Updated 30 October 2026

A website can attract more visitors while the business receives no more useful enquiries. Before paying for another batch of articles or links, find out where the journey from search to sale is breaking.

Picture a prospective customer comparing two suppliers on a phone. One website answers the important questions, explains delivery and makes it easy to ask for a quote. The other has a polished homepage, vague service descriptions and a form that appears to submit but never reaches the sales inbox.

That is an illustrative scenario, not a client case study. It exposes a problem with treating SEO as a checklist: completing more tasks does not necessarily solve the constraint that matters.

Our position is straightforward: prioritise SEO work by the problem it solves and the evidence behind it. An indexing fault, a weak offer and an unanswered enquiry require different responses. A business may have all three, but it still needs a sensible order of work.

Start with four questions before choosing an SEO fix

Use this diagnostic to form a working hypothesis. It is a decision aid, not a Google ranking model or proof of what caused a performance change.

A low click-through rate alone does not prove the title is wrong. Poor sales alone do not prove SEO is wrong. Compare several signals, check that tracking works and consider changes in demand, stock, prices and competition.

1. Buying SEO activity before diagnosing the problem

“We need more backlinks” is a proposed intervention. It is not yet a diagnosis. The same applies to “publish every week” and “redesign the website”.

Check: choose one commercially important service or product category. Trace its path from relevant search to landing page, enquiry and sales follow-up. Record where the evidence becomes weak.

Act: write a short hypothesis: “Customers cannot find this service”, or “Visitors arrive but cannot establish whether we serve their location”. Specify the evidence that would support or challenge it before committing the budget.

2. Treating Google Business Profile as an afterthought

For an eligible business serving a local market, inaccurate hours, unsuitable categories or outdated contact details can undermine discovery and customer confidence. A complete profile does not remove the effect of location: Google describes relevance, distance and prominence as central to local ranking. Read Google’s local ranking guidance.

Check: compare the profile with what customers can actually visit, buy or book. Confirm that the website, phone number, hours and business details are accurate.

Act: correct errors and maintain the profile as business information changes. Do not promise a map position or invent branches to target extra towns. For help with the broader local presence, see our local SEO services in Kenya.

3. Targeting a keyword without understanding the buying decision

A search for “fridge prices Kenya” can involve comparison, while a model-specific search may indicate a narrower requirement. Neither phrase automatically tells you whether that visitor will buy from you.

Check: compare Search Console queries with actual enquiries. Which searches concern products you stock, locations you serve and work you want?

Act: group queries by the job the customer is trying to complete. Use customer conversations to test your interpretation. Avoid assuming that Kenyan customers use a particular language mix without evidence from your own audience.

4. Sending every searcher to the homepage

A homepage introduces the business. A useful service or category page helps someone evaluate a specific purchase. One page rarely does both jobs equally well for every offer.

Check: does each important service have a clear destination explaining its scope, suitability, process and next step?

Act: improve those destinations before commissioning unrelated blog posts. For an online shop, category navigation and product information may deserve attention before another general buying guide. Our e-commerce SEO service addresses this type of store-specific work.

5. Confusing more content with more useful information

A longer article can still leave the reader’s actual question unanswered. A concise service page can be valuable if it explains exactly what someone needs to decide.

Google’s people-first content guidance asks whether content contributes original information, analysis or value beyond other sources. It does not set a preferred word count.

Act: add something your business can substantiate: a real comparison, a clearly explained process, a limitation, a documented lesson or an answer drawn from customer questions. If a paragraph could appear unchanged on any competitor’s website, ask what it contributes.

6. Adding town names without adding local usefulness

Changing “Nairobi” to “Nakuru” across otherwise identical pages does little to explain the practical differences for a customer.

Check: does each location page contain truthful information that changes the decision—coverage, delivery arrangements, access, availability or relevant work?

Act: use one honest coverage page when there is little meaningful difference. Separate pages need a reader-facing purpose. Google’s spam policies identify doorway abuse, including similar regional pages that funnel users to one destination. Location pages are not inherently a violation; their purpose and execution matter.

7. Making a mobile layout look good without testing the task

A responsive preview can miss the problem that costs the enquiry: an overlay covering a button, a difficult form or a product option that cannot be selected.

Check: on an actual phone, find a service, read its details and complete the intended action. Try ordinary mobile data as well as Wi-Fi where practical.

Act: fix the obstruction you observe. Record the device and conditions so someone else can reproduce the problem. Treat mobile usability as a customer requirement, rather than assuming every usability change produces a ranking increase.

8. Chasing a perfect speed score while the buying journey remains broken

Performance matters, but a score is a diagnostic measurement. It is not the business outcome. Google states that good Core Web Vitals do not guarantee top rankings and cautions against chasing perfect scores purely for SEO. See its page experience guidance.

Act: investigate delays on important pages, using field data where available alongside repeatable tests. Prioritise changes that improve actual loading and interaction. If the enquiry form is broken, repair it while investigating performance; do not let score polishing consume the entire project.

9. Publishing pages Google cannot properly access or index

A useful page cannot attract ordinary organic search visits if it is not indexed. That does not mean every URL on a website should be indexed.

Check: inspect important URLs in Search Console. Review access failures, indexing directives, canonical choices and internal links. A “noindex” instruction may be deliberate on utility pages and accidental on a key service page.

Act: investigate the specific reason before changing settings across the site. Google’s technical requirements explain basic eligibility; meeting them does not guarantee indexing or rankings.

10. Breaking existing search paths during a redesign

A new design creates avoidable risk when established URLs disappear, useful content is removed or navigation stops linking to important pages.

Check: before launch, inventory existing URLs, organic landing pages, downloads and enquiry routes. Decide which addresses remain and which genuinely need to change.

Act: preserve useful URLs where possible. Map necessary moves to relevant replacements, test redirects and compare the live site with the inventory. Keep a recoverable backup. Visual approval alone is not a migration check.

11. Buying links without examining their purpose

A package promising a large link count can distract from the questions that matter: why does the link exist, is it relevant, and does the arrangement comply with search policies?

Google lists buying or selling links for ranking purposes within its link spam policies. Legitimate advertising and sponsorship require appropriate link qualification.

Act: ask a supplier to explain placement methods, relevance and paid-link handling. Prefer genuine relationships and useful contributions over a numerical target. Do not assume every directory is harmful or every editorial-looking placement is sound.

12. Treating structured data as a visibility shortcut

Structured data can support eligibility for relevant search features, but valid markup does not guarantee a rich result. Google’s structured data guidelines also require it to represent the page truthfully.

Check: does the selected type suit the content, match what visitors can see and meet the applicable feature requirements?

Act: fix inaccurate markup and validate relevant implementations. Do not add invented reviews or mark up information simply because a plugin offers the option. Check current feature availability before making promises about search appearance.

13. Selling yourself a separate set of AI-search tricks

AI search deserves attention. It does not justify every newly labelled tactic. Google says there are no additional technical requirements or special optimisations necessary for AI Overviews and AI Mode. Its guidance continues to emphasise established SEO fundamentals.

Act: make important information accessible, accurate and useful. Publish evidence and comparisons that help people decide. Do not confuse an occasional AI citation with a dependable acquisition channel, or extrapolate Google’s guidance to every other AI service. Evaluate each platform separately.

14. Counting a WhatsApp click as a qualified lead

A click can open an app without producing a message. A message can ask for something you do not sell. A qualified enquiry can still fail to become a sale.

Check: distinguish four stages in reporting: interaction, received enquiry, qualified opportunity and completed sale. Define “qualified” with the people handling sales.

Act: label phone and WhatsApp clicks accurately. Where feasible and appropriate, reconcile received enquiries with sales records without exposing customer data. Keep unattributed business separate rather than assigning every sale to SEO. This distinction prevents an attractive activity report from concealing a weak commercial result.

15. Testing the form submission but not the received enquiry

A success message shows that something happened in the interface. It does not prove that the right person received the enquiry or can reply to it.

Check: send a clearly labelled test, confirm delivery and check the details needed for follow-up. For a shop, test the supported checkout route using an appropriate test method.

Act: assign an owner to enquiry checks and repeat them after relevant updates. If reliability is the issue, include it in your WordPress maintenance scope. Acquiring more visitors will not repair failed delivery.

16. Leaving buying questions for the sales team to answer repeatedly

A visitor may understand your service and still hesitate because the scope, exclusions, location coverage or process is unclear.

Check: ask whoever handles enquiries which questions recur and which misunderstandings waste time. For a tour, that might concern inclusions or pickup arrangements; for an appliance, delivery, installation or warranty terms. These are examples, not assumptions about your customers.

Act: put verified answers near the relevant offer. If a fixed price is unsuitable, explain the factors that determine a quotation. Specific information can also help unsuitable customers recognise the mismatch before enquiring.

17. Expecting instant results—or accepting an indefinite wait

Some changes can be verified immediately: a form reaches the inbox or a broken link works. Search visibility and commercial outcomes require observation over time. No universal timetable can account for every website, market and starting position.

Act: separate delivery milestones from outcome measures. Record what changed, when it changed, the expected effect and the review period. Compare comparable periods and note other changes. A before-and-after improvement may support a hypothesis, but it does not establish causation by itself.

How to decide what deserves the next budget

Start with faults that prevent access, enquiries or purchases. Then improve the pages closest to a real buying decision. Expand content and promotion around an offer that customers can understand and act on. For a new website with little visibility, discovery work will still be essential; this sequence is a starting point, not an inflexible rule.

For each proposed task, write down:

1. The observed problem: what evidence shows it exists?
2. The affected journey: which customers and pages does it concern?
3. The intervention: what will change, and who owns it?
4. The trade-off: what will wait while this work happens?
5. The verification: how will you confirm delivery and assess the outcome?

For example, a supplier might discover that its main product page attracts relevant visits but omits delivery charges. Adding accurate delivery information is a reasonable test before commissioning ten unrelated articles. If the page barely receives relevant visits, that same change may be useful but insufficient. The evidence changes the priority.

Bring the problem, not a shopping list of SEO tasks

If your website gets visits but too few useful enquiries—or you cannot tell where the problem begins—contact SEO Consulting Kenya. Share your website, the service or products you want to grow, and what you currently measure. That gives the conversation a concrete starting point.

Editorial note: the diagnostic and priorities in this article are editorial recommendations. Illustrative scenarios are not client case studies. Technical claims link to primary documentation; no ranking or revenue outcome is guaranteed.