SEO stands for search engine optimization and names an outcome rather than a product: a web page being found, read and chosen when somebody searches. The build makes a page findable and readable, the business’s own text makes it answer something, and the search engine decides whether it is shown. One SEO line in a quote puts all three owners in a single cell.
What SEO is, what it stands for and what it does
SEO names a result, not a product to install, license or switch on, and not a task that one supplier performs and ticks off: a particular page being found, read and then picked out from the alternatives for a particular search. The everyday answer to what it is for, putting your page in front of the person looking for what you do, is true as far as it goes, yet it hides the fact that the result is made of three separate things.
As for the letters themselves, they abbreviate search engine optimization. Typed alone into a search box, they also pull up personal names and place names that have nothing to do with websites, so it helps to say which sense this article takes: web pages, and the results a search engine returns for them.
Each of those three has its own owner, and the search engine is the easiest to name: it alone decides which page gets chosen. Whether a page can be found and read is settled by how it is built, so that falls to the development team, while whether it answers a question is settled by what it says, so that falls to whoever owns the content, which in practice means the business. Even where one person writes both the code and the copy, the jobs stay separate and the choosing still happens elsewhere.
So “getting the SEO done” is an order with no single addressee. What a page actually possesses is an address, a title and some text, the material a search engine works with, and SEO is not among its parts the way a contact form or a footer is. Rather than asking whether a site’s SEO has been done, ask who holds each of the three jobs and which of them has really happened.
A search engine does three things with a page: finds it, reads it, shows it
Everything in this section rests on one document, quoted rather than paraphrased so that each sentence can be checked against the original: Google’s guide to how Search works. It describes Google’s own system, and nothing here assumes that every other engine behaves identically. One line in the guide gives the shape of the whole process: “Google Search works in three stages, and not all pages make it through each stage”. It names those stages crawling, indexing and serving search results.
Followed through those stages one at a time, a single page is first discovered by the engine and downloaded, then has its title and text processed and filed away in the engine’s index. At the moment of a search the engine goes back to that index; in Google’s own words, “When a user searches on Google, Google returns information that’s relevant to the user’s query.”
Before it describes any stage, the guide also sets out what Google will not promise: “Google doesn’t guarantee that it will crawl, index, or serve your page, even if your page follows the Google Search Essentials.” Indexing gets a warning of its own further down: “Indexing isn’t guaranteed; not every page that Google processes will be indexed.” Those two sentences govern everything this article says about promises. A missing guarantee is the engine’s own published position, so a supplier who declines to offer one is repeating the engine rather than making excuses.
The owners come into view once you ask what each stage consumes. Serving takes nothing from the page directly; which pages come back is settled inside the engine. Crawling and indexing, by contrast, feed on the page itself: its address, its title, its text. Who controls those inputs is no mystery. The builder decides whether the address and title come out in a form a machine can read, and the content owner decides what the text says. You can inspect both from the page’s side whenever you like, while what the engine does with them stays out of view.
The first job: being found and read, which the build owns
Whoever signs off the acceptance can test the build’s share of SEO on handover day without any tools. The browser tab should name the page being viewed rather than repeat the company name, a link to an inner page sent to a colleague should open that same page on their screen, and the address in the browser bar should differ from one page to the next.
Those checks probe the inputs to crawling and indexing, which are also the only share of SEO that a quote can both promise and let you verify at acceptance. The least visible input is content that reaches the browser already assembled by the server, as in Plastik ve Makina’s results list. The other two you have just looked at: a permanent address for each page, and a title written for that page rather than for the whole site. Here the build’s share also stops. Addresses, titles and server output are what a development team hands over, however good the code; the answer to your customer’s question was never part of what it could deliver.
Çatalca Merkez Kasap’s menu project shows what the address part looks like in practice: send someone the link to one section and it opens there, not at the top of the menu. Why a section earns an address of its own is argued in our piece on replacing printed catalogues and menus with software, and that argument is not repeated here. The unit of address is not always a page, and in that menu it is a section. How the unit gets chosen belongs to that article.
When all of this gets settled matters as much as what it contains. One of the FAQs on our web development service explains why findability is decided while the pages are being built, rather than added as a package once they exist; this article simply takes that conclusion as given.
The second job: answering a question, which the business owns
Our piece on where an unanswered customer question goes when your own page stays silent follows a customer’s question to an unchecked source in the search results, where it is unclear who wrote the answer and nobody stands behind it, in an exchange the business never sees: the second job, left without an owner.
When an answer on a page turns out wrong or goes out of date, the business is the one that has to correct it, even if a developer typed every sentence. The facts were always the business’s: what it sells, in what condition and on what terms; the questions about them already arrive every day by phone, by email and through the contact form. A page the development team writes from start to finish out of facts the business supplied is no exception, since a wrong detail on it is still noticed and fixed by someone at the business. Ownership follows the origin of the facts.
Plastik ve Makina, a classifieds catalogue where sellers write their own listings, is organised on exactly that basis, with the platform supplying the catalogue to sellers who know their machines. That split also marks the limit of what a build can contribute. Planning a site around what visitors come looking for gives each answer somewhere to sit; writing the answer, in the business’s own words, remains the business’s work.
Our guide to the website contract clause that names who updates content after launch shows how the person responsible for keeping each page current gets written down during planning, before the build begins, rather than worked out after launch. The reason is that answers age: once a price changes, a page that was right on the day it went live is wrong from then on, with no code touched at all.
The third job: being chosen, which nobody can promise
Neither a development team nor a content owner, however skilled, can commit to this job, and the reason is mechanical rather than moral. Both of them supply inputs, since the first two jobs end in the page itself. Which page is chosen is the engine’s output, and nobody who controls only the input can commit to an output decided on another side.
Google’s guide confirms the point from the engine’s end. It opens its section on serving results with this: “Google doesn’t accept payment to rank pages higher, and ranking is done programmatically.” Which page appears for a search is decided by the engine’s own systems, and it is not for sale. A promise about the third job therefore describes a decision the person making the promise does not control.
We drew the same boundary for AI search in what makes a page citable by AI search, which measures only what can be checked on the page’s own side and promises nothing beyond it, and it holds for an ordinary results page as well.
Advertising slots, which can be bought, sit apart from the organic results and lie outside this article. On the organic side, a limit we apply to someone else’s quote has to hold in our own text before anywhere else, so we hold ours to it: the web development and corporate website pages promise pages that are findable and readable, and none of them names a position, a first page or a ranking.
The three jobs side by side: who owns each, where it shows, what the quote says
Read the table below as a map of who owns what rather than as a ranking: one job per row, with its owner, where it becomes visible, whether anyone can promise it, and an example from our own portfolio. One example cell stays blank on purpose. No case study of ours shows a page being chosen, and none could, since that choice is never made by whoever built or wrote the page; a project name there would claim a decision that was never ours.
| Job | Owner | Where you can see it | Can it be promised? | Portfolio example |
|---|---|---|---|---|
| Being found and read | The development team | At handover: each page’s own address, its own title, and content arriving ready from the server | Yes; you can check it at acceptance | Address side only: in Çatalca Merkez Kasap’s single-page QR menu, a shared section link opens that section |
| Answering a question | The content owner, meaning the business | After launch: whether the text answers what customers ask, and whether the person who updates it is named in the plan | Yes, but the business gives that promise, not the development team | Plastik ve Makina’s sellers write their own listing text |
| Being chosen | The search engine | Nowhere on the page’s side; the decision is taken inside the engine | No; the engine’s own documentation does not guarantee serving | Left empty on purpose |
Down the owner column, each job lands with a different party and no name repeats. The promise column is where a quote can slip. Two of the jobs can be promised, though not by the same people: the development team answers for being found and read, the business for answering. A developer’s proposal that promises the content row has handed the promise to someone who cannot keep it.
A single SEO line in a quote puts three owners on one row
When “SEO” sits on a quote as a single line, three owners share one row and the price gives no hint which of them it belongs to. A line that covers only the first job, and says so in writing, is perfectly readable. The difficulty begins when scope and owner are both left out.
Begin with the part that cannot be bought, because it is the quickest to settle and settling it accuses nobody. Any promise of a position, or of a guaranteed appearance in results, leaves the line unreadable: Google’s own guide neither guarantees serving nor accepts payment to rank pages higher, so that part of the price has nothing to buy and the line’s price comes apart from its work.
The first job is the one you can watch arrive. If the line includes it, acceptance day should show every page at its own address, under its own title, with its content coming from the server.
A line that names neither what the business will provide nor who keeps the pages up to date once the site is live has left the second job with nobody, even though the business assumes it has paid for a share that no supplier could carry out alone. Covering the job means writing both of those things down.
With an SEO line, the reading that goes missing is which of the three owners it was written for. That is one case of a wider pattern, which why software quotes differ, where a line’s name hides the reading behind it traces across an entire quote. Our own quote form leaves the question with you: SEO is not among its project types, and its scope step asks you to describe the work in your own words.
Take the SEO line in the quote you are holding and ask two things of it: which of the three jobs it covers, and whether the person who writes the content and keeps it current after launch is named. In our corporate website projects the first job is not a separate line; it comes inside the page plan, and the same plan names who updates which page.
