Search guide
A content plan built around what you sell, not around a calendar
Most content plans are a list of titles and a publishing frequency. That produces a blog. It does not produce a body of work where each page makes the next one rank.
11 min read · 7 steps · By Aman Mishra · Written August 2026
What you end up with
A plan where every page has a job, a primary term, an internal link path to a page that sells, and a date at which you will decide whether it worked.
Have these ready
- A mapped keyword list
- Search Console access
- An honest view of how much you can publish and sustain
What it takes
11 min to read
7 steps, in the order below. Skip to the one you are stuck on.
There is a version of content marketing that produces traffic and no customers. It is extremely common, and it comes from planning by calendar rather than by purpose.
The fix is not writing better articles. It is deciding, before writing anything, what each page is supposed to do and where it sends the reader afterwards.
-
Start from the pages that sell
Commercial pages come first and everything else exists to support them. A blog built before the money pages has nowhere to send anyone.
Commercial pages first. The pages describing what you do, for whom, at what sort of price, with the evidence attached.
This order matters because everything else is support. An article about how to choose a supplier is useful when it can send the reader to a page that makes the case for you. Without that page it is a good deed.
Businesses regularly have this inverted: two hundred articles and a services page that is a list of bullet points written in 2019. In that shape, more articles cannot help.
Get the commercial pages right first, then build around them.
-
Group topics around a buying decision
Cluster content by the decision a reader is making rather than by subject category, because the decision is what determines where they go next.
The usual clustering advice is to group by topic. Group by decision instead.
Somebody choosing a heating engineer is making one decision, and the questions around it are: what does this cost, how long does it take, what happens if it goes wrong, how do I tell a good one from a bad one, do I need a permit. Those belong together and they all lead to the same page.
Somebody deciding whether to repair or replace is making a different decision and lands somewhere else.
Clustering by decision rather than by subject gives you an obvious answer to where each page links, because pages in one cluster all point at the same commercial destination.
-
Decide what each page is for before writing it
Every page should have a stated job: rank for a term, answer an objection, capture a comparison, or support another page. A page with no job produces nothing.
One line per planned page, stating its job.
Rank for a term. There is demand, we can plausibly compete, this page targets it.
Answer an objection. No search volume necessarily, but every prospect asks it and having the answer on the site shortens the sales conversation.
Capture a comparison. People searching for alternatives, versus, or reviews are close to deciding.
Support another page. Depth on a subtopic, existing mainly to link to the commercial page and to make the site credible on the subject.
If a proposed page does not have one of these, it does not go in the plan. This is the filter that removes most of what would otherwise be written.
-
Write the internal links into the plan
Which page links to which, decided at planning time. Links added afterwards by whoever remembers are the reason most sites have orphaned content.
At planning time, decide which page links to which.
Two rules cover most of it. Every supporting page links to the commercial page for its cluster. Every commercial page links to the two or three supporting pages that answer its most common objections.
Doing this at planning time rather than afterwards is what prevents the standard outcome, which is a site with a hundred pages, no consistent linking, and a dozen pages nothing points at.
Internal links are also the mechanism by which one page helps another rank. A cluster of pages linking sensibly to a commercial page is doing real work for it.
-
Publish at a rate you can sustain for a year
Consistency compounds and bursts do not. Two good pages a month beats eight in January and nothing until June.
Pick a rate you can hold in a busy month.
Two solid pages a month for twelve months is a body of work. Eight pages in January and nothing until June is a burst that produces a spike and a decline, and it usually ends with somebody concluding content does not work.
Consistency matters more than volume because ranking takes months per page. A steady rate means something is always maturing.
-
Improve existing pages before adding new ones
A page in position nine has already proved Google will rank it. Improving it is usually cheaper than getting a new page to the same place.
Every month, before writing anything new, look at Search Console for pages sitting between positions five and twenty.
Those pages have already proved that Google will rank them. Improving one, adding the section it is missing, matching the search intent more closely, linking to it from somewhere relevant, is usually cheaper and faster than getting a new page to the same position.
Most content plans skip this entirely because new pages feel like progress and edits do not. The edits produce more.
-
Prune what never earned its place
Pages that have produced nothing in a year are not neutral. Deciding what to remove, merge or rewrite is part of the plan rather than an admission of failure.
Once a year, list every page and what it has produced.
Pages that have brought nothing in twelve months are not neutral. They dilute what the site is about and they take crawl attention.
Three options, in this order. Merge it into a stronger page on the same subject and redirect. Rewrite it properly if the topic matters and the execution was poor. Remove it and redirect to the nearest relevant page if neither applies.
Check before removing anything: does it have links pointing at it, and is it the only page covering something. A page with no traffic can still be holding something up, and finding that out afterwards is expensive.
When you have done all 7
A plan where every page has a job, a primary term, an internal link path to a page that sells, and a date at which you will decide whether it worked.