How long does SEO take to work? Realistic timelines by starting point

Plan SEO in months, and check progress from the very first one. Google says some changes show within hours, others take months and some never make a noticeable difference. For our clients, the median wait for a first qualified organic enquiry was about 4 months. The real timeline depends on your starting point, your release dates and what "working" means to you.

Anyone who quotes you an exact month for page one is guessing. A good plan names the work and the evidence that will show progress. It also sets the dates for each decision.

What does SEO working mean?

"Working" means different things at different stages. Agree which one you mean before you start counting months.

Google's SEO Starter Guide is honest about timing. It says some changes might take effect in a few hours, while others could take several months. It suggests waiting a few weeks to judge whether work helped. It also warns that not every change makes a noticeable difference.

That guidance is about changes showing up in Search. It says nothing about when a business gets leads or recovers its costs. Those are separate clocks:

A win at one stage doesn't guarantee the next. An indexed page isn't a ranking. A ranking isn't a sale. And a technically perfect fix can still have no business effect. Here's what counts as evidence at each stage:

Long sales cycles add their own delay. A builder might get an enquiry in March, sign in May and get paid in August. So record the enquiry date, the sale date and the payment date separately. Our ROI guide shows how to measure SEO return.

Agree three things with whoever runs the money: the goal, a funding limit and a decision date. No universal percentage of revenue or fixed number of months fits every business.

What usually moves first

The order of evidence is fairly predictable, even when the timing isn't. Work goes live first. Then Google discovers and recrawls the changed pages. Impressions often appear next for specific, longer searches. Clicks follow, then broader and more competitive terms, if they come at all.

That's a common sequence, not a rule. A fix to a broken page can bring traffic back quickly. A new page in a crowded market may never reach the broad terms. And rising impressions don't mean Google is "testing" your business.

How does the starting point change the timeline?

Start with where your site is today. Each of these four starting points needs a different plan:

A new site has no track record in Search yet. It needs pages that answer real questions, a credible offer and working measurement before rankings are worth judging.

An established site may have one fixable barrier holding it back. A blocked section, thin service pages or a mismatch with what people search for are common examples. Fix the barrier and the timeline can shorten, because the rest is already in place.

A migration has a different job. The first goal is keeping what you had, from the URLs to the measurement history. Our migration guide shows how to preserve SEO during a migration.

Penalised or just declining?

Recovery cases need a diagnosis before anyone promises a date. Not every drop is a penalty. Check which one you're dealing with:

• A manual action is a real penalty. It shows in Search Console's Manual actions report, with a documented fix and review process. Even after it's lifted, former positions aren't guaranteed to return.

• A core update is different. Google's guidance on core updates says they don't target specific sites. A drop after one doesn't prove you did anything wrong. Other pages may simply be doing a better job for searchers.

Recovery takes time too. Google says it could take several months for its systems to confirm a site has improved. There's no fixed recovery date.

Competition matters in every case. A plumber in a regional town faces a different field than one in inner Sydney. The specific searches you target matter just as much.

Why agencies quote six months

Six months is a common proposal length. There's a fair reason for it. Most plans need several months of work plus time to observe the results.

But a six month contract doesn't validate a six month promise. Ask what will be delivered sooner, and what depends on you. Ask which assumptions could change the plan. And ask what happens if the agency doesn't deliver. Our agency guide shows how to choose and assess an SEO agency, question by question.

What should happen at months 1, 3, 6 and 12?

Checkpoints turn a vague wait into a set of decisions. Treat these as conversations to book in advance, not deadlines Google has to meet:

Each checkpoint has its own questions and its own decision to record:

A useful month 1 review shows four things:

• the baseline numbers, saved with their dates

• confirmed access to the site, analytics and Search Console

• measurement that records real enquiries

• the first work released, with dates

The month 3 review is where many plans wobble. It should show completed work, checks that the work is correct and a reasoned next step. "Nothing can be judged for six months" isn't good enough. But "every good campaign grows by month three" isn't supported either.

Watch for red flags at that review. Reports that show only rankings, with no released work, are one. So is enquiry tracking that still doesn't work, or a plan nobody can explain. Our guide to what the SEO work includes shows what delivered work should look like.

For our own SEO clients, the median wait was about 4 months. That's from the first fix going live to the first qualified organic enquiry. It's our record, not a promise.

What can make progress faster or slower?

SEO runs on three workstreams at once. This is our planning framework, not three Google timers:

• Technical work removes the barriers to access and use.

• Content work matches useful pages to real questions and buying needs.

• Reputation work earns credible references from other sites.

Each one has an output you can check and something it depends on. A technical fix needs a developer to release it and a live recheck. A new service page needs subject knowledge, approval, publishing and links from other pages. Outreach can't force another publisher to reply.

More staff can shorten a queue. They can't buy a recrawl, a ranking or someone else's endorsement.

Judge work from the day it went live

A common mistake is judging every page from the day the invoice started. Work released in week 11 hasn't had three months to show results. So log what really happened:

Here's a simple cohort example. A campaign starts on 1 July. Group A goes live on 8 July and group B on 16 September. At a 30 September review, group A has had about 12 weeks of exposure. Group B has had two.

Judge delivery against the agreed schedule first. Then judge results over the time each group actually had. A slow release is still a delivery problem, even if it explains thin results.

Compare like with like when you do:

• matching page groups

• enough complete weeks to see past daily noise

• the same devices and places, where it matters

Keep branded searches apart from the rest. Our Search Console guide shows how to compare search performance fairly.

How long does Google take to notice a change?

Google says a recrawl request can take days to weeks. A request doesn't guarantee inclusion, and asking again won't speed it up. So before blaming time, check the page is accessible and allowed into the index. Check it's linked from your other pages too.

Some problems can't wait at all. A broken enquiry form, an accidental noindex tag, lost access or a sudden sitewide drop needs investigating now. A noindex tag is a line of code that keeps a page out of Google. Google's guide to debugging traffic drops covers the usual causes. A long SEO horizon is never a reason to ignore a fault.

Clicks without enquiries need a different check. The fault may sit on the page, not in Search. Our conversion guide shows how to investigate traffic without enquiries.

A worked example over 12 months

Here's an example Melbourne physio practice. It isn't client data, and none of these numbers is a benchmark or forecast. Each row is one comparable month, observed at that checkpoint:

These aren't running totals, and the other months aren't shown. The enquiry ranges are examples, not confidence intervals.

The chart's scale starts at the baseline, not zero. That makes the climb look steeper. The table above has every value.

Here's the arithmetic against the baseline:

• month 3 impressions: 2.57 times (5,400 divided by 2,100)

• month 6 impressions: 5.24 times

• month 6 clicks: 4.47 times (380 divided by 85)

• month 12 clicks: 10.24 times (870 divided by 85)

Be careful what you read into that. The checkpoints don't show impressions quadrupling before enquiries rose. By month 6, both impressions and clicks were over four times their baseline. Nobody can see the exact week either crossed that line.

There's no cost data here either. So the example can't show a cost per lead or a payback date. Don't add the five rows up as if they were a year of enquiries.

Several things changed over the year, too. In this example log, months 1 and 2 fixed technical faults and the booking form.

Months 2 to 5 added service pages for each treatment. The practice also tidied its Business Profile and kept replying to reviews. The example shows how to monitor progress. It doesn't isolate what caused the growth. Nor can it predict another practice's results.

How long does local SEO take?

There's no fixed number of months. Local SEO has two surfaces that move separately: the Maps listing and the website. Neither is guaranteed to produce the first or cheapest leads.

Google ranks local results on relevance, distance and prominence, according to its local ranking help. Distance can't be bought with a bigger SEO fee. And review velocity isn't a confirmed ranking factor, so ignore anyone selling a review quota. Our guide to what affects Maps visibility covers the factors.

Track the two surfaces apart:

• profile calls, direction requests and website clicks from the listing

• search clicks to the website

• qualified enquiries, by suburb or service area

A strong Maps position in one suburb isn't coverage across a city. Our profile guide shows how to manage your Business Profile.

Several locations

Plan each genuine location on its own. Start with an inventory:

• the real premises or service arrangements

• who keeps each location's details right

• the pages and profiles that already exist

Then plan each location in turn:

Don't create extra profiles to target suburbs. A shared website platform can cut repeated work. Each location still needs its own facts and its own approvals.

Industries differ too, but not by a fixed number of months. An urgent trade job and a considered professional service sell differently. A big online catalogue differs again. Regulated sectors may also need claims reviewed by a specialist before anything goes live.

What can budget and the team change?

Money and people change how fast work gets done. They don't change how fast Google responds:

The business has its own part to play while the work runs. Approve priority content, supply accurate service details and test the enquiry route yourself every few weeks. Record the quality of each enquiry, and unblock access when it's needed. Give every dependency an owner and a due date.

The delivery model changes the shape of the queue. An agency brings breadth and continuity. It still needs the owner's input on facts.

In house staff know the business, but may lack time or specialist skills. DIY costs the least cash. It costs the most of the owner's evenings.

Set the funding limit before the work starts. Here's an example allowance, in Australian dollars before GST:

That's six monthly fees of $2,000, plus a one off $1,500 for implementation. Internal staff time sits outside it. It's a sample outlay, not a price benchmark or a sufficient budget for every business. If delays stretch the plan, revisit the funding decision before adding months. Our pricing guide shows how to plan the cost of SEO.

How do seasons and maintenance affect the plan?

Work backwards from when customers decide, not from the busy month. A common mistake is starting seasonal pages when the busy month arrives. By then, the research window may have passed. An example heating business wants its pages ready before winter research starts. Bookings come after that.

So set dates for content, approval, release and observation, with room for delays. Say enquiries usually build from May. The pages should be live and checked weeks before that. So approvals need to happen earlier still. There's no universal rule like starting six months before the peak.

Compare equal periods. Use the same season last year too, where you have it. Note public holidays, promotions, weather and any capacity limits. Google Trends can show whether wider interest in your service rose or fell over the same period. It isn't a record of your sales, though.

Maintenance isn't a growth curve

Useful pages and internal links can keep helping users. But growth isn't automatic, and content doesn't earn trust just by getting older. Demand, offers and competing pages all keep changing. So keep reviewing:

Review a page when its facts or services change. Do the same when its search intent drifts, a conversion route breaks or performance shifts for weeks. No evidence supports a compulsory refresh every 12 to 18 months. And an updated date needs a real update behind it. Our content guide shows how to review your content strategy.

Do provider changes or shortcuts reset progress?

Changing provider doesn't reset Google's history. Losing access or breaking pages during the handover can still undo real work, though.

A clean handover preserves:

• the domain and the CMS

• analytics and Search Console access

• live content and URL maps

• open fixes and the change log

The new provider should learn the baseline before replacing anything that works.

Shortcuts don't buy time either. Bulk content and bought links can't purchase an exact timeline. Google's spam policies cover manipulative links and scaled low value content, however it's produced. AI assistance alone isn't a breach. But a predictable temporary boost is a sales pitch, not a plan.

When should you wait, investigate or change course?

Every review should end with a decision. Start by naming the situation you're in:

Then follow the checks in order:

An inconclusive result isn't a pass. Record the next observation, who's responsible, the review date and the most you'll spend before then. Neither "wait" nor "continue" means unlimited funding.

A pause can be the right call when cash, operations or the offer needs attention. Keep access, content and essential maintenance going. A pause doesn't wipe out what you've built. Staying with a weak plan isn't automatically cheaper either.

If a Google update is suspected, check the official dates and the affected page groups. Google recommends waiting at least a full week after a core update finishes before analysing it. That doesn't stop you fixing a separate broken page today. Our algorithm updates guide shows how to investigate a Google update.

Do Ads and AI search follow the same timetable?

Google Ads can bring paid visibility sooner, once an account is approved and live. Budget, auction demand and conversion quality still decide whether it pays. Organic visibility can keep going without paying per click. But the website and content still cost money to maintain.

Compare channels on the same terms. A crossover date only means something with like for like costs and outcomes:

To compare SEO and Google Ads in depth, see our comparison guide.

AI answers can change how many clicks a ranking brings. They don't change how long SEO takes. There's no fixed timetable for AI visibility either. Rankings, AI citations and enquiries are separate outcomes. Our guide to how AI search changes visibility covers the bigger picture.

Set a plan for your business

Five questions set a realistic plan:

• What outcome matters most?

• Where are you starting from?

• What work must go live, and who can release it?

• What evidence and comparison period will you use?

• What can the business fund before a decision is due?

Our estimator turns those answers into a rough planning range you can test against real dates. Treat its range as a starting assumption, not a forecast.

What we recommend at Elev8d

FAQ

Next steps

Written by Ajay K.Co-founder, Elev8d

Ajay is the co-founder of Elev8d and writes most of what you will read here. Psychology grad turned marketer, which is a fancy way of saying he cares about why people click, not just where things rank. Every guide is written in plain English for Australian businesses, tested on real campaigns before it earns a spot on this site.