
Strategy
Part of Nine steps to build a technical SEO strategy: outcomes, sampled crawls, scoring and review points
Why a technical SEO strategy framework beats chasing algorithm updates
Why a technical SEO strategy framework beats chasing Google algorithm updates, and how to build one with named tools, scoped audits and a quarterly review.
Reactive technical SEO means chasing each Google update as it lands: a core update, a helpful content update or a spam update. A strategy framework replaces that cycle with a planned set of objectives, scoped audits and named tools, so you change what the evidence supports rather than what the latest announcement implies.
What to take away
- The GOV.UK service manual on user research lists methods such as interviews and usability testing that give a strategy evidence rather than guesswork.
- A framework fixes scope before tools are chosen, so an audit answers a defined question with a clear scope.
- Google's SEO Starter Guide treats crawlability and helpful content as fundamentals, keeping planning anchored.
- The IAB UK introduction to creative best practice keeps messaging consistent as pages are rewritten.
- Chasing Google core updates, helpful content updates or spam updates without a baseline leads to reactive fixes you cannot measure.
- A written checklist turns the plan into weekly tasks.
Why chasing updates costs more
Google's core updates, the helpful content update and spam updates each trigger a wave of reactive fixes. Teams that chase them often rewrite titles, block URLs or change internal links without a baseline, so they cannot tell whether the update caused the movement or their own changes. A framework avoids that by holding a dated evidence base: crawl stats, Google Search Console index coverage and Core Web Vitals field data from PageSpeed Insights and the Chrome User Experience Report (CrUX). When an update lands, you compare the new data with the baseline and act on the templates that actually changed.
Set the objective first
Start with one measurable objective, agreed before any crawl. Decide whether you need more indexed pages, faster rendering, fewer duplicate URLS or better internal linking. Give it a deadline and an owner.
Example objectives
- 10%lift in non-brand organic sessions
- 20%reduction in crawl waste
- 2 quartersdeadline for retailer objective
Connect it to a business outcome, not a vanity metric. A retailer might target a 10% lift in non-brand organic sessions over two quarters; a publisher might target a 20% reduction in crawl waste. Both are illustrative figures.
Map objectives to audit phases in a single planning document that you own and version.
Choose the audit scope
Scope decides what the audit covers and what it leaves out. List domains, subdomains and templates, marking which are commercially important and which are legacy.
Audit scope time boxes
Template audit
- Duration
- Two weeks
- Focus
- Templates
- Stakeholder ease
- Early estimate
Migration review
- Duration
- Six weeks
- Focus
- Migration
- Stakeholder ease
- Early estimate
Google's starter guide explains that search engines must crawl, understand and serve content, so scope those three areas separately. Crawl budget problems and content quality problems need different evidence.
Set a time box. A scoped template audit might take two weeks; a migration review can take six. Publishing that estimate early eases stakeholder conversations.
Match tools to questions
Tools are not the strategy. Pick each one because it answers a question already written in the scope. For crawling, Screaming Frog SEO Spider and Sitebulb are common choices; for log analysis, Screaming Frog Log File Analyser or Botify; for rendering, Google Search Console's URL Inspection tool shows the rendered HTML.
Avoid buying overlapping tools. Two crawlers reporting the same orphan pages add duplication, not clarity. Ahrefs and Semrush are useful for backlink and keyword data, but check whether a crawler already covers the same ground. Record which tool owns each question in a table.
Keep manual checks in the mix. Google Search Console's index coverage report, server logs and a handful of template URLS often reveal what automated crawls miss.
Build the evidence base
Evidence is data you can show. Pull crawl stats, Google Search Console index coverage, Core Web Vitals field data from PageSpeed Insights and the Chrome User Experience Report (CrUX), and internal link counts into one dated folder.
Where evidence is thin, use the user research methods in the GOV.UK service manual rather than assumption. Short interviews with content editors often explain why pages duplicate.
Write findings as problems with owners, not tickets. Each should say what is broken, who it affects and the expected gain.
Plan delivery and reporting
Turn findings into a delivery plan with quarters, owners and dependencies. Separate quick fixes from projects needing engineering time. A redirect cleanup may ship in a sprint; a rendering overhaul can take a quarter.
Agree reporting before work starts: which metrics move, how often they are reviewed, who signs off. A monthly one-page report beats a quarterly deck nobody reads.
Record tasks and owners in a simple template you can reuse each quarter.
Keep content and creative aligned
Technical fixes only pay off when the pages they serve are worth reading. When templates change, brief editors on tone, format and calls to action.
The IAB UK creative best practice guidance supports consistent standards across formats, which matters when teams rewrite titles, descriptions and copy during an audit.
Pair each technical change with a content check. A faster page with thin copy still underperforms a slower page that answers the query.
Review and adjust quarterly
Set a quarterly review. Compare the objective, the evidence and the delivery plan. Retire tasks that no longer matter and add new ones only with a stated reason.
Keep a short decision log recording what changed, why and who approved it.
A framework is a loop, not a document.
Checklist
- Objective written with metric, deadline and owner.
- Scope lists domains, templates and exclusions.
- Each tool mapped to a question.
- Baseline evidence saved with dates.
- Findings include impact and owner.
- Delivery plan split by quarter.
- Reporting cadence agreed.
- Content standards briefed with technical changes.
- Quarterly review booked.
Common questions
How long should a framework take to build?
Allow two to four weeks for a first version covering one site. Larger estates with many templates need longer, so time-box the first pass.
Do small sites need a framework?
Yes. A single-page checklist still forces an objective and an owner, which prevents random fixes.
How do we prove the framework works?
Compare the baseline with the same metrics each quarter. Report movement against the stated objective rather than a general claim of improvement.
Who should own the framework?
A named person in marketing, product or engineering should own it. Shared ownership without a lead usually means the plan stalls.



