For most sites, SEO takes two to three months to show rankings on low-competition searches and four to twelve months to produce significant results, according to Orbit Media's breakdown of SEO timelines. Searches dominated by strong competitors can take two to three years. So the useful version of the question is narrower. Which searches are you chasing, and how quickly can a search engine find and trust the pages you build for them?
That second part is where most of the delay hides. It's also the part you can do something about.
The timeline, broken down by competition
Orbit's Zac Harris gives three bands, and they're a sensible way to plan.
Low-competition searches take two to three months. These are specific phrases that few strong sites target, like a service plus a city or a narrow product comparison. Orbit says most brands can expect rankings here within two to three months.
Significant success takes four to twelve months. Orbit attributes this range to Google itself. It lines up with what Orbit says most agencies quote, which is four to six months to rank. That quote isn't wrong. It just describes the early end of a longer curve.
High-competition searches take two to three years. If you want to rank for a broad term where established sites already sit at the top, Orbit puts the realistic window at years, not months.
The ranges are wide because the inputs vary so much from site to site. Orbit points out there's no governing body in SEO, so the field is full of opinions and stories about sites that ranked within a few months. Treat any fixed promise with suspicion. Orbit specifically warns about agencies that told clients they would "rank in three months", and a timeline stated with that kind of certainty is a sales line.
Why the clock runs slower than you'd like
Google closed most of the shortcuts years ago. Orbit's example is meta keywords, a late-2000s tactic that once moved rankings and now does nothing. Each improvement to Google's ranking made low-effort pages less useful and raised the bar for everyone else, which stretches the time it takes a new page to earn a place.
Orbit names three things that set the pace.
Content that matches what the searcher wants. Orbit ranks content next to backlinks as one of the biggest factors in how quickly you'll rank. Word count and the number of indexed pages aren't ranking factors, so a hundred cheap articles won't speed anything up. The right length is whatever the topic needs, which you can judge by reading the pages that already rank well for it.
Backlinks. Links from other sites still matter, but Orbit is blunt that thousands of links won't rescue pages with little value.
Technical health. If Google can't crawl your site properly, Orbit says you're fighting a losing battle no matter how good the content is.
The part of the timeline you control
Here's the position this post takes. For most businesses, the slowest part of SEO is the stretch before Google has properly read your pages at all. Competition matters, but you can't change it. Whether your pages get fetched and whether they look distinct from each other are fully in your hands.
The timeline starts when a search engine fetches a page, and that can be weeks after you publish, or never. Heddle's method was built on one of our own sites, where an SEO agent and a content agent run on a schedule, and one finding from that site makes the point. Thirteen of twenty-five pages in one family were never fetched, because every page linked to the same nine siblings. Those thirteen pages had no timeline. Heddle now rotates sibling links so each page in a family gets found.
Duplication is the other quiet delay. Two city pages on one of our own sites once shared 67% of their sentences, which is exactly the kind of near-copy that gives a search engine little reason to rank either one. Heddle now measures uniqueness on the built HTML of every page against an 80% ceiling, so a page that's too close to its siblings fails before it's published.
Neither of these speeds up Google. They remove reasons for Google to ignore you, which is the only honest way to shorten the wait.
Measure from where your buyer searches
Some sites that look slow are already working, and the report is reading from the wrong place. A page on one of our own sites that a tracking tool reported at #48 was #3 when the results page was read from the city it targets. For a service firm that sells locally, that gap is the difference between "SEO isn't working" and "SEO is working and the dashboard is wrong."
If your business serves specific cities, check rankings from those cities before you judge the timeline. Otherwise you may cut work that's already paying off.
Planning a realistic first year
Plan for twelve months before you judge results, and stage your targets so you get early signal.
If you run a small or mid-sized business, start with low-competition searches you can win in Orbit's two-to-three-month band, meaning your services in the places you actually serve. Those searches tend to come from people closer to buying than broad informational queries do, and early wins tell you the pages are being read.
If you're launching a software product with no traffic, avoid aiming your first pages at the biggest category terms. Comparison and alternatives pages for named competitors are narrower and more specific. fabric.pro, a sister product, is a Heddle install taking this route. Its new site has ten comparison and alternatives pages generated from sourced datasets, though they aren't live yet, so there's no result to report.
If you're the engineer who owns the site, put the checks in the build. A duplication ceiling or a broken internal link that fails the build gets fixed the same day, while a dashboard warning can sit for months and quietly add them to your timeline.
One thing that won't change your Google timeline is answer-engine files. Heddle writes llms.txt and a Markdown twin of every page, but Google said on 15 May 2026 that Google Search ignores llms.txt and Markdown files. Build them for other tools if you want them, and don't expect them to move Google rankings.
Frequently asked questions
How long does SEO take for a new website?
A new website should plan on the same bands Orbit gives, about two to three months for low-competition searches and four to twelve months for significant results. Start with narrow, specific phrases, and make sure every page is linked well enough to be fetched, because a new site has no history to make up for pages search engines never reach.
Can SEO work in one month?
It's unlikely, since even low-competition searches typically take two to three months to rank according to Orbit. Anyone promising rankings by a fixed date is making a claim nobody can back up, because no one controls how Google ranks pages.
Why is my SEO not working after six months?
Six months is still inside the four-to-twelve-month window Orbit cites, so it may simply be early. Before you change course, check whether search engines have actually fetched your pages, whether your pages are too similar to each other, and whether your rankings are measured from the location your buyers search from.