Skip to content
search-engine-optimizedby Netsolution

Where to start on a new site

On a new site there is almost nothing to do in order to be found: publishing is enough in the great majority of cases. Your time is better spent checking that the pages are reachable, writing two or three genuinely useful pages, and waiting long enough to judge without fooling yourself.

Do you have to do anything to be found?

It is everybody’s first question, and the documented answer surprises almost everybody.

There is generally nothing to do to be found beyond publishing your site: the great majority of sites present in the results were discovered and added automatically.

officialSource : Google Search Central, “SEO Starter Guide”

So there is no registration to complete, no form to fill in, no entry-to-the-index service to buy. A site that is published, reachable and linked from somewhere will be discovered. That promises nothing about what comes next — being discovered, being crawled, being kept and being shown are four distinct things — but it settles the question of getting started, permanently.

Claiming your site in Search Console stays worthwhile, for a reason that has nothing to do with discovery: it is what gives you access to the queries actually recorded. You sign up to obtain data, not to obtain a place.

An archival print mounted on card: a foundation site, timber formwork and steel reinforcement in place, the caption and date printed under the image.
The moment on a building site when nothing yet shows above ground, and where what the building will carry is nonetheless decided. The photograph carries its date on itself, which is the second useful thing here.Stuart D. Pike, Metropolitan District Water Supply Commission — Domaine public · Wikimedia Commons, relevé le

How long before you see an effect?

The second source of waste on a new site, after the imaginary registration: impatience.

Some changes take effect within hours, others can take months; waiting a few weeks before assessing a change is advised.

officialSource : Google Search Central, “SEO Starter Guide”

The gap between “a few hours” and “several months” is not vagueness, it is the information itself: no duration can be announced, so any promise of a delay is invented. And to that you have to add the background noise, also documented:

Small position changes happen at any time, upwards included, with no action needed: results are dynamic by nature.

officialSource : Google Search Central, “Debugging drops in Google Search traffic”

Those two sentences, taken together, condemn the most common practice of the early days: change a page, look three days later, conclude, change again. Every turn of that cycle destroys any chance of knowing what produced what.

On a site with no history, most decisions taken during the first weeks are judged too early to be judged at all, which pushes people to fix things that were not broken.

hypothesis

Which three checks are worth making, and in which order?

One. Are the pages reachable? Not “does the site display on my machine”, but: a page served without authentication, without a block in the robots exclusion file, without a no-index instruction left behind by mistake. That last error is the most frequent one on a new site, because the staging environment carries one and the launch forgets to remove it.

Two. Does each page have a stable address and a title that describes its content? An address you are going to rename three times in a month costs more than every optimisation you could apply to it. A title that genuinely describes the page serves the reader before it serves anything else.

Three. Are there two or three pages that answer a real demand seriously? It is the only one of the three that asks for work rather than verification, and it is the one that decides the rest.

Checking that pages are reachable before any optimisation

hypothesis

Effort
low
Time to effect
immediate
Risk
none
Durability
structural

Nothing on that list guarantees an outcome, and it is better said plainly than left to hope:

Indexing is not guaranteed: not every page the engine processes ends up in the index.

officialSource : Google Search Central, “In-depth guide to how Google Search works”

These checks remove the avoidable obstacles. They do not make a page deserving.

What does not deserve your time at launch?

A blunt list, because this is where most of the energy of the first months goes.

Fine-tuning tags on a site nobody visits yet. Paid tools before you have any data to compare them against. Automated audits producing dozens of alerts, not one of which describes a real problem. Daily position tracking of figures that, as we saw, move on their own. And producing volume, whose supposed effect is explicitly denied:

Content length on its own plays no part in ranking.

officialSource : Google Search Central, “SEO Starter Guide”

What does not work, or no longer works

Publish thirty articles in the first month to show Google your site is active, then watch your positions daily and adjust.

Publishing pace does not buy attention, and indexing is not guaranteed in any case. As for daily tracking, it measures variations the documentation itself describes as normal: you read signals in noise, then correct at random.

What is left, once all that has been taken away, is unspectacular and fits in three lines. Publish reachable pages. Write a few that genuinely answer a demand you know how to handle. Wait long enough for what you observe to mean something, then look at the queries actually recorded rather than at positions collected by hand.

It is a disappointing programme to read and a far cheaper one to follow than its alternative. Its main quality is that it asks you to believe nobody: each of the first four decisions can be checked in a public document, and the fifth — waiting — costs nothing but patience.

Sources

  1. SEO Starter Guide: The Basics — Google Search Centralaccessed on
  2. Debugging drops in Google Search traffic — Google Search Centralaccessed on
  3. In-depth guide to how Google Search works — Google Search Centralaccessed on
  4. Performance report (Search results): Overview and basic setup — Google Search Consoleaccessed on

Block revalidated on

Frequently asked questions

Do you have to submit your site to Google to be listed?
No, in the great majority of cases. The official documentation says publishing your site is generally enough, discovery being automatic. Claiming your property in Search Console stays useful, but for receiving data rather than for being found in the first place.
After how long can you judge a new site?
A few weeks at the very least, and the official source recommends that span before assessing a change. Some changes act within hours, others over months. Judging after ten days means deciding on noise, then fixing what was working.
How many pages should you publish at launch?
Fairly few, but genuinely useful ones. Two or three pages that handle a demand seriously beat twenty produced to fill a sitemap. Quantity shows up when you read, and it never makes up for having nothing to say.
Do you need a blog from day one?
Only if you have something to write there regularly. A blog abandoned after four posts does not harm a site, but it does not help it either, and it eats the time that should have gone into the pages answering real demands.