Ranking a new website: the first 90 days, step by step

A freshly launched website starts from nothing: no history, no inbound links, no accumulated trust. That makes the first 90 days important, though not in the way many people expect. This is not the period in which you reach page one; it is the period in which you lay the foundations and find out whether Google is understanding your site. What follows is the sequence we use, in order of priority, with no promises of results by a fixed date.

Days 1-7: make sure Google can get in

Before writing a single line of content, check that nothing is blocking access. It sounds trivial, but a site left with the "discourage search engines" option switched on after development is a classic. Check, in this order:

• robots.txt with no Disallow: /, and no noindex tag on the pages you want found;
• HTTPS working, with a single version of the site (with or without www) responding and the others redirecting with a 301;
• an XML sitemap published and submitted in Search Console;
• Search Console connected: without it you have no real data on what Google sees.

For a quick automated check of these points you can use our free SEO analysis: it returns a prioritised list, with no sign-up.

Weeks 2-3: decide what you want to be found for

A new site cannot compete on everything. Choose a handful of concrete searches, the ones a real customer would type when they have the problem you solve, and assign each to one specific page. Two practical rules:

• one page, one topic. If two pages answer the same question, they compete with each other;
• a specific search beats a generic one. "Consultant for a clothing e-commerce shop in Bergamo" is more reachable than "marketing consultant", and brings enquiries closer to a purchase.

You do not need expensive tools to begin. The related searches at the bottom of a Google results page, and the questions customers actually ask you on the phone, are already a good starting point.

Weeks 3-6: the pages that bring enquiries

This is where you write the core pages: services, about, contact and, if you cover several locations or sectors, one page for each. Every page needs a title and description written to be clicked, a single clear H1, and copy that answers what the visitor wanted to know: what you do, for whom, on what terms, and what to do next. A long text is not worth more than a short, complete one; what counts is the one that settles the doubt.

Link the pages to one another. Internal links are the simplest way to tell Google which pages matter: from the homepage to the services, from the services to the supporting pages and back. A page that no other page links to can be ignored for months.

Weeks 6-10: content that answers real questions

Only now does it make sense to open a blog or a guides section, and only if you have something to say that is not already on ten other sites. A useful article starts from a real customer question, gives an honest answer (even if it is "it depends, and on what") and leads to the relevant service page. A few good articles beat many filler ones, and text produced in bulk and published without review is exactly the kind of content likely to stay "discovered, currently not indexed".

If you also want to be cited by AI-generated answers, the foundation is the same, with a little more care over structure and clarity: we cover it on the page about SEO for AI.

Weeks 10-13: measure, don't hope

At three months, read your Search Console data realistically. What is reasonable to expect:

• most of your important pages are indexed;
• the first impressions appear, often on very specific searches and at low positions;
• a few queries start to climb, and that is where you decide where to invest next month.

What is not reasonable to expect: first place for competitive searches. For a site with no history, the slowest part is authority, meaning links and mentions from other credible sites, and it cannot be hurried by buying links: it is built with useful content, listings in serious industry directories, partnerships and genuine mentions. Anyone promising page one in 30 days is promising something nobody can guarantee.

What you can do yourself and what is worth delegating

Almost all the work of the first 90 days is within reach of anyone who knows their own business well: choosing the searches, writing the service pages, answering customers' questions. It is worth getting help with the technical side (redirects, sitemap, structured data, speed) and with the final check, because these are errors you cannot see by eye and that cost months if they stay. A good consultant, at this stage, hands you a priority list with the reason for each item and leaves you to do what you know better than they do. Be wary of anyone proposing a monthly retainer before they have even looked at the site: at the start you need an analysis, not a subscription.

A concrete example of a calendar

For a local business with three services, a realistic calendar is: week 1, technical checks and Search Console; week 2, choosing three to five main searches; weeks 3-5, one page per service plus a contact page with address, opening hours and map; week 6, your Google Business Profile and consistent details (name, address, phone) wherever they appear; weeks 7-10, two or three articles answering the most frequent questions; weeks 11-13, reading the data and making corrections. It is not the only possible calendar, but it has the merit of being checkable: at each stage you know whether you have finished it or not.

The mistakes we see most often in the early months

• changing titles and URLs every week, giving Google no time to assess them;
• copying descriptions from suppliers or competitors;
• duplicating the same page for ten cities and changing only the name;
• ignoring mobile, where most of the traffic arrives;
• judging the results after three weeks.

In short

Week 1: no technical blocks and Search Console live. Weeks 2-6: a few well-chosen searches, one page for each, coherent internal links. Weeks 6-10: content that answers real questions. At 90 days: measure, adjust, and do not expect page one on competitive searches.

If you want to know where your site stands, start with the free SEO analysis; if you would rather talk through a plan for your case, get in touch. The cost of each option is set out on the pricing page.