A manual action is the worst message you can open in Search Console. It means a human reviewer at Google has looked at your site and decided it breaks the rules. For the site in this case study, it meant disappearing for the one search term the entire business ran on.
Within six months of rebuilding, that site was back at position one for its core commercial term. Today it produces close to 200 leads a month.
This post walks through how I did it: the decision that mattered most, the rules I followed to avoid bringing the penalty with me, and the SEO strategy I ran from a standing start. The niche is a high competition agency space, and I have kept it anonymous for confidentiality. The process is what matters, and it applies to any business.
Key takeaways
- The first decision is recover or restart. Getting it wrong costs months.
- A penalty lifted only gets you back to zero. Either way, you are rebuilding.
- If you restart on a new domain, bring nothing across that could carry the penalty with it.
- Build the site around the money keyword from day one: technical foundations, then content, then authority.
- Measure position, traffic and leads. Nothing else tells you if the recovery is working.
What a manual action actually is
Most ranking drops are algorithmic. Google's systems reassess your site, and nobody tells you anything. A manual action is different: a person at Google has reviewed the site, and you get a notice in Search Console's Manual actions report saying what the problem is and whether it affects some pages or the whole site.
Commercially, the effect is brutal. The site in this case ran on one core term. Ranking for it meant a steady flow of enquiries. Losing it meant the organic channel went from the business's best lead source to effectively nothing, overnight.
If you think you have been hit but have no notice in Search Console, you are probably dealing with an algorithmic drop instead. My breakdown of the September 2026 spam update explains how to tell the difference.
Source: Google's Manual Actions report documentation, which lists every type, from unnatural links to thin content to pure spam.
Decision one: recover or restart?
There are two routes out of a manual action.
Route A: clean the original site and submit a reconsideration request. You fix whatever the notice describes, document it, and ask Google to review the site again.
Route B: rebuild on a new domain. You start clean, with none of the history that caused the problem.
Most guides only cover route A. But route A has a catch people do not mention: even when a reconsideration request succeeds, you do not get your old rankings back. You get permission to compete again. Whatever was propping up the old rankings is usually what caused the penalty, so once it is removed, the site starts close to zero anyway.
Here is how I weigh the two:
| Factor | Points to recover (Route A) | Points to restart (Route B) |
|---|---|---|
| Cause of the penalty | Narrow and fixable, like a batch of thin pages | Deep in the site's history, like years of manipulative links |
| Brand value of the domain | Strong direct traffic, brand searches, offline recognition | The domain name itself does not carry the brand |
| Time to clean | Weeks | Months of link removal and cleanup, with no guarantee |
| Certainty | High chance the request succeeds first time | Real risk of rejection and resubmission cycles |
| What is left after | A strong site underneath the penalty | Little worth keeping once the violating parts are gone |
For this site, a new domain was the right call. Rather than spending months cleaning and waiting on reviews, with the old rankings not coming back regardless, I put the time into building something with a clean record that I controlled from the first page.
If you are weighing this for your own site, be honest about the right hand column. Owners often cling to a domain out of attachment when the numbers say restart.
Decision two: do not carry the penalty with you
A new domain only works if it is genuinely new. The rules I work to on any restart:
- No 301 redirects from penalised pages. A redirect passes signals, and that can include the problem you are escaping. If you need people to find the new site, point them there with messaging, not with redirects from the penalised URLs.
- No reusing the old link profile. Do not ask the sites that linked to the old domain to switch their links wholesale, and never move any link you are not certain is clean.
- No copying content across. If content contributed to the penalty, moving it just moves the problem. Everything on the new site gets written fresh.
- No repeating whatever caused it. Obvious, but it is the most common reason a restart fails. The tactics that got the old site penalised do not become safe on a new domain.
The new site starts with no authority, but also with no debt. That trade is the whole point.
The strategy: rebuilding from a standing start
From there, it was my own SEO strategy, run in a deliberate order. Each stage depends on the one before it, and skipping ahead is how rebuilds stall.
Stage 1: technical foundations from day one
Retrofitting technical SEO onto a messy site is slow. Building it in from the first page is fast. Before any content went live, the new site had:
- Clean, logical URL structure planned around the keyword map
- Correct indexing controls, so only pages worth ranking were indexable
- An XML sitemap submitted to Google and Bing from launch
- Fast load times and a fully responsive build
- Organisation schema so Google understood the brand as an entity from the start
- Search Console and analytics tracking conversions from day one
The last point matters more than people think. If you cannot attribute leads from the first week, you cannot prove the recovery to anyone, including yourself.
Stage 2: a keyword map built around the money term
The whole business ran on one core commercial term, so the site architecture was built to win it.
- The money page targets the core term, and every structural decision supports it
- Supporting service pages target the close variations and related services, each with a distinct intent so they do not compete with each other
- Supporting content answers the questions buyers ask before they are ready to enquire, and every piece links back to the money page
This is the topic cluster model I explain in my SEO for small business playbook. On a new domain it is even more important, because internal links are the only authority you control at the start. Every page on the site existed to make the money page stronger.
Stage 3: content that proves expertise
In a high competition niche, every competitor has pages targeting the same term. Mine had to be better on substance, not just on keywords. That meant:
- Answering the searcher's actual question immediately, not after three paragraphs of intro
- Showing real expertise: how the service actually works, what it costs, what results look like
- Covering the objections buyers raise before they enquire
- Structuring pages so every section stands alone, which also makes them easier for AI tools to cite
That last point turned out to be a bonus. The same structure that wins Google rankings earns citations in ChatGPT and AI Overviews. I break down how in my generative engine optimisation guide.
Stage 4: clean authority, built slowly
This is where a restart is won or lost, and where the temptation to cut corners is strongest. A new domain with good content still needs links to compete for a commercial term.
- Relevance over volume. A handful of links from sites genuinely connected to the niche beat dozens of generic ones.
- Nothing bought in bulk. No link packages, no networks, nothing that would make sense to a reviewer as manipulation. This site already had one manual action. A second would be fatal.
- Steady pace. Natural link growth is gradual. A sudden spike on a young domain is a red flag.
- Brand mentions count too. Being discussed across the web builds trust with Google and with AI search tools.
Stage 5: turn rankings into leads
Ranking was never the goal. Leads were. So the money page was built to convert:
- A clear call to action visible without scrolling
- Proof close to the call to action
- A short, simple enquiry form
- Every enquiry tracked back to the page and search that produced it
A ranking that does not convert is a vanity metric. Close to 200 leads a month is the number that tells you the recovery actually worked.
Stage 6: measure three numbers
Throughout the rebuild, I tracked three things and ignored almost everything else:
| Metric | Why it matters |
|---|---|
| Position for the core term | The single number that showed whether the strategy was working |
| Organic traffic | Confirms rankings are turning into real visitors |
| Leads | The only number the business actually cares about |
Impressions, domain rating and visibility scores are interesting, but they do not pay anyone. Keeping reporting to these three kept every decision focused on what moved the business.
The result
Within six months of launching the new domain, the site reached position one for its core commercial term, and it has held that position since. It now produces close to 200 leads a month, more than the channel was worth before the penalty.
The penalty turned out to be the moment the site stopped relying on something fragile and started ranking on foundations that do not need defending every time Google ships an update.
Lessons for anyone facing a penalty
- Diagnose before you act. Manual action, algorithmic hit, or a technical fault that looks like one? The fix for each is completely different.
- Be honest about recover versus restart. Attachment to a domain costs businesses months.
- A lifted penalty is a starting line, not a finish line. Plan the rebuild either way.
- Never bring the cause with you. Not through redirects, not through links, not through tactics.
- Build for the money term first. Everything else supports it.
- Measure leads, not activity.
If you are staying on your domain: writing a reconsideration request
Restarting is not always right. If your penalty is narrow and your domain has real brand value, a reconsideration request is the route. What gets them approved:
- Fix everything first. Partial cleanup gets rejected. Google wants to see the problem gone, not reduced.
- Be specific. State exactly what was wrong, what you did about it, and when.
- Show your working. For link penalties, keep a record of every removal request and outcome, and disavow only what you could not get removed.
- Take responsibility. Blaming a previous agency without explaining what you have changed reads as evasive.
- Explain prevention. Say what stops it happening again: new processes, a new agency, new content standards.
Vague, apologetic requests get rejected. Clear, documented ones get approved.
If you are under a manual action, or you have seen a sharp drop and cannot tell why, actual SEO's first pass audit is free. I will tell you honestly what caused it, whether Google penalty recovery or a restart makes more sense, and what it will realistically take. And if it is something you can fix yourself, I will say so.
Frequently asked questions
How long does Google penalty recovery take?
A manual action can be lifted within a few weeks of a thorough reconsideration request, though rejected requests add time. Rebuilding rankings afterwards, or on a new domain, typically takes three to six months for a competitive term.
Should I move to a new domain after a Google penalty?
Only if the cause runs deep into the site's history, the domain carries little brand value, or cleanup would take longer than rebuilding. If the problem is narrow and fixable, cleaning the original and requesting reconsideration is usually faster.
Can I redirect a penalised domain to a new one?
It is risky. Redirects pass signals between sites, which can include the problem you are trying to escape. Rebuilding cleanly and pointing people to the new site through messaging is safer.
How do I know if I have a manual action?
Check the Manual actions report in Google Search Console. If there is a manual action, it is listed there with the type and scope. If the report is clean but traffic dropped sharply, you are likely looking at an algorithmic update or a technical issue.
Will my rankings come back once the manual action is lifted?
Not automatically. Lifting a manual action removes the restriction, but any rankings built on what caused it will not return. Expect to rebuild from close to zero either way.
Where this leads
If this is the bit you want handled rather than explained:
- Google penalty recovery services. This process, run on your site, starting with an honest diagnosis.
- Organic SEO. The rebuild stages run as a monthly engagement.
- Results. The numbers behind this and other accounts.
