promote-your-next-startup: a curated link list for launching a small product

An opening argument that the list itself undercuts

The first thing in the README is a quote and it is not a link. It argues that the main advertisement channel is people who are in love with your product, that their sharing is the best advertisement, and that the correct order is to build something useful, usable and attractive before spending any effort telling people it exists.

A repository of promotion links that opens by arguing promotion is downstream of product quality is either self-aware or inconsistent, and the text does not resolve which. What it actually does is set the order of operations. Build first, then work down the tables. Every section that follows is a distribution channel, and none of them is a strategy for deciding what to build.

That framing has an honest appeal for a small project. Launching into a few well-matched communities beats being ignored in a large one, and the tables are organised so a reader can find the channel that matches the audience rather than the channel with the largest audience. It also means the list is worth reading for its taxonomy rather than for its links, because the categories are the part that will not go stale.

Posting news, grouped by what the audience is

The Post News table is the longest section and the most evenly mixed. It leads with reddit.com described as the front page of the internet, then Hacker News, Lobsters and Slashdot, which is an implicit ranking by how technical the audience is rather than by size. The Hacker News entry is labelled Hacker news, Lobsters gets its own name, and Slashdot is described as news for nerds.

Then the mix broadens: Facebook is annotated with post news to your niche groups, Twitter gets Of course, and the rest are community-specific. There is a developer community (DevRant), a data science board (DataTau), a designer network (Designer News), and two link roundups that publish on a weekday schedule, Sidebar.io for design and CSS Weekly for CSS articles and experiments.

Facebook and Twitter standing next to each other with no argument for either is a useful signal about the age of the list. Both are listed without qualification, which tells you the entries were assembled when posting there was unremarkable. Nothing in the README explains how to choose between them or what makes a submission land, which is the limit of a list this shape.

Product pages, and the Gartner cluster in the middle of it

The Create a Product/Tool Page section is the one with the most interesting interior structure, because it is not uniformly one kind of site. It opens with Product Hunt and SaaSHub, then AngelList and AlternativeTo, then Indie Hackers. Those are directories where a submission gets you listed.

Then the character changes. Capterra is annotated as a Gartner-owned software directory with high commercial-intent traffic, and two more entries follow it with the same structure: Getapp described as a Gartner Digital Markets sister to Capterra, and Softwareadvice as a third Gartner property. SaaSworthy, Tekpon and G2 follow, then the design galleries, Awwwards, One Page Love and Land-book. Rounding it out are Terminal Trove for command line tools, Peerlist for makers with a Launchpad, and Stackshare for publishing your tech stack.

The Gartner entries are annotated with traffic intent, which is the only place in the whole README that a channel is described in terms of the audience it sends rather than what it is. That annotation is worth reading as the author's judgement about which listings bring people who can pay. It is also the single most dated-looking part of the list, since the commercial-intent framing of those directories has shifted considerably. The design galleries and the launchpad entries serve different purposes again, and the section never says which category a given product belongs in.

Articles and niche platforms as a longer bet

The Post an Article section is broader still and includes the same sites that appear elsewhere, Medium, LinkedIn and Dev.to among them, alongside a spread of regional and vertical communities. DZone, CodeMentor, Hacker Noon and Morioh sit in the developer column. KDNuggets and Towards Data Science carry the data science audience, and Smashing Magazine and the Free Code Camp Forum carry the web development one.

Two entries stand out for annotation rather than name. LinkedIn is described as people and companies posting and looking for job, with the instruction to post to your niche groups as well, which identifies it as a place to find people rather than readers. MuckRack is listed as a service for journalists and public relations, which is the only entry in the entire list aimed at getting covered by someone else rather than at readers arriving directly.

The Niche Platforms subsection underneath is a small table of three: GeeksforGeeks as a computer science portal, Codeforces as a programming contests platform, and Kaggle as a home for data science. These are places where a technical audience already gathers. Nothing in the README connects them to the article table above, so the practical reading is that a project with a data science or competitive programming component has three more places to write, and everything else does not.

Writing for news blogs and the parts of the list that are not about promotion

The Write for News Blog table is a different kind of list from the ones above, because most of these are publications you pitch to rather than sites you post on. It links to Medium's Daily JS newsletter, ITNext, A List Apart, The Next Web, Developer Tech, ProgrammableWeb, Hakin9 security, DataCamp Community, EchoJS and CSS-Tricks. Several entries go directly to a contribution or contact page rather than the homepage, which is the tell: these are open submission venues.

What the table does not contain is any guidance on what to pitch, how to pitch, or what the editors want. A reader gets a set of URLs and no template. The same is true of the sections that follow, and at some point the README stops being a promotion resource and becomes a bookmark collection.

Two things in the repository sit outside that. The tree holds `.gitignore`, `CODE_OF_CONDUCT.md`, `CONTRIBUTING.md`, `LICENSE`, `README.md` and a single `assets/` directory, which holds the header image. The Unlicense is the license field, and for a list of links that is a reasonable choice, though it means the list carries no warranty of any kind and no mechanism obliges anyone to keep it current. Contributions are invited through the usual community health files, which is how a list like this stays alive at all.

What a repository with no code can and cannot tell you

There is no language field, no release history and no source code here, and the topics are the honest summary of what the project is: free, links, pitch, promote, promotion, spread, startup, startup-directories and startups. A reader looking for a tool will find none, and a reader looking for a curated directory will find exactly what was promised.

The strength of the format is coverage. Few single pages assemble this many channels across news aggregators, product directories, design galleries, article venues and niche communities in one place, and having them in one file you can grep is more useful than having them scattered across bookmarks. The weakness is that a link list ages silently. Nothing in the README carries a date per entry, nothing marks a dead domain, and the absence of a `last verified` convention means a reader has to check each link themselves.

The last push was 2026-06-19 and the repository is not archived, with 2226 stars and 193 forks. Ten open issues is a low number for a repository of this kind and suggests the maintenance burden is real but small. The honest summary is that this is a well-organised starting point for one day of launch work, maintained by a single author on no schedule, and that its categories are more reusable than its links.

Editorial conclusion

This list is best read as a reference for where to post on the day you launch, not as a promotion strategy. Its own opening claim is that word of mouth from people who like the product is the real channel, and everything after that is a distribution list for when the product already deserves the attention. Two things are worth knowing before you rely on it: the entries are unranked and undated, so nothing tells you whether a given directory still has traffic, and the repository is a README with an Unlicense and an assets folder, meaning there is no code to break but also no mechanism keeping the links alive. Work through the tables by channel type, submit to the handful of directories that match your category, and spend the remaining effort on the product, which is the argument the README opens with and never contradicts.

Frequently asked questions

Official sources

1. Issues
2. License: Unlicense
3. README
4. trekhleb/promote-your-next-startup on GitHub