SEO: Simplify, Then Add Lightness

He wasn't talking about websites. He was talking about racing cars. But the principle is identical and the failures are identical and the excuses are identical. Too much weight. Not enough useful work being done by any individual component. Impressive numbers on paper. Slow everywhere that matters.

I've been doing SEO since 1994. The single most common problem I find in inherited sites isn't that they lack horsepower. It's that they're hauling too much dead weight to use what they have.

Philosophy

Power-to-weight ratio is the metric that actually tells you how fast something will go. Raw horsepower is a vanity number.

A 500-horsepower SUV loses to a 150-horsepower Lotus Elise through every corner on every circuit. Not because the SUV lacks power. Because it's dragging its own mass through every turn and the Elise isn't dragging anything. The Elise weighs 798 kilograms. That's the Elise Sprint with the carbon fiber bits. Every gram in that car is doing a job. Nothing is there for comfort or appearances.

SEO works the same way. A site with 400 pages, 67 plugins, three conflicting schema implementations, duplicate metadata on half the URLs, redirect chains four hops deep, and a Core Web Vitals score that makes Google wince has plenty of horsepower. It just can't use any of it. The weight is killing it everywhere—not just on slow pages but in crawl budget, authority signals, internal link equity bleeding into 404s, structured data Google ignores because it contradicts itself three different ways.

The question isn't how much you have. It's how much of what you have is actually pulling its weight.

Practice

When I audit a site I'm calculating something like SEO horsepower per ton. How much crawl budget is being spent on pages that will never rank and never convert? How many plugins are adding load time without adding function? How many redirects are in the chain that could be one direct hop? How many schema types are implemented incorrectly and fighting each other?

Every one of those is dead weight. And dead weight costs you everywhere—not just on the slow pages, not just in the bloated crawl, but in every signal you're trying to send.

Simplify first. Find what's working and protect it. Find what's blocking it and remove it. Then—only then—think about adding anything.

Example

Client comes to me with a WordPress site carrying three years of agency work. Forty-one active plugins. Seven different SEO tools installed simultaneously, several generating conflicting meta tags. A redirect map that looks like someone dropped a plate of spaghetti. PageSpeed in the low thirties on mobile.

They want to know why their new content isn't ranking.

The content is fine. The site is eating itself. Every new page is fighting for crawl budget against hundreds of orphaned pages from a product line they killed in 2022. Every good signal is being diluted by structural noise underneath it.

We don't add anything for the first six weeks. We remove. Consolidate. Redirect properly. Kill the redundant plugins. Fix the canonical chaos. Let Google reindex a site that finally has a clear structure to index.

Traffic moves before we publish a single new word.

Philosophy

Chapman also said: any car which holds together for a whole race is too heavy.

Brutal. He meant if nothing failed you left weight on the table. Every gram of margin you kept was a gram slowing you down for no reason.

In SEO: if your site has zero thin content, zero structural debt, zero redundancy—you haven't audited it seriously. Every site accumulates weight over time. The question is whether anyone's paying attention.

Current Formula 1 cars have a minimum weight of 800 kilograms including the driver. The drivers complain they feel sluggish in slow corners. 800 kilograms and they're complaining about being too heavy. That's the standard you're being held to whether you know it or not. Google's crawler has seen everything. It knows what light feels like.

Practice

Content weight is the one clients resist most. They wrote it. They paid for it. It exists. It feels like an asset.

Ten pages on the same topic, none comprehensive, all competing with each other in search—that's not ten assets. That's one asset shredded into ten pieces too small to rank for anything. Consolidation is weight reduction. One authoritative page outperforms five thin ones every time because the authority concentrates instead of diluting. You're not losing content. You're losing drag.

Example

Local service business. Twenty-three location pages. Fourteen are copy-paste variants with the city name swapped in. Google has stopped indexing eight of them entirely. The ones that are indexed aren't ranking because they offer nothing a user in that city couldn't get from the homepage.

We keep six. Rewrite them with actual local specificity—real service area detail, real schema, real differentiation. Properly redirect the rest.

The six outperform the twenty-three within ninety days.

Fewer pages. More horsepower per ton.

Philosophy

The Lotus Seven made around 40 horsepower in original spec. It weighed roughly 500 kilograms. It ate Ferraris for breakfast at track days because the power-to-weight ratio was obscene relative to everything else on the road. Not because of what it had. Because of what it didn't have.

No noise deadening. No carpets. No back seat. No heater worth mentioning. No spare tire. No A/C. No passenger seat if you ordered the single-seater. Everything that wasn't making the car go faster or turn harder or stop shorter got left in the bin.

The fastest web page is one that already exists. A flat HTML file served from a CDN edge node is the Lotus Seven of web delivery. No database query. No PHP execution on request. No plugin stack firing twelve processes to render a page that could have been pre-built and cached. The server hands you a file that already exists. That's the whole transaction.

Movable Type understood this in 2001. The CGI scripts would rebuild the site into flat HTML files whenever content changed. Visitors got pre-rendered static pages. Near-instant load. Minimal server load. hillmole.com still runs on a 2006 MovableType install on Perl and still loads faster than most WordPress sites built last year.

The industry spent fifteen years getting fat on dynamic generation and then reinvented flat files as JAMstack and called it a revolution.

Practice

Strip the production car for the track. Lose the noise deadening, the back seats, the headliner, the A/C, the spare, the carpets. Pull everything that isn't moving the car forward. What's left is faster everywhere because there's nothing left that isn't working.

For a website that means: static generation where possible, CDN edge serving, zero database calls on the critical render path, deferred or eliminated third-party scripts, properly compressed images, clean HTML that a crawler can parse without executing JavaScript first.

Example

Client migrates from WordPress to a git-based JAMstack build. The architecture is correct. Pre-rendered. Edge-deployed. The HTML is flat and fast.

Then the marketing team gets involved. HubSpot tracking pixel. Hotjar session recorder. Intercom live chat. Google Tag Manager loading four more tags. A cookie consent manager that blocks render until the user responds. A Calendly embed on the contact page. Two retargeting pixels from different ad platforms.

The HTML still loads in 180 milliseconds. The browser spends the next six seconds executing JavaScript before anything is interactive. Largest Contentful Paint: 6.4 seconds. Worse than the WordPress site they migrated away from.

The architecture was the Lotus Seven. The marketing stack turned it into a Rimac Nevera. 5,000 pounds of battery and torque vectoring. Impressive in a straight line. A handful everywhere else.

Philosophy

The Rimac Nevera weighs 5,071 pounds. It does 0-60 in 1.85 seconds. It costs $2.4 million. It is genuinely one of the fastest accelerating vehicles ever built and it corners like it's hauling a grand piano because it is, essentially, hauling a grand piano. The battery pack alone weighs over 1,500 pounds. All the torque vectoring and rear-wheel steering in the world is managing weight that shouldn't be there.

That's the big agency SEO play. Obscene budget. Obscene horsepower. Impressive on the straight—traffic dashboards, keyword counts, backlink volumes, monthly deliverable decks. But the weight of the apparatus kills it everywhere that matters.

A recently client chose to work with me instead of a large agency. The agency had proposed a six-month content plan. Forty new pages, a backlink campaign, monthly reporting. No mention of the indexing issues I found in a twenty-minute crawl. No mention of the canonical errors affecting their top ten pages. No mention of the fact that Google had soft-deindexed their blog entirely because of thin content signals.

The content plan was the Nevera. Impressive digits. Expensive. Fast in a straight line.

Practice

Cloudflare and CDNs can add real horsepower to a site. But there's an honest version of that conversation and a dishonest one.

The honest version: Cloudflare on a clean, well-structured, properly optimized site is a legitimate restomod. You take a solid chassis, strip it, rebuild it right, then bolt on modern running gear. Real performance gains everywhere.

The dishonest version: Cloudflare on an unreformed pig is a pig served faster. You're caching a mess and delivering a mess with better latency. The PageSpeed number moves. The client feels good. The redirect chains, the plugin sprawl, the duplicate content, the thin pages eating crawl budget—all still there, just delivered more efficiently to a crawler that sees all of it anyway.

A restomod starts with a car worth restoring. Do the structural work first.

Example

Same client. We spend month one on nothing but technical cleanup and consolidation. No new content. No outreach. No Cloudflare configuration theater. Redirect audit. Plugin cull. Canonical fix. Schema cleanup. Image compression. Render-blocking script deferral.

Month two we configure Cloudflare properly on a site that now deserves it.

Organic traffic up 34% before we write a single new word.

Chapman would have recognized the sequence immediately.

Fix the chassis. Strip the weight. Then see what the engine can actually do.

Then—and only then—bolt on the supercharger.

Chris Abraham is a senior SEO consultant at Gerris Corp in Arlington, Virginia. He has been building and optimizing websites since 1994. If your site is a pig, he can help. If it's already a Lotus, he can make it faster.