Eight problem-solving frameworks from industry, applied to SEO
Manufacturing and engineering solved the problem of diagnosing faults in complicated systems decades ago. A website is a complicated system. Here are eight of their methods, what each one is good for, and how to run it on your own site this afternoon.
Most SEO advice treats symptoms
Traffic is down, so publish more. Bounce rate is high, so rewrite the page. Rankings slipped, so build links. Each of those might be right, and each is a guess dressed as a diagnosis.
Industry stopped guessing a long time ago. When a production line underperforms, nobody starts replacing parts at random — there are methods for finding out which part is actually the problem. Those methods transfer to websites more cleanly than you would expect.
Three ways to work out what is actually wrong
- The 5 Whys, from ToyotaAsk why five times and the answer usually moves somewhere unexpected. Rankings down, because traffic dropped, because articles are not ranking, because they are too generic, because the brief was a keyword list rather than a question. The fix is at the fifth answer, not the first. Run it on one underperforming page and see where the trail goes — it often ends somewhere technical rather than in the content.
- The Fishbone diagram, from IshikawaMost ranking problems have several causes at once, and the 5 Whys can march you down one branch while ignoring three others. A fishbone maps them side by side: content, technical, local signals, links. Poor performance is rarely one thing.
- Theory of Constraints, from GoldrattA production line runs at the speed of its slowest station. Improving anything else changes nothing. Sites work the same way: if Google cannot crawl a section, no amount of writing for it matters. Find the constraint, fix that, then look again — because the constraint will have moved.
Practical version of all three: run your homepage through PageSpeed Insights and open the Pages report in Search Console before you touch any content. If something is blocking crawling or the site takes eight seconds to load, that is your constraint and everything else is decoration.
Three ways to prioritise once you know
- Pareto, the 80/20 lensNot a law, and the numbers are never exactly 80 and 20, but the shape holds: a small handful of your pages carry most of your traffic. Ahrefs found around 90% of pages get no search traffic at all. Open your analytics, find the two or three pages doing the work, and improve those before writing anything new.
- Kaizen, small continuous changeThe alternative to the expensive quarterly campaign. Refresh one page, fix one broken link, add one internal link. Individually trivial, and after a year the site is unrecognisable. It also survives contact with a busy month in a way that campaigns do not.
- Jobs To Be Done, from ChristensenNobody searches for keywords. They search to get something done — solve a problem, compare two options, book somebody today. Ask what job each page is hired for. If you cannot answer in one sentence, that is why the page is not converting.
Two frameworks for the long run
- PDCA, from DemingPlan, do, check, act. The step everybody skips is check. Most SEO work is published and never looked at again, which means nobody ever finds out that the article ranking beautifully sends no enquiries because it has no call to action. Pick one post, look at whether traffic converts, and act on the answer.
- The S-curve, for expectationsResults do not arrive in a straight line. The first few months look flat while technical fixes and new pages get indexed, then movement compounds, then it plateaus and needs new work to move again. This matters because the flat part is where most people conclude it is not working and stop — usually a month or two before it would have started.
Which to reach for
| Framework | Use it when | What it gives you |
|---|---|---|
| 5 Whys | One page is underperforming | The actual cause, not the symptom |
| Fishbone | Several things seem wrong at once | All causes mapped side by side |
| Theory of Constraints | Effort is going in, nothing is moving | The one blockage to fix first |
| Pareto | Too much to do, too little time | The few pages that matter |
| Kaizen | You cannot fund a big project | Compounding change that fits round the day job |
| Jobs To Be Done | Traffic arrives but nobody enquires | What each page is actually for |
| PDCA | Work gets published and forgotten | A loop that closes |
| S-curve | Month three and nothing has happened | A reason not to give up too early |
What it looks like applied
Paramount Space Cleaning in Coventry started with us in October 2025 with no search visibility. The constraint was technical, not editorial — writing more would have changed nothing until it was fixed. Seven months on they hold position one for “deep cleaning coventry”, rank on page one for 55 search terms, and traced a £7,000 commercial contract back to organic search.
The caveat we attach to everything: one business, one city, one sector that is competitive but not brutal. Not a forecast for anyone else. Nora approved every figure before we published it.
Simon & Lisa Batchelor run Get Your Website Seen. Simon spent ten years as owner and editor of a magazine, then built and sold Commerce Computers Ltd, a trade computer supplier that traded online as shop4it.co.uk. So we have been on the buying side of marketing that did not work, which is most of why this site reads the way it does. You deal with us directly, and every figure published here belongs to a client who approved it.
Or let us run the diagnostics
The free audit does most of the above on your site and tells you what the constraint is, what to fix first, and what you can do yourself. Written by one of us. It is lead generation and we are not pretending otherwise, but you get it without a call first.
More articles · How SEO works · Health Score Checker · Broken Link Checker · All free tools