Negative split SEO
Controlled early. Earned late.
Most SEO is built to look busy in month one. NSEO sequences the work so the diagnosis and the structure are finished before a penny goes into content. A slower first half, and a far steeper second.
Get Your Free WalkthroughFree, about 15 minutes, and no call unless you want one. I do four a month.
Same race. Different finish.
Conventional SEO peaks early and drifts. NSEO stays quiet through the first two phases, then climbs once the review gate is passed.
What NSEO is.
NSEO (Negative Split SEO) is a proprietary search methodology developed by Rick Ashworth, founder of Yellow Lava. It sequences SEO work in five phases — Diagnostic, Foundation, Review Gate, Acceleration and Consolidation — structured so that the diagnostic and structural work is complete before content investment begins. The result is a slower initial output curve and steeper later results, because the foundation is stable when acceleration starts.
The name comes from the endurance sport principle of the negative split: running the second half of a race faster than the first. Go out too hard and you fail before your competition does. The athletes who win read the race correctly, hold back early, and have the discipline not to blow the strategy in the first mile.
Most SEO burns hard and fades fast.
Two ways to run the same race. Only one of them is built to still be climbing in month twelve.
Fast out of the blocks
Keyword research on day one. Content calendar on day two. Technical problems discovered six months in, by which point the content sitting on top of them has already been paid for.
It produces a reassuring first report and a flat second half. The process is optimised for looking busy, not for producing results.
Impressive early. Stalled late.
Diagnostic first
No content investment until the structural picture is clear. The entity model, the technical health, the architecture and the authority layer go in before a single pillar page is commissioned.
The early reports look thin. That is the trade being made deliberately, and it is the reason the second half is steep.
Patient early. Fast late.
What actually happens.
Every phase has one job, and nothing moves forward until that job is done. Here is each one in plain English: what I do, what it looks like on a real site, and what I need from you.
Pre-race analysis
I find out what is actually stopping you being found — before I suggest fixing anything.
Before we touch a single piece of content, I map exactly where your site stands: technical health, crawlability, keyword opportunity, competitive gaps, and the entity model sitting underneath all of it. No guesswork. There will be no movement in the headline numbers this phase. That is deliberate — this phase buys information, not rankings.
What I actually do
- Crawl the whole site and log every technical fault that stops Google reading it properly.
- Pull your real rankings from Google Search Console and Ahrefs — including the terms sitting just off page one, on pages two to four, where the quickest wins usually are.
- Ask ChatGPT, Perplexity and Google’s AI Overviews the questions your clients ask, and record who gets named instead of you.
- Compare you against the three or four competitors who are winning the searches you want.
- Check what Google and AI actually understand your business to be — who you are, where you work, what you do.
On a real site
A finance recruiter might find its location pages all open with the same paragraph, its consultants aren’t on the site at all, and ChatGPT names three competitors when asked who recruits finance directors in the North West.
What I need from you
Access to Google Search Console and Analytics, and an hour on who your best clients are and what they buy.
Output: a clear picture of what’s holding you back — and a prioritised plan to fix it in the right order.
Building race fitness
I fix what’s broken and build the structure, so everything after this has something solid to stand on.
This is where most agencies skip straight past. I fix the structural problems first: technical errors, site architecture, internal linking, entity signals and schema, and the slow-burning content problems that quietly cost you rankings. Clients used to agency decks full of deliverables find this phase uncomfortable. That discomfort is evidence the methodology is working.
What I actually do
- Fix the technical faults in the order that matters most — pages Google can’t index, slow load times, broken links and redirects.
- Give every page one job — one sector, one location, one question — so your pages stop competing with each other.
- Rewrite the opening of key pages so the first lines answer the search directly. That’s the part Google and AI lift and quote.
- Add structured data (schema): code that tells search engines who you are, where you are and what you do, without them having to guess.
- Link your pages together so the ones that make you money carry the most weight.
On a real site
The Leeds page says something the Manchester page doesn’t from its very first line, and the consultants who actually fill those roles are named on the pages for their sectors.
What I need from you
Someone who can publish changes to the site (or access for me to do it), and sign-off from anyone whose name goes on a page.
Output: a site Google can actually crawl, trust and rank — before we pour fuel on it.
The checkpoint
We stop, look at the evidence together, and only push on when the foundations are holding.
Before acceleration, we stop. This is the structural innovation, and it is deliberately longer than the phases either side of it. We review what has been implemented, measure what has moved, and confirm the foundations are solid. Nothing accelerates until four questions all answer yes:
- Are the technical foundations stable?
- Is the content architecture established?
- Is the authority layer in place?
- Do you understand and commit to what Acceleration requires?
What I bring to the table
- Before-and-after on the pages Google has indexed and how it reads them.
- Movement on the search terms we agreed in the Diagnostic — not a vanity list of everything you rank for.
- Whether AI search names you yet, against where we started.
- Anything still outstanding, and what it would cost to leave it.
On a real site
The location pages are indexed and climbing, but AI still isn’t naming you. The gate stays shut until that’s explained, because pushing harder won’t fix it.
What I need from you
An hour, and a straight answer on whether you can commit to what the next phase takes.
If one of the four is a no, we stay here and fix it. This is the gate that separates structured SEO from endless activity with no direction.
Output: a shared decision point — go, adjust or pause — with full transparency on what the data says.
The negative split
Now we go after the searches worth winning, on a site that can hold on to them.
Now we push. With solid foundations in place, content and link investment compounds instead of leaking away. New pillars aimed at the terms worth winning, intentional link building, and social corroboration feeding the citation signals that AI search reads. This is the second-half kick — and it is earned, because the foundation held.
What I actually do
- Build new pages around the questions your buyers actually ask — fees, rebates, time to hire — not just descriptions of what you do.
- Put your people’s expertise on the page: named authors, real examples, first-hand knowledge that AI can’t fake.
- Earn links and mentions from industry sites, directories and press.
- Back up what your site says on LinkedIn and elsewhere, because AI checks more than your website before it recommends you.
On a real site
Search “what happens if a candidate leaves after three months” and the answers come from Reddit and LinkedIn, not from recruiters. That’s a page waiting to be written by an agency that knows.
What I need from you
Twenty minutes a month from the people who know the market, and quick reviews before anything goes live.
Output: measurable organic growth that builds momentum rather than plateauing after 90 days.
Crossing the line strong
Keep what you’ve won, and keep adding to it as search changes underneath you.
Protecting what you have built. Defending rankings, refreshing content, closing the gaps competitors are targeting, and monitoring that watches AI citation signals and not just SERP positions.
What I actually do
- Track your rankings, traffic and AI mentions every month, against your competitors rather than in isolation.
- Refresh pages that start slipping before they fall off page one.
- Spot competitors moving onto your terms, and respond.
- Add new pages as new questions, services and locations come up.
On a real site
An AI Overview starts quoting a competitor for a term you held. You hear about it that month, with a fix, rather than finding out at next year’s review.
What I need from you
A monthly check-in, and a heads-up when the business changes — a new office, a new sector, a new service.
Output: an organic channel that works harder every month — not one you rebuild from scratch each year.
Already on LinkedIn.
The whole method as an eight-card carousel — the big idea, all five phases, and what each one produces. Same content as this page, in the time it takes to swipe.
NSEO is not for everyone.
It asks for patience in the first half in exchange for a steeper second. If that trade does not suit where your business is right now, I would rather say so before you have paid for a diagnostic.
It fits if
- You are a recruitment agency or an ambitious SME with a real market to take share in.
- You can commit to six to twelve months rather than a quarter.
- You want to be visible in AI search as well as in Google, which rewards structure and entity clarity above volume.
- You would rather know what is actually wrong than be handed a content calendar on day one.
It does not fit if
- You need leads this month. That is a paid media problem, not an SEO one.
- Your board measures the agency by deliverables shipped per month.
- You want someone to write the blog posts and leave the architecture alone.
- You have already decided what the problem is and want it implemented.
Not sure where you stand?
The AI Citation Readiness Check is a short set of questions on whether AI search can currently see, understand and quote your business. It’s the fastest way to find out if the Foundation phase is where your problem actually lives.
The plain answers.
Who developed NSEO?
Rick Ashworth, founder of Yellow Lava. The term was coined in 2022, in a LinkedIn post about the Wilmslow Half Marathon, and the methodology was built out from sixteen years of performance coaching before it was ever applied to a website.
Why is it called Negative Split SEO?
A negative split is running the second half of a race faster than the first. NSEO applies the same discipline to search: controlled early, earned late.
How long before I see results?
Later than conventional SEO, and then faster. The early phases are structural and deliberately quiet; the acceleration phase is where the curve turns. NSEO is a six to twelve month commitment, not a quarterly one.
What makes the review gate different from a progress call?
A progress call reports. The review gate decides. Acceleration does not begin until the technical foundation, the content architecture and the authority layer are all in place and you have committed to what comes next.
Does NSEO cover AI search?
Yes, and that is a large part of why it is sequenced this way. Being cited by an AI assistant depends on entity clarity, structured data and corroboration — exactly the work that the Foundation phase front-loads.
Who runs the work?
I do. One person, no account manager layer, which is also why the diagnostic comes first — I only take on what I can actually see through.
Phase one starts with a look at your site.
Send me your site and I’ll record a video going through it — what’s working, what’s costing you, and what I’d fix first. It’s a short version of the diagnostic, and it tells you whether any of this is worth your money.
Get Your Free WalkthroughFree, about 15 minutes, and no call unless you want one. I do four a month.
