A content programme has an output number that everybody watches and an outcome number that almost nobody checks. The output is how many posts went live. The outcome is how many of them exist as far as a search engine is concerned. When those two numbers diverge, every report you produce is measuring effort.
Answer first
Open the Pages report in Google Search Console, compare the indexed count against what you have actually published, and read the reasons given for the excluded URLs. If a large share of your library is not indexed, stop the publishing calendar. The fix is almost never more posts. It is consolidating the pages that duplicate each other, linking properly to the ones worth keeping from pages that already have authority, and removing or merging the ones that were only ever written to hit a monthly count.
Publishing is a claim, indexing is an acceptance
Publishing puts a page on your server. That is the entire achievement. Everything else is a sequence of decisions made by somebody else.
A crawler has to discover the URL, usually because something linked to it or a sitemap listed it. It has to be willing to spend a request on fetching it. Then, and this is the step people skip, a system has to decide the page is worth keeping in the index at all. That last decision is a judgement about value, and it is made against everything else already in there on the same subject, including the rest of your own site.
A page that fails at any of those points is not a page. It cannot rank, it cannot be cited by an answer engine, and it cannot receive a link from anywhere that matters. It is a file in a filing cabinet in a room nobody has the key to.
The check, in about ten minutes
Go to Search Console, open Indexing and then Pages. You get two totals, indexed and not indexed, and beneath them a list of reasons.
Compare the indexed total with the number of posts your team believes it has published. Those two numbers being far apart is the finding, and it is worth doing before you read a single reason, because the size of the gap decides whether this is a tidying job or a strategy problem.
Then read the reasons. The ones that matter most in a content library are usually these. “Crawled, currently not indexed” means the page was fetched and judged not worth keeping, which is a quality verdict rather than a technical fault. “Discovered, currently not indexed” means it was not even worth fetching yet, which usually points at crawl budget and weak internal linking. “Duplicate without user-selected canonical” and “Alternate page with proper canonical tag” mean something else on your own site is doing the same job. “Excluded by noindex tag” is the easy one, and it is worth checking first because it is occasionally the whole answer and takes a minute to rule out.
Two habits make this diagnosis more useful. Filter to the URL prefix for your blog rather than the whole site, so shop pages and paginated archives do not drown the signal. And export the list, because the pattern in which pages failed is usually more informative than any individual reason.
Do not use a site: query in Google as your measurement. It is an estimate, it is generous, and it has ended more of these investigations prematurely than any other single habit.
Why it happens to content programmes specifically
The causes cluster, and they are largely produced by how the content was commissioned.
The pages are near-duplicates of each other. A brief that produces twenty posts on twenty tiny variations of one question produces twenty pages that answer the same thing. A search engine only needs one, and it will pick one, and the other nineteen become the exclusion list. This is the most common cause in a large library and the hardest one to hear, because those nineteen pages were paid for.
Nothing links to them. A post published, shared once and then reachable only from page seven of an archive is effectively orphaned. Internal links are how importance is communicated, and a page with no internal links from anywhere that matters is a page you have told the crawler not to care about.
Crawl budget is spent on the wrong URLs. Tag archives, category pages, paginated variants and parameter versions of the same page can absorb an enormous share of the attention a site receives. The crawler is not refusing to look at your good page. It is busy looking at forty near-copies of a listing.
The page is genuinely thin. Six hundred words assembled from what was already on page one of the results, with nothing specific in it, is not a page a search engine needs. This is the same standard an answer engine applies when it decides whether to cite you, and it is only getting stricter.
Why publishing more would have made it worse
The natural response to weak content performance is more content. In a library with an indexation problem, that response is actively harmful.
If the reason your pages are excluded is that they compete with each other, another twenty posts on adjacent topics adds twenty more competitors for the same crawl attention and the same slot. If the reason is thin pages, more thin pages lowers the average quality of the site as a whole, which is a signal that applies to everything on it, including the pages that were working.
There is a compounding version of this that catches good teams. The monthly publishing target is the thing being measured, so it keeps being met, and every month the gap between published and indexed grows a little larger while the report stays green. Volume feels like progress precisely because it is the only number moving.
The order that works is unglamorous. Fix indexation on what exists. Consolidate the pages that overlap into one page that deserves to exist, and redirect the rest into it. Link to the survivors from pages that already have authority. Only then start publishing again, and at a rate the site can actually get indexed.
What we found, and what we cannot claim
We audited the content of a home services company in New Jersey. It had published 243 posts. When we checked Search Console, 110 of them were not indexed.
Nearly half a content programme, paid for, written, uploaded, and invisible. Nobody had done anything wrong in an obvious way. The calendar had been met every month, the posts were on relevant subjects, and the reporting had been counting the wrong thing for long enough that the gap had become structural.
Being straight about the limits of that: this programme is proposed rather than delivered. We have not run the fix and we therefore have no results number to publish, and we would rather say that than borrow a percentage from somewhere. The 110 is an observation from an audit, which is a different and much weaker claim than an outcome, and we will publish an outcome when there is one to publish.
That distinction is the same discipline we apply to attribution, where we grade the evidence into tiers rather than presenting one confident number. A finding is not a result.
Where this sits
Indexation is a chain of custody problem wearing different clothes. In advertising, the break is between a click and a sale record. In content, the break is between the page you paid for and the page a search engine will actually serve, and in both cases the reporting keeps looking healthy because it is measuring the half that still works.
The method we run starts the same way in both: establish what can be proved before spending anything else. For content that means one number, indexed pages, checked against one other, published pages, before a single new brief is written.
The rest of the programme is downstream of that answer.