Earlier this year, I launched two websites, remote job boards with almost identical technical setups.One indexed reasonably well.The other was indexed briefly, then almost completely removed from Google.The comparison has changed how I think about launching websites with large numbers of pages.The mistake, I believe, was making too much content indexable too quickly on a completely new domain.The first site indexed without major problems.In February 2026, I launched a remote legal job board.Like most job boards, it contains a lot of pages; however, at the time I launched it, there were just a handful. Whilst Google did not rank the site particularly well immediately, it crawled and indexed a reasonable proportion of it. Over time, its search visibility has started to improve.As the site grew, I created more pages, and now the site has pages for individual jobs, job categories, locations, specialisms, and various long-tail searches. They are all indexing and serving nicely.That was roughly what I expected.I have another job board that has been online for around two years, and the experience there taught me that Google can take a very long time to trust a new site.For a long time, that older site struggled to rank for anything important. Gradually, however, Google started sending it more traffic.It now ranks in the top three for many of the main search terms I was originally targeting.Most people probably do not wait long enough to see this happen.They launch a website, receive almost no Google traffic for the first few months, and conclude that the project has failed.My experience has been that patience does eventually pay off, provided the website remains active and useful.So when the site I launched in February started indexing reasonably well and showing signs of growth, I decided to launch another one in May 2026 using almost the same system.That is where the problem began.I copied the successful site too closely.The second website was another niche job board, only this time targeting remote accounting jobs, rather than legal.Because the first site already had many long-tail landing pages indexed, I created similar pages, but targeting accounting keywords, for the new website.The technical infrastructure was almost identical.The sites used the same general:Application architectureSitemap generationStructured dataInternal linking systemJob ingestion processPage templatesIndexing rulesAt first, Google indexed a substantial number of pages on the new site.Then a few weeks later almost all of them disappeared.The site dropped to just two indexed pages. At times, it looked like effectively nothing was indexed.It has remained in that state for 2 months now!Google knows about the pages. It has crawled many of them. The sitemap works. The pages are not blocked. The technical setup is identical to the other site.Google has simply decided not to keep most of them in the index.This gave me an unusually clear comparison.When a website is not indexing, the normal response is to question the content or technical setup.People will tell you to:Improve the contentFix the canonical tagsImprove page speedAdd internal linksCorrect the sitemapRemove duplicate pagesImprove structured dataThat advice can be useful.But in this case, I have two websites running on the same infrastructure and using the same content model.Google considers the pages on one site good enough to index.Yet, on the other site, they're apparently not good enough.That tells me Google is not simply looking at the content of each page and making an isolated decision about its quality.The reputation, history, or perceived trustworthiness of the domain must also influence the decision.A page can apparently be good enough for one website but not good enough for another, even when content quality and SEO setup are close to identical. My theory about what happened.I think the newer site looked too large, too quickly.It was a completely new domain that appeared with thousands of indexable pages almost immediately.Google initially crawled and indexed many of them, then it appears to have reassessed the website and removed almost everything.I cannot prove that the number of pages was the reason.Google does not explain these decisions clearly enough for anyone outside the company to know exactly what happened.But this is my best explanation based on the comparison between the two sites.The first site launched in February and developed more gradually.The second site inherited a mature information architecture from the beginning. It launched with many of the long-tail pages that had worked on the first site.That looked efficient from my perspective.From Google’s perspective, it may have looked like a brand-new domain suddenly producing thousands of highly structured pages without any history, backlinks, traffic, or evidence that users valued them.Google may not have viewed that as an opportunity.It may have viewed it as a risk.Good content is not judged in isolation.This experience has made me sceptical of the idea that Google simply reads every page, assesses its quality, and decides whether it belongs in the index.Context clearly matters.Google may also be considering:How long the domain has existedWhether the site has a history of useful contentHow quickly the number of pages increasedWhether external websites link to itWhether the pages receive any real trafficWhether users appear satisfied with the resultsWhether the site looks like a genuine long-term projectWhether Google has any reason to trust the publisherThe content on the newer site is not substantially worse than the content on the older one.The main difference is that one domain has started to establish a history, while the other appeared almost overnight with a larger number of pages.What I would do differently nowIf I launched another large website, I would not allow Google to index everything immediately.I would launch most of the programmatic pages with noindex.I would then choose a small number of pages that gave Google the clearest reason to index the site.That would probably include:The homepageA main category pageA few important landing pagesA small number of strong long-tail pagesSeveral genuinely useful editorial articlesI would only include those pages in the sitemap.Once they were indexed and receiving impressions, I would gradually make more pages indexable.The process might look something like this:Launch with 10 to 20 indexable pages.Wait for Google to crawl and index them.Monitor whether they receive impressions.Add another small group.Continue expanding as the domain develops history and trust.This feels slower than launching everything.In reality, it may be much faster.The alternative is what happened to me: launch thousands of pages, get many of them indexed briefly, then lose almost the entire site and spend months trying to recover.I would also target the smallest keywords firstAnother mistake was focusing too much on the keywords that could theoretically bring the most traffic.That feels logical.If a broad keyword gets thousands of searches every month, it appears to be the most important target.But a new domain has very little chance of ranking for the most competitive terms.It may create a good page, but Google already has many established websites it can show instead.The result is that the new site receives no traffic at all.A better strategy may be to target extremely narrow searches first.For example, rather than starting with a broad phrase like:remote accounting jobsA new website could begin with much smaller searches relating to specific roles, locations, qualifications, or industries.Those keywords might only receive 50 or 100 searches per month.That is not necessarily a problem.Ten clicks from a small keyword are more valuable than zero clicks from a large one.If you win enough small searches, the site begins receiving real visitors.Google then has opportunities to observe whether users click the result, use the page, and appear satisfied with what they find.Without any search traffic, the website has very little opportunity to demonstrate that its pages are useful.The goal should perhaps be:zero visitors to tenten visitors to one hundredone hundred visitors to one thousandRather than trying to go directly from zero to thousands by targeting keywords, you have almost no chance of winning.A non-indexed page may still be known to Google.Another point I have been thinking about is the difference between a page being unknown and a page being excluded from the index.Google may have already crawled and analysed a page without making it available in search results.In other words, “not indexed” does not necessarily mean that Google does not know what the page contains.It may mean that Google knows what the page contains but has not seen enough reason to make it searchable.This makes highly specific long-tail content potentially useful.If a page directly answers a narrow query for which Google has few good results, it may be easier for Google to justify indexing that page than another generic programmatic landing page.That could create an initial foothold for a domain that is otherwise struggling to index.I am currently testing this idea by focusing more heavily on distinctive editorial content and very specific search queries.The Sites InvolvedThe newer site experiencing the indexing problem is Find Remote Accounting Jobs.The site launched in February is Find Remote Law Jobs.The sites are useful for comparison because they share almost the same underlying system while receiving completely different treatment from Google.The older job board I mentioned is Zero Tax Jobs. It has been online for around two years and now ranks near the top of Google for many of its main target searches, despite making very little progress during its early months.The LessonMy main conclusion is that a large website should probably earn the right to expand its indexed footprint.A new domain may contain thousands of technically valid and useful pages, but Google has no existing reason to trust them.Making every page indexable immediately can therefore work against the site.If I were starting again, I would:Launch with very few indexable pagesKeep most programmatic pages set to noindexTarget narrow, low-competition searches firstAdd pages to the sitemap graduallyExpand only after the first pages begin receiving impressions and clicksTreat indexing as something that must be earned over timePublishing everything immediately felt like the fastest option.It may have been the exact reason the site became stuck.I hope this helps others avoid the mistake I made. Happy job board building!