Blog / Web management /
How I plan a client’s SEO year
A client asks what SEO work will happen this year. “We’ll keep an eye on it” is not a reassuring answer.
Planning gives the work a shape: business priorities connected to pages, technical tasks and a regular review. Working across client sites in my Nivonix role has made one thing clear to me. The plan has to fit the site and the people who maintain it.
Start with context, not a universal checklist
Before setting tasks, understand the business and its site. What does it offer? Who does it want to reach? Which pages already help? Who can edit the CMS, approve a change and check the result? A Shopify shop and a WordPress service site both need SEO, but their workflows and risks differ.
I also ask what changed recently. A redesign, new theme, domain change, catalogue update or new analytics setup shifts what deserves attention. Without that context it’s easy to recommend a long list of plausible tasks nobody can schedule.
Agree on the business outcome
SEO is a route to discovery, not the outcome. Ask what the business needs more of: qualified enquiries, product sales, awareness of a service, fewer support questions. Then agree how the team will recognise progress. With no reliable baseline, say so and make measurement phase one rather than inventing a target.
A clear outcome helps separate useful work from activity. Publishing more pages is not automatically useful. A site may need to improve the pages it already has, fix an indexing problem or make its offer easier to understand. A business owner should be able to see why a proposed task belongs in the plan.
Review the site in layers
I look at the foundations first: important URLs, crawling and indexing, redirects, canonical signals, mobile experience, page templates and analytics. Then I inspect the pages that matter to the business. Do they explain the offer, answer practical questions, link to the next step and reflect how the service actually works?
Technical audits throw up a lot of alerts. Each finding needs context before it becomes a task. Is the URL important? Is the behaviour intentional? Does it affect customers or discovery? If nobody knows, the task is investigation, not a fix. That stops clients paying for changes based on a tool’s severity label.
Turn findings into owned tasks
Every action needs an owner, a dependency and a definition of done. “Improve product pages” is too broad; “review the top collection’s product information for missing compatibility and delivery details” is clearer. “Fix the site” is no handover. Name the template or URL and the behaviour expected afterwards.
Some tasks need the client’s knowledge, such as service areas or product specifications. Others need developer access, hosting support or a content review. I make those dependencies visible, and when I need information from a client I name the missing input rather than assume. A plan becomes more credible when it acknowledges what is not yet known.
Put review dates on the calendar
A yearly plan should not mean one large annual meeting. Break work into a sequence that the team can deliver and review. A technical fix may need to happen before a content update; analytics validation should follow a tracking change. Keep a short record of when changes went live so later reporting has context.
At each review, ask what was completed, what changed, what the evidence says and what should be adjusted. Compare appropriate periods and note seasonality, campaigns and other releases. Search Console can show visibility, while GA4 can report collected sessions and key events. The business’s own records may explain whether the enquiries were suitable. No single dashboard gives the full answer.
Leave room for real life
A plan that assumes unlimited time will fail. Client sites have busy periods, approvals, dependencies and unexpected technical work. Prioritise a few high-value actions and keep a backlog for the rest. If an urgent DNS or SSL issue appears, the plan may need to pause while the site is made reliable. That is not poor planning; it is maintenance responding to reality.
A yearly plan should give the client clarity without pretending every unknown is solved in advance.
Make approvals part of the workflow
Many SEO tasks need approval because they affect how a business describes itself. A writer may need the service owner to verify claims; a developer may need the client to approve a redirect map; an analytics change may require consent review. Put those approvals beside the task and allow time for them. Otherwise the plan quietly assumes that the person doing SEO can make decisions on behalf of everyone else. I also try to define a rollback or escalation path for changes with technical risk. A simple note about who to contact when a page breaks can prevent delay and confusion. Planning is not only about what should happen; it is also about making sure the people involved know how to move the work forward.
What’s the first SEO action your team can own and complete this quarter?