Blog / Case studies /
What I learned about client websites in 2024
No client website arrives as a blank, tidy project. There’s a theme someone chose years ago, plugins with overlapping jobs, content from five different people and an owner who needs the site to keep working while you improve it.
A year of client sites at Nivonix has taught me that technical SEO is often as much coordination as code.
Every change belongs to a real site
The work at Nivonix includes WordPress, Shopify and other CMS client sites, technical SEO audits, uptime, DNS and SSL, Cloudflare, plugin conflicts, client training and coordination with designers and writers. Each area touches another. Changing a theme can affect headings and internal links. A new plugin may alter page output. A DNS adjustment can make a site unavailable if the records are wrong. The work is not a neat list of isolated tasks.
That is why I try to understand the current setup before recommending a change. What platform is in use? Who manages hosting and DNS? Which plugin or app controls the relevant feature? Is there a staging environment? Who needs to approve the change? A technical recommendation that ignores ownership and workflow may be correct in theory and impossible to carry out safely.
Audits are useful when they lead to decisions
An audit can produce a long inventory of warnings. The useful part is deciding what matters, who can act and what evidence confirms the fix. A broken canonical on a key landing page isn’t the same priority as a minor warning on a throwaway URL. I want each recommendation tied to a page or task the client understands.
The audit also needs context. A crawler may report URLs that are deliberately excluded, duplicate by design or used for one campaign. The report isn’t wrong; the reviewer has to check how the site is meant to work. Findings without that review create busywork and erode trust.
Reliability is part of the experience
Uptime, DNS and SSL sound distant from content strategy until a visitor cannot reach a page or a browser shows a security warning. Monitoring and careful changes help teams notice issues, but no setup removes every risk. When something does break, the response starts with confirming scope and checking what changed, not with a dramatic assumption that SEO is the cause.
The same steady approach applies to plugin conflicts. Two tools can both try to manage redirects, caching or structured data. Disabling things at random may make the immediate symptom disappear while introducing another problem. I prefer a controlled check: identify the affected pages, review recent updates, isolate the likely conflict and record the outcome. If a rollback is needed, the team should know how to do it.
Communication keeps work moving
A designer, writer, developer and site owner may use different words for the same page. Before a change, agree on the URL, the intended outcome and what must stay intact. During handover, explain what was changed and where to look if it needs maintenance. A short note can prevent someone from undoing a fix because they did not know why it existed.
Client training matters for the same reason. A team should be able to make routine updates without fearing the CMS. That does not mean everyone needs to know every technical detail. It means showing people which fields they own, which changes affect search or layout, and when they should ask for help. Training is a way to make the site less dependent on a single person’s memory.
Work with the system you have
It’s easy to recommend an ideal stack that ignores budget, skills and existing commitments. Client work asks what the safest improvement is for this site and its maintainers. Sometimes that’s a technical fix; sometimes a clearer page brief or a better approval process. Small, understood improvements outlast a big change nobody can support.
I am still learning from the differences between sites. The same symptom can have different causes depending on the CMS, theme, hosting and history. That is a reason to investigate, not a reason to reach for a favourite tool immediately. A good process leaves room for evidence and for the client’s knowledge of the business.
The lesson I’m carrying into another year is simple: a healthy website is maintained by people who understand the decisions behind it.
Respect the boundary between learning and proof
Working on several kinds of CMS gives me a perspective on recurring operational questions, but it does not mean every site behaves the same way. A lesson from one audit can suggest what to check elsewhere; it cannot prove the same cause or outcome on another client site. I try to keep that boundary clear when discussing experience. The useful takeaway may be a process—check the deployment history, identify who owns the DNS, test the actual page—rather than a claim that a particular change produced a specific ranking result. That kind of process is portable without exposing client details or overstating what I know. It also leaves room for the client’s own history, which often explains why a site is configured the way it is.
What part of your site would benefit most from a clearer handover or a more careful change process?