The fixed three-to-six-month promise is the wrong contract
A founder approves six months of SEO. The monthly report shows more published pages and a longer keyword list, but nobody can explain when those outputs should become qualified demand.
Search results are crowded with confident timelines. The problem is that they blend four different events: Google finding a change, a page appearing for a relevant question, a suitable buyer visiting and that buyer becoming revenue. Those events do not share one clock.
Google says crawling a changed page can take from a few days to a few weeks. Broader improvements may take a few days or several months to affect search results, and inclusion or impact is not guaranteed. That makes a universal deadline impossible—but it does not make SEO unaccountable.
My verdict: replace the promised ranking date with an evidence contract. Define the commercial opportunity, work to be shipped, proof expected at each layer and the decision date for continuing, repairing, narrowing or stopping.
The Search Investment Proof Chain
I use four proofs to stop early activity from being presented as final success. Each stage depends on the one before it, so a revenue problem cannot be repaired by publishing more pages when the priority pages are not accessible or aligned with buyer intent.
Can search systems use the asset?
Priority pages are crawlable, indexable, canonical and internally connected.
Is the right market finding it?
Relevant non-brand questions produce impressions, citations or suitable entry pages.
Do appropriate buyers progress?
Organic visitors reach service, product, proof or contact paths with useful intent.
Does the channel justify investment?
Accepted enquiries, opportunities, sales or defensible assisted value support the case.
Do not skip the middle. Rankings without suitable visits can signal the wrong questions. Visits without progression can reveal weak positioning, proof or conversion. Enquiries without commercial fit can expose a segment problem rather than an SEO problem.
Set the review window from the starting condition
The same work can produce different evidence on two websites. An established site improving an indexed commercial page starts closer to discovery than a new domain entering a competitive market. The owner should approve the starting condition before approving the timeline.
| Starting condition | First evidence to require | Main uncertainty | Owner question |
|---|---|---|---|
| Established site; important pages already indexed | Relevant query and page movement after specific improvements | Whether the page answers the right commercial decision better | Are we improving existing demand capture or creating another content project? |
| New site or new market category | Access, index coverage, clear entity signals and narrow discovery | Whether the market associates this business with the target problem | Is the opportunity valuable enough to earn authority over time? |
| Traffic exists but enquiries are weak | Qualified entry-page behaviour and progression to commercial pages | Intent, offer, proof and conversion fit | Would more traffic amplify the same leak? |
| Local business with known service demand | Accurate business information, relevant local discovery and suitable calls or enquiries | Location, service coverage, reviews and operational capacity | Can the business serve the demand it wants to win? |
| Large content library with declining performance | Prioritised improvements, consolidation and recovery by page group | Content usefulness, overlap and changing buyer behaviour | Which assets deserve protection, rebuilding or retirement? |
If leadership is deciding between immediate demand capture and a longer owned-discovery asset, use the Paid Ads vs SEO decision framework. If the wider question is whether SEO still deserves investment as AI answers grow, use the Search-to-Revenue Portfolio.
Use review windows without turning them into ranking guarantees
Verify decisions and delivery
Confirm the market, priority questions, baseline, commercial pages, measurement and shipped technical fixes. Accountability is the outcome.
Look for directional evidence
Review crawl and index status, relevant impressions, improved entry pages and early qualified journeys. Diagnose gaps rather than demanding revenue on every model.
Make an investment decision
Compare completed work with discovery and demand evidence. Continue, repair or narrow the programme with one named constraint and owner.
Judge commercial movement
Connect organic cohorts to accepted enquiries, opportunities, sales or assisted journeys only after buyers had time to progress.
These are management checkpoints, not universal performance benchmarks. A provider should show what changed, what evidence should follow and what decision will be made if it does not. “SEO takes time” is not an explanation for unshipped work, unclear intent or missing measurement.
Use the same principle when leadership asks whether marketing is working: connect channel evidence to a defined business movement with the Marketing Evidence Chain.
Choose continue, repair, narrow or stop from the evidence
| Observed pattern | Likely meaning | Decision | Do not do |
|---|---|---|---|
| Agreed work is shipped and relevant visibility is improving | The proof chain is progressing | Continue to the next evidence checkpoint | Scale content volume before learning which pages matter |
| Work is shipped, but pages remain inaccessible or unindexed | Technical access or quality problem | Repair the foundation and recheck priority URLs | Commission more content to hide the blockage |
| Impressions rise, but suitable visits do not | Weak intent, title, proposition or search-result fit | Narrow the question set and improve the definitive pages | Celebrate visibility as commercial success |
| Qualified organic visits rise, but pipeline does not | Offer, proof, conversion or sales handoff problem | Repair the buying path before expanding acquisition | Blame rankings for a downstream leak |
| Completed, credible work shows no improving evidence across agreed reviews | Opportunity, positioning or execution may be wrong | Stop, change the market or redeploy the budget | Renew solely because results are described as long term |
Require the evidence behind any case study used to justify patience. Check the starting point, timeframe, market, work completed and business outcome with the Case Study Evidence Ladder. A successful result elsewhere is context, not a forecast for this business.
Do not start a separate AI-search clock
Google states that its established SEO practices remain relevant for AI Overviews and AI Mode, with no additional technical requirements or special schema needed. A page must still be indexed and eligible for Google Search, and inclusion is not guaranteed.
OpenAI says any public website can appear in ChatGPT search and advises publishers not to block OAI-SearchBot if they want content considered for summaries and snippets. Access creates eligibility; it does not create a promised date for a mention.
Keep one discovery investment and inspect different evidence surfaces. Use Search Console, analytics, CRM outcomes and a controlled set of commercially relevant AI questions. The AI Visibility Evidence Ladder explains how to measure mentions and citations without confusing them with revenue.
For the broader foundation—technical access, business identity, useful decision pages and corroborated proof—use the Business Discoverability Stack. Review AI Growth Systems, relevant case-study evidence and my operator background before choosing the next investment.
Sources and evidence notes
Sources and current search results were checked on 22 August 2026. Search priority is qualitative; no search volume, ranking deadline, traffic uplift or client result is claimed. The Search Investment Proof Chain, review windows and decision matrix are original ThomPerformance analysis.
Frequently asked questions
How long does SEO take for a new business website?
There is no reliable universal deadline. A new site must first become crawlable, indexed and relevant for specific buyer questions before traffic or enquiries can follow. Set staged evidence checks rather than accepting a guaranteed month: access, discovery, suitable visits and commercial movement.
What is the first sign that SEO is working?
The first useful sign depends on the starting problem. It may be priority pages becoming indexed, relevant impressions appearing, suitable non-brand queries growing or qualified organic visitors reaching commercial pages. Treat these as leading evidence, not proof of profitable growth.
Can SEO work within 30 days?
Some changes can produce evidence within 30 days, especially on an established site with known technical or page-level problems. Google says crawling may take days to weeks and broader improvements can take much longer. An early gain is a signal; it is not a durable forecast.
When should a business stop investing in SEO?
Stop or narrow the programme when the target market is commercially weak, execution remains unaccountable, the site cannot support the required offer or evidence, or completed work produces no improving signal across an agreed review window. Do not continue solely because SEO is described as a long-term channel.
Does AI search make SEO results faster?
Not automatically. Google says its established SEO practices also apply to AI Overviews and AI Mode, with no special technical shortcut. ChatGPT search also requires accessible public pages for inclusion. AI search adds discovery surfaces, but relevance, evidence, access and buyer value still need to be earned.
Fund SEO against evidence, not patience
SEO can be slow without being vague. Define the starting condition, move through access, discovery, qualified demand and commercial value, and decide what happens when a layer stalls. The business should know what it is waiting for, why that evidence matters and when waiting becomes repair or redeployment.
Which proof is currently missing from your SEO investment: access, relevant discovery, qualified demand or commercial value?