Most SEO advice for new websites is written for sites that already have some authority. It tells you to publish consistently, build links and track rankings, which is all true and none of which explains why a brand new site can do everything right for two months and see almost nothing.
This is a ninety day plan with honest expectations attached. If you have not set the site up yet, start with what to do after you buy a domain, which covers hosting, SSL, Search Console and the rest of the groundwork. This picks up from the moment the site is live.
Why a new site is slow, and what is actually holding it back
A new domain has almost no links pointing at it. Search engines use links both to find pages and to judge whether a site is worth ranking, so a site with none starts with two problems at once.
Content does not fix this on its own. You can publish genuinely good work for months and watch it sit on page five, because the limit is not how useful the page is but how little reason a search engine has to trust the site carrying it.
That is uncomfortable but it is useful, because it tells you where effort belongs. A plan that is entirely publishing, with nothing that earns a mention from another site, is working on the wrong variable.
Days 1 to 30: get found, and find out what you can win
The first month is about being crawlable and choosing targets you can actually reach. Rankings are not the goal yet and measuring them will only discourage you.
- Confirm Google can reach the site on a phone. Since 5 July 2024 every site is crawled with the mobile Googlebot, and content that is not reachable on a mobile device is not indexable at all. Source: Google Search Central.
- Do not worry about crawl budget. Google aims that guidance at sites over a million pages changing weekly, or over 10,000 pages changing daily. A new site is nowhere near it. Source: Google Search Central.
- Pick keywords by looking at the results page, not the difficulty score. If the top ten are all large brands, or all software, you cannot win it yet. Reading a SERP properly takes two minutes and saves months.
- Publish five to ten pages in one narrow topic. Depth in one area beats one page each across ten areas, because a cluster gives you something to link internally and a reason to be seen as being about something.
- Write down your baseline. Zero clicks, zero impressions, however many indexed pages, with today’s date. You will want this later and Search Console only keeps 16 months.
Days 31 to 60: fix indexing before anything else
The most common second month problem is not ranking. It is that pages are not in the index at all, and no amount of optimisation helps a page Google has not stored.
Open the Pages report in Search Console and look at the reasons under “Why pages aren’t indexed”. Two of them matter most on a new site.
- Discovered, currently not indexed means Google knows the URL exists and has not come to fetch it. Our guide to discovered, currently not indexed covers what moves it.
- Crawled, currently not indexed means Google fetched the page and chose not to store it, which is a judgement about the page rather than an access problem. The fix is covered in crawled, currently not indexed.
Also check the indexed count against the number of pages you have published. If you have twenty pages and six are indexed, that gap is the whole problem and nothing else is worth touching until it closes.
This is also the month to start earning the first few mentions. Not link schemes, just the ordinary ones: a directory that genuinely lists businesses like yours, a supplier page, a profile somewhere real.
Days 61 to 90: read the early signals
By the third month there should be something to read, as long as you are reading the right thing. Clicks are the wrong metric this early.
- Impressions. These move first and move far more smoothly than clicks. A rise from 40 to 400 impressions is real progress even if clicks stay at zero.
- Query count. The number of distinct queries you appear for matters more than position. One query at position 20 is fragile, forty queries at position 60 means Google has found a topic it associates with you.
- Average position, by page. A page moving from 80 to 45 produces no traffic and is a genuine signal.
- Indexed pages. Still the most important number if it is not keeping pace with what you publish.
What you should not expect is meaningful traffic. If you get some, good, but a plan that requires it at ninety days is a plan that will be abandoned at day ninety-one.
What a realistic ninety day result looks like
On a new domain with no link profile, a reasonable outcome is a few hundred to a few thousand impressions a month, most pages indexed, average position somewhere between 40 and 70, and clicks in the single or low double digits.
That is not a disappointing result, it is the normal shape of the first quarter. The sites that appear to do better in public case studies almost always had something extra at the start, such as an existing audience, an aged domain, or links from a parent company.
If you need a number for a budget conversation, build a range rather than a point estimate, the way our SEO forecasting guide sets out, and write the assumptions next to it.
Mistakes that cost new sites the most time
- Checking rankings daily. At this stage positions bounce for reasons that have nothing to do with you, and watching them produces anxiety rather than information.
- Targeting head terms. A new site that targets the two-word version of its topic is competing with every established site in the market at once.
- Publishing across unrelated topics. It spreads thin authority thinner and gives you nothing to link together.
- Buying links early. A new domain with a sudden burst of paid links is the easiest pattern in the world to spot.
- Rewriting pages every few weeks. Give a page long enough to be crawled, indexed and settled before judging it. Changing it constantly resets your own evidence.
- Treating volume as the plan. Doubling output does not double results when the constraint is trust rather than coverage.
What to do at the end of the ninety days
Compare against the baseline you wrote in month one, not against what you hoped for. Impressions up and indexing healthy means the approach is working and needs time, not a change of direction.
Impressions flat with pages indexed means the content is being seen and ignored, so the targets were too competitive and the next quarter should go narrower.
Pages still not indexed after three months is the one result that demands a different plan, because it says the site is not yet considered worth storing, and that is an authority problem rather than a content one.
Frequently asked questions
How long does SEO take for a new website?
Expect very little in the first three months, and treat impressions rather than clicks as the early signal. A new domain has few links, which limits both how often it gets crawled and how much a search engine trusts it, and neither of those changes quickly. Meaningful traffic usually takes longer than a quarter, and how much longer depends mostly on how fast the site earns genuine mentions elsewhere.
Why is my new website not showing up on Google?
Check indexing before assuming it is a ranking problem. In Search Console, “discovered, currently not indexed” means Google knows the URL and has not fetched it, while “crawled, currently not indexed” means it fetched the page and chose not to store it. If your indexed count is well below your published count, that gap is the problem and nothing else is worth working on first.
How many pages should a new website publish first?
Five to ten pages inside one narrow topic beats the same number spread across unrelated subjects. A tight cluster gives you pages to link together and a clear subject for a search engine to associate with the site. Breadth can come later, once something in that first cluster starts earning impressions.
Should a new site focus on content or links?
Both, but a plan containing only publishing will stall. Content gives a search engine something to rank, and links give it a reason to trust the site enough to rank it, so neither works alone on a brand new domain. If you only have time for one thing in the first quarter, make it indexing, because an unindexed page cannot benefit from either.
What should I measure in the first three months?
Impressions, the number of distinct queries you appear for, average position by page, and indexed pages against published pages. Clicks are too noisy at low volumes to tell you anything useful. Record a dated baseline at the start, because Search Console keeps only 16 months of Performance data and you will want the comparison later.
Does a new website need to worry about crawl budget?
No. Google aims its crawl budget guidance at sites with over a million pages changing weekly, or over 10,000 pages changing daily, and a new site is far below that. If pages are not being crawled, the cause is usually too few links pointing at the site rather than any budget limit.