From Idea to Launch: How IndiesRead.it Built a Successful Niche Directory
A retrospective on IndiesRead.it, the curated directory that led to DirectoryFast: what drove its traffic, what broke, and what would have to change to build it in 2026.
Originally published in 2025. Revisited in July 2026 with what held up and what did not.
IndiesRead.it is a curated directory of books for entrepreneurs and indie makers. It reached around 2,000 monthly visitors and steady affiliate revenue, and it is the reason DirectoryFast exists at all. Building it surfaced every problem a directory owner runs into, which is what the template, and later the platform, was built to solve.
This is the honest version of that story, including the parts that would not work if you started today.
Why this niche
Books for entrepreneurs is a niche that passes the basic tests, though not all of them cleanly.
The decision is genuinely hard. There are thousands of business books and most are bad. Someone who wants a book on pricing for a solo consultant is not served by a bestseller list.
The data takes effort. Curating means reading, or at least reading enough to judge. That is slow, which is precisely why the set had value.
Where it is weaker: the entries do not pay to be found. Authors and publishers were not going to buy placements, so the monetisation path was affiliate from the start. That capped the ceiling in a way a service directory would not be capped.
If you are choosing a niche now, this is the trade-off to notice. Content directories are easier to fill and harder to monetise. Service directories are the opposite.
What actually drove traffic
Not the homepage. Individual entry pages and tightly scoped category pages, consistently.
Pages like "books on pricing for freelancers" or "books about building an audience" attracted people who knew exactly what they wanted and could not find it any other way. That is the long tail working as designed, and it is the same pattern in every directory niche.
The homepage got branded traffic and referrals. It was never the acquisition engine, and treating it as one wastes effort that belongs on category pages.
What worked
Depth in one category before breadth. The first version covered a narrow slice properly rather than every business topic thinly. That is the single most transferable lesson, and the one most new directory owners ignore.
Curation as the product. The value was in what got left out. A directory that includes everything is a search engine with worse results. Entries were judged, not aggregated.
Community distribution. Indie maker communities were where the audience already gathered. Sharing there produced early traffic and, more usefully, early feedback about which categories people actually wanted.
Affiliate as a fit. Books have affiliate programmes. The model matched the entries, which is the only reason affiliate worked here at all.
What did not work
Manual maintenance. Every update meant editing content and redeploying. That friction is invisible in month one and decisive in month twelve. Corrections that took five minutes got postponed, then forgotten, and the data drifted.
This is the failure mode that killed more of the directories built on the original DirectoryFast template than anything else, and it is why instant publishing became a hard requirement for the platform version. A change should be live at the next read, not after a rebuild.
Loose entry structure. Entries had a description, a link, a few tags. That was enough to read and not enough to compare. Real filters were impossible because the underlying data was not typed consistently.
Retrospectively this is the biggest miss. If every entry had carried the same typed attributes from the start, topic, stage, format, depth, applicability, the filtering would have been genuinely useful and every generated page would have carried structured data worth indexing.
That lesson is why DirectoryFast now designs the schema before it fills anything.
Monetisation ceiling. Affiliate on books is low margin. Traffic went up and revenue barely moved. A niche where entries pay for visibility would have converted the same traffic into several times the revenue.
What would be different in 2026
Three things have changed since this was built, and they change the playbook.
Thin aggregation stopped working. Google's March 2026 core update enforced its scaled content abuse policy hard, and directories built on template pages with minimal differentiation lost most of their rankings. The original entry pages would sit uncomfortably close to that line today. Typed attributes per entry are no longer optional polish, they are the indexation requirement.
AI Overviews absorbed the informational queries. "What are good books for founders" is now answered without a click. What still ends in a click is filterable comparison, which is an argument for building real facets rather than curated prose lists.
The setup is no longer the hard part. Six months of building in 2025 is an afternoon now, on any of several platforms. The work has moved entirely to the schema, the curation and the maintenance, which were always the parts that mattered.
The transferable lessons
Five things that apply to any directory, in any niche.
- Design the typed schema before you fill anything. It determines what your filters can do and whether your pages are differentiated enough to index. It is close to impossible to retrofit.
- Go deep in one category first. Fifty well-filled entries in a tight category beat five hundred thin ones across fifty.
- Pick entries that pay to be found, unless you are genuinely fine with affiliate-level returns.
- Judge how easy maintenance will be in month eighteen, not month one. Every platform is pleasant to use on day one.
- Category pages are the acquisition engine. Write real introductions on them. The homepage is not where the traffic comes from.
FAQ
Could you replicate this today?
The niche is more crowded and the SEO bar is higher. A narrower slice with properly typed entries would work better than a broad curated list.
How long did it take to get traction?
Months, not weeks, which is normal on a new domain. Meaningful organic traffic on a fresh directory generally takes three to six months.
Was affiliate worth it?
It covered costs and proved the traffic was real. It was never going to be a business on its own, and that was a niche choice rather than a model failure.
What is the one thing you would do differently?
Type the data model first. Everything else follows from it.
Describe your niche and see the data model an agent designs for it →