels themselves. Each one points at something genuinely different. The rest of this page is what that something is.
What Holistic Actually Describes
Treating search as connected to everything else the business does rather than as a separate channel. Content, site, brand, service and offline activity as one system rather than as departments.
Why it earns its name. It describes a position, not a context.
Unlike the other three, this is not a circumstance a business finds itself in. It is a decision about how the work is organised, which makes it the only one of the four that changes what is actually done.
What the separated version looks like. Familiar.
A search agency producing content nobody in the business reads. A brand campaign the search team learns about afterwards. A site rebuild that nobody told the people responsible for rankings about until launch week.
Why separation is expensive rather than merely untidy. The channels feed each other.
Brand activity produces people searching your name. Service quality produces reviews. Offline presence produces mentions. A search programme cut off from those is competing without its own strongest inputs.
What it requires practically. Somebody who sees all of it.
Not a bigger team. One person with visibility across marketing, the site and the operation, with enough standing to be told before things happen rather than after. That is an organisational arrangement rather than a technique.
Enterprise Is Slow Rather Than Complicated
Enterprise work is sold as technically harder. It rarely is. The constraints at that scale are organisational. Confusing the two is why enterprise projects are staffed wrongly.
What actually makes it hard. Permission.
Approval chains, competing teams with their own objectives, legacy platforms nobody is allowed to touch and change processes measured in quarters. The recommendation is frequently obvious within a fortnight. Implementing it takes a year.
Why that changes who you need. Different skills entirely.
A technically brilliant specialist who cannot navigate an organisation achieves nothing at this scale. Someone competent who can build agreement across four teams achieves a great deal. Most enterprise failures are staffing decisions made on the wrong criteria.
The consequence for planning. Sequence around the constraint.
If the platform change takes three quarters to approve, plan the work that does not depend on it first rather than waiting. Enterprise plans fail by assuming a technical order when the real order is an approval order.
What genuinely is more complicated. Two things.
The number of things happening simultaneously, which makes attribution harder. And the number of people who can change something without telling you, which makes anything you fixed liable to come back.
B2B Is A Measurement Problem
Selling to businesses does not require different search work. It requires different measurement. Every serious problem in this area follows from that rather than from the work itself.
The shape of the market. Tiny volumes, enormous values.
A term searched a handful of times a month can be worth more than one searched thousands of times, because one of those searches becomes a contract. Volume-driven prioritisation gets this exactly backwards.
Why traffic reporting misleads here. The numbers are too small to read.
A programme producing a dozen genuine enquiries a year is either transformative or a failure. Monthly traffic cannot tell you which. Judging it by visits produces panic in good months and complacency in bad ones.
The buying cycle breaks attribution. Months, sometimes longer.
Somebody researches in March, gets budget in September and contacts you in January. The reporting period that receives the enquiry is not the one that earned it. Shortening the window to make reporting tidier destroys the picture entirely.
Several people, one decision. The part that changes content.
A researcher, a user, a technical assessor and somebody signing. They search different things and need different pages. A site addressing only the decision maker misses the person actually doing the looking. Our measuring performance material covers the reporting consequences.
International And Global
These are treated as separate disciplines and the distinction between them is thin enough to ignore. Both describe serving people in more than one market. The genuine considerations are the same in each case.
The first decision. What you are actually targeting.
A language, a country or both. Those are different intentions with different consequences. They get conflated constantly. That is important enough to have its own block below.
The second. Which version serves whom.
If several versions of a page exist for different markets, something has to establish which is intended for which audience. Left unsignalled, the wrong version reaches people, else the versions compete with each other.
The third. How duplication across markets is handled.
Several markets sharing a language produce pages that are near-identical by necessity. That is legitimate duplication rather than a problem. It needs signalling rather than avoiding. Our technical material covers the mechanism.
What makes this expensive when done late. It is structural.
All three decisions are about how the site is arranged, so changing them afterwards is a migration with everything that carries. Our site migrations material covers that. The practical consequence is that this is a decision to make before the build rather than after.
What is not different. Everything else.
The content still has to be good, the site still has to be reachable and the subject still has to be covered properly. Nothing about crossing a border changes any of that.
Language Is Not Country
Targeting a language and targeting a country are different decisions with different consequences. Conflating them causes more damage in international work than any technical error.
Why they are not the same. Neither maps onto the other.
A language is spoken across many countries, with different currencies, regulations and expectations in each. A country contains many languages. Assuming one determines the other misses people in both directions.
What targeting a language gets you. Reach without relevance.
One version serving everybody who speaks it. Efficient, though it cannot mention a price, a delivery time, a legal position or a local reference, because those differ between the markets it serves.
What targeting a country gets you. Relevance at a cost.
A version that can say exactly what applies where the reader is. More versions to maintain, plus more decisions about which one somebody should receive.
How to decide between them. One question.
Does anything you need to say change between the countries sharing that language. If prices, terms, delivery or regulation differ, you are targeting countries whether you meant to or not. If genuinely nothing changes, one language version is sufficient and cheaper.
The error to avoid. Building for language and writing for country.
One version per language containing country-specific claims is the worst of both. It is extremely common in businesses that expanded gradually.
Translation Is Not Localisation
People in different markets search differently for the same thing. A translated page frequently targets a phrase nobody uses, which makes this a research problem rather than a language one.
Why accurate translation still misses. Correct is not current.
A translator produces the correct term for a thing. The market may search for a brand name used generically, an abbreviation, a borrowed English word or an entirely different concept. All of those are what people type and none is what a dictionary returns.
The commonest specific failure. Borrowed vocabulary.
Many markets search using English terms for technical or commercial concepts even where a perfectly good local word exists. A faithfully translated page uses the local word and is invisible to the people typing the borrowed one.
What localisation actually involves. Research in the market.
Establishing what people there genuinely search, in their own phrasing, before anything is written. That is the same work as any keyword research and it has to happen per market rather than being translated from the first one.
The consequence for cost. More than translation.
Businesses budget translation and receive localisation-shaped disappointment. The research is the expensive part and it is the part that decides whether the pages reach anybody.
What else moves between markets. More than words.
What people worry about, what proof they expect, how directly they can be sold to. A page that reads well in one market can read as pushy or evasive in another.
What Actually Scales And What Does Not
Structure, templates and technical decisions scale. Judgement about what is worth publishing does not. That single asymmetry explains why large sites accumulate large amounts of useless content.
What scaling means here. Decide once, apply everywhere.
A template built well serves ten thousand pages as easily as ten. A technical decision made properly applies to everything. That is genuine leverage and it is why large sites can be built at all.
What refuses to scale. Deciding whether a thing is worth doing.
Somebody has to establish that a page answers a real question, that the question is not already answered elsewhere on the site and that the business has something to say. That judgement is per page and no template performs it.
What happens when the two get confused. Volume replaces decision.
An organisation with the capacity to produce a thousand pages produces a thousand pages, because the constraint that used to limit output was production rather than judgement. Nobody decided each one should exist. Nobody decided any of them should not.
Why this is a scale problem rather than a quality problem. The people are not worse.
Small teams are forced into judgement by cost. Large teams have that constraint removed and rarely replace it with a deliberate one, which is a structural outcome rather than a failure of care.
Where this leads. Two other pages.
The production method is covered in our programmatic material and the remedy in our content pruning material, both of which argue this from their own direction.
When You Genuinely Need A Different Approach
Having spent a page arguing that these are contexts rather than disciplines, here are the situations where the work really does differ enough to justify treating it separately.
When approval is the binding constraint. The enterprise case, genuinely.
Where nothing can be changed without several parties agreeing, the project is a change management exercise with a search objective. That needs different people and a different plan. It is covered further in how SEO strategy changes as websites grow.
When the site is large enough for allocation to bind. A real threshold.
Above a certain size the problem stops being what to publish and becomes which pages get attention. That is genuinely different work and it needs the structural thinking in what is information architecture in SEO.
When markets need separate research. Not separate methods.
Several markets means the research happens several times. The method does not change and the work multiplies, which is a resourcing question rather than a disciplinary one.
When a sale takes longer than a reporting period. The B2B case.
Measurement genuinely has to be built differently. That difference is real rather than a label.
Everything else. Ordinary work with a name on it.
If a supplier cannot say which of these constraints applies to you, they are selling the label. The full series is on the advanced SEO guide.
Which constraint
actually
applies to you?
Approval chains, site size, several markets or a sale longer than a reporting period. Those are the four situations where the work genuinely differs. Knowing which one you are in decides what to buy. We will tell you plainly if none of them applies.
What the national tier covers:
The national tier is £1,550 a month. It suits businesses competing beyond one town. We will say so if a lower tier would serve you better.
Every guide.
One practice.
Semantic search, machine learning, topical authority, programmatic production, rendering, log files, crawl allocation, information architecture, content pruning and how strategy changes with scale.
