On November 5, Ahrefs previewed DR4, a new model for its Domain Rating score. It reaches the standard Ahrefs interface on December 1. Founders pay directories partly for that number, so we wanted to know what it does to the places they list on.
We read both scores for 2,026 sites on November 7: launch platforms, SaaS and AI directories, startup databases, marketplaces, newsletters, communities and press. The result is sharper than we expected, and it includes us. Aback Launch is a launch platform, and we fall from DR 43 to DR 25. Keep that interest in mind as you read.
The short version
- Launch platforms fell the most. The median one dropped 28 points, from DR 45 to DR 10. Across our sample, 93% of them scored lower.
- High scores became rare among them. Launch platforms at DR 50 or above went from 80 to 18. Those at DR 70 or above went from 34 to 7.
- Popularity separated who held from who fell. Directories at DR 70 or above that appear in a public top 1 million popularity list had a median change of 0. Those that do not appear fell about 40.
- Link-back directories fell hardest. SaaS directories that list you in exchange for a link back fell a median of 27 points. The others fell 4.
- Dofollow or nofollow made no difference. Launch platforms that give dofollow links fell 28 points; those that give nofollow links fell 27.
- Young startups' own scores may collapse too. In a small sample of 51 recent launches, the median site went from DR 26 to almost zero.
- None of this changes your Google rankings. DR is a third-party estimate. Ahrefs is changing how it measures, not how Google ranks.
What DR4 changes
Domain Rating estimates the strength of a site's backlink profile on a scale of 0 to 100. Ahrefs describes two main changes in DR4:
- Links from sites with search traffic count for more. DR is still calculated from links, but more weight goes to links from sites Ahrefs estimates get organic search traffic. Ahrefs says about half as many sites have a DR4 score as had a DR3 score, so fewer sites can pass a score on.
- Subdomains get their own score. A site on a subdomain no longer inherits its parent's number. Ahrefs' own example is help.ahrefs.com at DR 75, against DR 92 for ahrefs.com.
DR4 is a preview until December 1, and Ahrefs says you can still compare DR3 through its API for a short period afterwards. If you want to see any site's two scores now, our free domain authority checker shows them side by side, and the bulk checker handles up to 20 sites at a time.
How we ran the study
We combined two open lists of places founders submit to: Listaify's dataset (CC BY 4.0), which records each place's type, price model and link type, and the awesome-directories-list (MIT). After removing duplicates and sites those lists marked as dead, that left 2,074 hosts.
For each host we asked Ahrefs' free Domain Rating API for the default score (DR3) and the DR4 score, on the same day. 2,026 answered on both. To test the traffic question independently of Ahrefs, we matched each domain against the Tranco top 1 million, a research ranking of popular domains. We report medians, because a few extreme sites would distort an average. The limits of all this are listed further down.
Finding 1: launch platforms fell the most
| Type of site | Sites | Median DR3 | Median DR4 | Median change | Scored lower | Lost 20+ |
|---|---|---|---|---|---|---|
| Launch platforms | 199 | 45 | 10 | -28 | 93% | 66% |
| SaaS and review directories | 374 | 48 | 28 | -6 | 69% | 28% |
| AI directories | 634 | 26 | 12 | -5 | 74% | 21% |
| Newsletters | 135 | 54 | 42 | -5 | 76% | 13% |
| Startup databases | 65 | 54 | 48 | -2 | 62% | 11% |
| General directories | 84 | 48 | 39 | -2 | 58% | 13% |
| Marketplaces and app stores | 176 | 87 | 83 | -2 | 65% | 6% |
| Communities | 40 | 68 | 53 | -2 | 52% | 5% |
| Press | 319 | 76 | 76 | 0 | 37% | 2% |
Domain Rating by Ahrefs. The median change is the middle site's own change, so it does not always equal the gap between the two median scores.
The gap between launch platforms and everything else is large. Among sites that started between DR 40 and DR 79, the median launch platform fell 34 points, and 79% of them lost 20 or more. The median SaaS directory in that range fell 7.
The score bands moved with it:
| Launch platforms on their own domain (197) | On DR3 | On DR4 |
|---|---|---|
| DR 70 or above | 34 | 7 |
| DR 50 or above | 80 | 18 |
| Under DR 10 | 12 | 98 |
Nearly half of them now sit under DR 10. Across all 1,356 directories in the sample (launch platforms, SaaS, AI, general and startup databases), the number at DR 50 or above fell from 448 to 279, and the number under DR 10 rose from 223 to 537.
Finding 2: well-known names moved very differently
Here is a sample of platforms founders will recognise, sorted from the smallest change to the largest.
| Site | DR3 | DR4 | Change |
|---|---|---|---|
| Tiny Startups | 71 | 74 | +3 |
| DevHunt | 70 | 71 | +1 |
| Product Hunt | 91 | 91 | 0 |
| Peerlist | 78 | 72 | -6 |
| BetaList | 77 | 70 | -7 |
| PitchWall | 74 | 67 | -7 |
| Uneed | 77 | 62 | -15 |
| Microlaunch | 64 | 49 | -15 |
| Fazier | 85 | 69 | -16 |
| TrustMRR | 73 | 57 | -16 |
| Launching Next | 56 | 40 | -16 |
| Aback Launch (us) | 43 | 25 | -18 |
| PeerPush | 79 | 60 | -19 |
| TinyLaunch | 74 | 52 | -22 |
| StartupBase | 75 | 46 | -29 |
| Outbid | 75 | 41 | -34 |
| Open Launch | 71 | 29 | -42 |
| ScrollLaunch | 73 | 30 | -43 |
| What Launched Today | 70 | 8 | -62 |
Domain Rating by Ahrefs, read on November 7, 2026. Outbid (outbid.lol) was checked separately and is not part of the 2,026. DR4 is a preview, so scores can change before it rolls out. Ahrefs itself says a lower DR is not a judgment on a site, and a fall here says nothing about how a site is run or how many people visit it.
Look at the scores in the DR3 column. Sites between DR 70 and DR 77 looked interchangeable on the old model. On DR4 the same group runs from 74 down to 8. A buyer comparing them by score last month would have seen no difference.
Finding 3: popularity separates who held from who fell
Ahrefs says DR4 favours links from sites with search traffic, which should reward sites connected to real audiences. We cannot see search traffic for 2,000 sites, but we can check a separate measure of popularity. Within each score band, we split directories on their own domain by whether they appear in the Tranco top 1 million.
| Starting score (DR3) | In the top 1M | Median change | Not in the top 1M | Median change |
|---|---|---|---|---|
| DR 70 and above | 158 sites | 0 | 34 sites | -40.5 |
| DR 50 to 69 | 69 sites | -3 | 174 sites | -24.5 |
| DR 30 to 49 | 23 sites | -7 | 318 sites | -17.5 |
| DR 10 to 29 | 9 sites | -3 | 325 sites | -11 |
This is the clearest pattern in the data. Two directories that both showed DR 75 on the old model were very different underneath: the popular one kept its score, and the unpopular one lost about half of it. The smaller groups are thin (9 and 23 sites), but the direction is the same in every band.
The effect shows across popularity tiers too. The median change was 0 for directories in the top 10,000, +2 from 10,000 to 100,000, -4 from 100,000 to 500,000, -7 from 500,000 to 1 million, and -9 for those outside the list. Among the 260 ranked directories, the link between score and popularity rank also got a little tighter on DR4 (a rank correlation of 0.76, up from 0.72).
Launch platforms are a special case: even the 23 in the top 1 million fell a median of 18 points. Being known helped them, but less than it helped other kinds of site.
Finding 4: link-back directories fell hardest
Many directories list you for free if you put their badge or a link on your site. Listaify's dataset records that as a reciprocal price model, so we could test it.
| Type of directory | Needs a link back | Median change | No link back needed | Median change |
|---|---|---|---|---|
| SaaS directories | 56 sites | -27 | 313 sites | -4 |
| AI directories | 83 sites | -17 | 538 sites | -5 |
Among launch platforms, all 35 that ask for a link back fell, by a median of 34 points. That group includes us: our free plan's dofollow link depends on the founder adding our badge.
Comparing directories that started at a similar score (DR 40 to 79) makes it plainer:
| How the listing works | Sites | Median change | Lost 20+ |
|---|---|---|---|
| Free in exchange for a link back | 122 | -28.5 | 63% |
| Free, with a paid upgrade | 191 | -22 | 52% |
| Free, nothing asked | 64 | -3 | 12% |
| Paid only | 52 | -0.5 | 8% |
Price points the same way. Among directories with a listed price, those charging under $100 had a median change of -16 (574 sites), and those charging $100 or more had a median change of -2 (65 sites). The median launch platform with a listed price charges $12.
This does not make link-back listings bad, or expensive ones good. A big paid directory may simply be an older, better-known business. It does show that a badge model and a high score were closely tied together, and that DR4 pulled them apart.
Finding 5: dofollow or nofollow made no difference
You might expect directories that hand out dofollow links to be treated differently. Among launch platforms, they were not: dofollow platforms fell 28 points (96 sites) and nofollow platforms fell 27 (78 sites). The share losing 20 points or more was 66% and 67%.
A directory's link policy affects what it gives you, not what it receives, so this is what you would expect. It still matters to you: a dofollow link can pass ranking value and a nofollow one generally passes little. Just do not read a dofollow promise as a sign of a healthy site.
Finding 6: subdomains lost their borrowed score
Thirty directories in our sample live on a subdomain. Their median score went from DR 48 to DR 6, and 90% fell. Some of the largest drops in the whole study are here:
- aitools.neilpatel.com: DR 91 to DR 6
- index.dodopayments.com: DR 82 to DR 11
- sell.appsumo.com: DR 83 to DR 27
- techindex.law.stanford.edu: DR 93 to DR 35
Domain Rating by Ahrefs. Directories hosted on a form builder or a shared notes page show it most starkly. One list on a Typeform address went from DR 93 to DR 0. That 93 was never the list's own: it belonged to Typeform.
Finding 7: young startups' own scores may collapse
This is the finding that may affect you most directly, and our sample is small, so treat it as a warning and not a measurement. We checked 51 recent launches on Aback Launch that run on their own domain.
| Starting score (DR3) | Sites | Median change | Now under DR 1 |
|---|---|---|---|
| DR 50 and above | 5 | -23 | 2 |
| DR 30 to 49 | 14 | -31 | 7 |
| DR 20 to 29 | 13 | -21 | 6 |
| DR 10 to 19 | 8 | -14 | 6 |
| Under DR 10 | 11 | -1.1 | 9 |
Overall, 47 of the 51 fell, and 30 are now under DR 1. The median site went from DR 26 to almost zero. A founder can spend months building a DR 30 and watch it disappear on December 1 without losing a single link.
We now read both scores for every startup launched on Aback Launch each Monday until the switch. See which launches held their rating best, and check your own site in the free domain authority checker.
Early-stage sites are often linked mainly from directories and from other new sites, and those are the links DR4 appears to count for least. We cannot see each site's links through the free API, so that is our reading, not a measured cause.
Why this probably happened
We can show what moved, not exactly why. Here is the explanation we think fits best.
Launch platforms form a tight network. Founders submit to dozens of them, many list each other, and many ask each new product to link back. Under DR3, a steady stream of those links could lift a young platform to a high score, whether or not many people visited it. The linking sites were often new sites themselves.
DR4 gives more weight to links from sites with search traffic, and removes about half of all sites from the scoring pool. Any network built on links from sites with little traffic loses much of what held it up, and each platform's fall lowers what it passes to the next. That fits all of our findings: launch platforms fell most, link-back models fell hardest, unpopular sites fell further than popular ones at the same score, and young startups fell too.
Is listing on directories a scam?
Some founders are saying so, and we can see why. Anyone who paid for a listing because of a score of 70 now holds a link from a site that scores 30. That is a fair complaint about how those listings were sold.
The conclusion goes too far in three ways:
- The list of places being dismissed is not one group. In our data, Product Hunt, G2, Crunchbase and Hacker News barely moved. Treating them like a board that fell 60 points misses the main result.
- A score is not the value of a link. Google has never used Ahrefs' score. Ahrefs' own test found that across 10 million sites, DR4 tracks Google rankings better than DR3 does, yet for root domains alone, the group most directories belong to, DR3 still correlated better. Neither figure says what one link from one directory does for your site.
- Links are not the only reason to launch. Early users, feedback, a page that ranks for your name and a first few customers do not depend on DR. A platform with a real audience can be worth it even with a nofollow link.
So the honest verdict is narrower. Paying for a number was always a poor deal, and DR4 shows why. Listing somewhere people actually look still makes sense.
How to choose where to list now
- Check DR4, not the number on the sales page. Run the site through the domain authority checker, or paste your whole shortlist into the bulk checker and sort by DR4. A badge reading DR 70 may be out of date by next month.
- Look at the gap between the two scores. A site that holds its score has links from sites with real traffic. A large fall suggests the old number came from somewhere weaker, whatever the new number is.
- Check its popularity. The same checker shows the Tranco rank. Being outside the top 1 million is normal for a small site, but it is odd for one charging you for reach.
- Be careful with subdomains. Check the exact address your listing would appear on, not the parent brand.
- Look at recent launches on the site. Do they have comments and votes from real accounts? Would you find this page if you were looking for your own product? If nobody is reading, the link is all you are buying.
- Check what you actually get. Open a real listing and use the nofollow checker and the meta tag checker to confirm the link type and that the page can be indexed. Our launch platform database records link types where we have checked them.
- Treat a link-back request as a cost. A badge on your homepage is a real price, and it sends visitors and link strength away from you. Pay it for platforms you would recommend anyway.
- Measure what comes back. Tag your listing links, watch referral traffic and signups for a month, and only then decide whether to pay for an upgrade or a repeat.
- Stop buying bulk submission packages on DR promises. A list of 200 sites chosen for their old score is the thing our data undercuts most directly.
What to do if your own score falls
First, check it now with the checker, so December 1 does not surprise you. Then:
- Do not panic about rankings. A lower DR on December 1 does not mean Google has changed its view of your site. Judge progress in Search Console: impressions, clicks and the pages that bring signups. SEO takes months regardless of what any score says.
- Warn anyone who relies on your number. If you quote DR to advertisers, partners or investors, tell them about the change before they find it.
- Change where you earn links. Aim for sites that people search for and read: articles in your field, resource pages, data others want to cite, and tools. Our guide on getting backlinks for a new website covers these in order.
- Count referring domains that have traffic. Ten links from sites with real readers are likely to do more for you than a hundred from sites without them.
Where this leaves Aback Launch
We are not a neutral observer. We run a launch platform, we are in the group that fell most, our score goes from 43 to 25, we are not in the Tranco top 1 million, and our free plan's dofollow link asks for a badge. Those four facts match the pattern in this study closely.
Our homepage shows our live rating, so it will change when Ahrefs' number does. We would rather you judged a launch here by what you can check: who is voting and commenting, and what your listing brings you. The same goes for every other place on your list.
What this study cannot tell you
- DR4 is a preview. Scores can change before December 1. These are all from one day, November 7, 2026.
- The sample is not the whole web. It comes from two lists built for founders, weighted towards AI and startup products. Site types are as labelled in those lists, and we mapped the second list's categories onto the first's. A directory could be labelled differently by someone else.
- Popularity is not search traffic. Tranco ranks overall popularity across several sources, is measured per domain, and only covers the top million. It is a stand-in for the traffic Ahrefs describes, not the same thing.
- These are correlations. Link-back directories, cheap directories and launch platforms overlap heavily, so these effects are not independent. Our explanation is a reading of the pattern, and we did not inspect each site's links.
- The founder sample is small. It is 51 sites from our own platform, all recent, so it may not represent older or better-known startups.
- We have an interest. We sell launches, and we are named in the data. You can check any score here yourself.
FAQ
What is Ahrefs DR4?
It is a new model for Ahrefs' Domain Rating. It gives more weight to links from sites with organic search traffic and rates subdomains on their own links. It rolls out in the standard Ahrefs interface on December 1, 2026.
How do I check my DR4 score?
Enter your domain in our free domain authority checker. It shows your current score and your DR4 score together, with no sign-up. For a list of up to 20 sites, use the bulk checker.
Will a lower DR hurt my Google rankings?
Not by itself. DR is Ahrefs' estimate, and Google does not use it. Your rankings depend on your pages and your actual links, which do not change because a third party rescored them.
Why did so many launch platforms lose their score?
We cannot prove the cause, but the pattern points at where their links come from. They tend to be linked from founders' new sites and from each other, and DR4 gives less weight to links from sites with little search traffic. Platforms that ask for a link back and platforms outside the top 1 million fell the most.
Are launch directories still worth using?
Some are. A platform with an active audience can bring users, feedback and a page that ranks for your product name, whatever its score. Choose by who will see your launch, check the DR4 score before you pay, and measure what the listing brings you.
Should I still pay for a listing because of its DR?
We would not. Scores are moving now, and a high one was never a promise of value. If you do pay, pay for audience you can verify, and use the DR4 score only to rule places out.
Does a dofollow link from a site that fell still help?
It can still count, but it probably carries less than the old score suggested. If that site gets little traffic and few other sites link to it, do not expect one link from it to move your rankings.
Written by
Devvrat Hans
Founder
By subscribing, you agree to get our newsletter by email. Unsubscribe any time in one click. See our Privacy policy and Terms.