More listings, higher domain rating. Direction unproven.

We track products by reading launch directories: which directory lists a product, whether the outbound link is followed, and when the listing appeared or went away. Separately we store a domain rating for each product, read from Ahrefs' free public endpoint with the date we read it. One number is ours and one is borrowed, and this post is about what happens when they are put next to each other.

The two together answer a question founders ask in the form of advice they have been given: does getting listed on more directories build authority. Our store can show whether the association exists. It cannot show the direction, and the second half of this post is about why.

The association

Every product below holds at least one live listing on a directory we crawl, and carries a domain rating read on 20 August 2026. That is 17,292 of the 17,298 products we track; six lookups failed on network errors and those products carry no rating rather than a guessed one.

Median domain rating rises at every step. A product on one directory sits at DR 19; a product on seven or more sits at DR 86. Nothing in the data is ambiguous about the pattern itself.

WITH p AS (
  SELECT pr.id, pr.dr,
         count(l.id) FILTER (WHERE l.disappeared_at IS NULL) AS live
  FROM products pr
  LEFT JOIN listings l ON l.product_id = pr.id
  WHERE pr.dr IS NOT NULL
  GROUP BY pr.id, pr.dr
)
SELECT live, count(*), percentile_cont(0.5) WITHIN GROUP (ORDER BY dr) AS median_dr
FROM p WHERE live > 0 GROUP BY live ORDER BY live;

Why we are not telling you to submit more

Read as advice, the table says: get listed in seven places and your domain rating goes to 86. Our data does not support that sentence, and three things in it point the other way.

The direction is unestablished. Every measurement here was taken on one day. A cross-section cannot separate listings raising a rating from a rating attracting listings, and the second reading is the one with a mechanism behind it: directory curators add products they have heard of, and products people have heard of already hold links from elsewhere. The domains in the top bucket are recognisable ones. That is what selection looks like.

The top bucket is 60 products, out of the products we have seen listed, read on 20 August 2026. A median of 86 across 60 rows is not a rate, it is an observation about 60 domains. It moves if a handful of them change.

Domain rating is not a measurement of ours. It is Ahrefs' estimate of a backlink profile, and their profile is built from links across the whole web, of which directory listings are a small and mostly nofollow part. A number that mostly reflects links we did not observe cannot be used to price the links we did.

What would actually answer the question

The store is append-only: a listing row records when a link appeared and, if it goes, when it went. Ratings carry the date they were read. So the honest version of this question is answerable later, by measuring the same products over time: take products whose first listing appeared after we started watching, and see whether their rating moves in the months after, against products that stayed unlisted.

That study needs months of history and repeated ratings on the same domains. We hold one rating per product today. When there are enough of them to compare against themselves, the result will be published here whichever way it points.

What this data cannot tell you

The corpus is products discovered through directory listings, so it is not a sample of startups in general. It is a sample of startups that submitted somewhere, which is exactly the population most likely to have submitted more than once.

We deliberately did not cross domain rating with our headline metric, the share of a directory's listing pages that Google has indexed. Two of the 206 directories in the catalog currently hold a measured indexation rate, and a claim about the relationship between two numbers cannot rest on two rows. That gap is ours and it is being closed by measurement, not by writing around it.

Listing counts are what our crawler read on the directories it can read. A product listed somewhere we do not crawl is undercounted here, which pushes products down the table rather than up it.

Every rating shown is Ahrefs' number, dated, and shown as given. The listing counts are ours. The catalog always holds the current values with their own checked dates, and how each one is measured is written down separately.