The short answer
Add three UTM parameters to the link you submit to every directory: utm_source for the directory, utm_medium for the kind of listing, and utm_campaign for the batch. Log each tagged link in one sheet the day you submit it, then read visits and signups per row after 30 days and drop whatever sent nobody.
By the time you have submitted to thirty directories, you have thirty links pointing at your site and no way to tell which one brought a single person. I have watched this happen to my own launches. The submissions feel like progress, a few visits trickle in, and a month later the only honest answer to "which one worked?" is a shrug.
The fix costs about ten minutes. Tag the link before you submit it, write it down, and look at the result once. This post gives you the naming convention, a tracker with ten platforms pre-filled, and the 30-day read that tells you what to keep.
Why do untagged submissions lose their data?
An untagged link tells your analytics only what the visitor's browser reports. Some visits arrive with a referrer you can read. Many do not: a click from an app, from an email digest of a directory's weekly picks, or from a page that hides its referrer lands in your reports as direct traffic. A UTM tag travels in the link itself, so it survives all of those routes.
Even when the referrer does show up, it names a domain, not a listing. A launch platform can send you visitors from its front page, a topic page, a newsletter and a comment thread, and all of them look like one referrer. If you listed on the same site twice, for example a launch and later a directory profile, a tag is the only thing that separates them.
Google's guide to campaign URL builders describes the mechanism plainly: you add campaign parameters to the destination URLs you use in referral links, and then you can see which campaigns referred traffic. A directory submission is just a referral link you hand to someone else's website.
What UTM naming convention survives a month?
The convention matters more than the tool, because the failure is never a missing tag. It is the same directory spelled three ways. Analytics tools generally treat ProductHunt, producthunt and product-hunt as three different sources, so one listing ends up split across rows and none of them looks like it did anything. Use these four rules and write them at the top of your sheet.
- Lowercase only, no spaces. Hyphens inside a value, never underscores or capitals.
- utm_source is the platform, one word.
producthunt,saashub. Pick the spelling once and copy it from the sheet every time. - utm_medium is the kind of listing. Three values cover almost everything:
launchfor a dated launch event,directoryfor a permanent profile, andcommunityfor a post people reply to. - utm_campaign is the batch with a date.
launch-2026-11says which round of submissions this was, so a second round next year does not mix into the first.
Skip utm_content and utm_term. They exist for ad variants and paid keywords, and a one-time listing has neither. Two other limits keep your data clean. Never put UTM tags on links between pages of your own site, because the tag replaces the original source of the visit. And give each tagged page a canonical tag that points to the clean address, so the tagged and untagged versions are read as one page.
A tagged link then looks like this, with your own domain in front: ?utm_source=saashub&utm_medium=directory&utm_campaign=listings-2026-11. If you want the longer reference for what each parameter means, the UTM parameters overview on Wikipedia covers the five standard fields and their history.
How do you build the tracker sheet?
One spreadsheet, one row per listing, six columns. Create the columns below in any spreadsheet and paste in the rows from the table that follows.
Date submitted
The day the listing went live, not the day you filled in the form.
Platform
The name you will recognize in six months.
Tagged URL
The exact link you pasted into the form.
Visits (30 days)
Sessions from that source, read from your analytics.
Signups (30 days)
Whatever your one conversion is: signup, trial, waitlist.
Verdict
Keep, refresh, or drop. One word.
Here are the first three columns pre-filled for ten platforms. The tags are my suggested values, and the visits and signups are for you to fill in, so I left them out rather than invent numbers. Append the whole string to the URL you submit, after your own address.
| Platform | utm_source | utm_medium | utm_campaign |
|---|---|---|---|
| Product Hunt | producthunt | launch | launch-2026-11 |
| BetaList | betalist | launch | launch-2026-11 |
| Uneed | uneed | launch | launch-2026-11 |
| MicroLaunch | microlaunch | launch | launch-2026-11 |
| Peerlist | peerlist | launch | launch-2026-11 |
| Indie Hackers | indiehackers | community | launch-2026-11 |
| Hacker News (Show HN) | hackernews | community | launch-2026-11 |
| SaaSHub | saashub | directory | listings-2026-11 |
| AlternativeTo | alternativeto | directory | listings-2026-11 |
| G2 | g2 | directory | listings-2026-11 |
For the launch platforms in the first rows, the guides to launching on Uneed and submitting to MicroLaunch walk through the forms, and the launch directories guide covers where each one fits. Work through the sheet in this order.
01Tag the link before you open the form
Build every tagged URL in one sitting, in one document. Doing it per form is how a typo in one submission turns into a source nobody can group later.
02Paste the tagged URL into the website field
Use it in the main URL field and any "visit website" field the listing offers. If a form asks for a clean URL and rejects parameters, submit the clean one and write that row down as untracked.
03Log the row the same day
Date, platform, tagged URL. Thirty seconds now, and the whole sheet is the only record you will have of where you listed.
04Check your analytics after two days
Open the acquisition report, look for the source value you chose, and confirm one visit from your own click shows up under the right name. Fix the tag now, while the listing is new.
05Come back at day 30
Fill the visits, signups and verdict columns for every row. That single pass is the point of the exercise.
One note from my own setup. Favors.dev saves a project's URL without its query string, so a tag added to your project page URL inside a Favors.dev listing would be removed. Tag the links you submit to outside directories, and check that your tag survives wherever you paste it.
How do you read the results after 30 days?
Open your analytics, find the acquisition or traffic-source report, and filter by campaign to the value you used, such as launch-2026-11. Group by source. Copy the visit count for each source into the sheet, then do the same for signups.
Three things to watch for while you read. A source with zero visits is a result, not a mistake, as long as your own test click registered. A source with a few visits and no signups is a discovery listing that did its job as a public page but not as a door. And any row with an unusual number of visits and no signups is worth a look at the landing page it points to before you blame the directory.
Resist averaging. Launch platforms send traffic in the first days and then stop, while a permanent directory profile drips in a little at a time for years. A launch row and a directory row are different shapes, which is the reason the medium tag exists. Read the two groups separately. The launch metrics to track post covers what to measure beyond visits.
Which directories should you drop?
Drop a directory when it sent nobody in 30 days, your test click confirmed the tag works, and the listing does not serve another purpose for you. Keep a zero-traffic listing only if it is one of the trusted pages that describe your product consistently, which is a reason I laid out in whether directory backlinks help SEO. Everything else goes in the "drop" pile, which means you stop renewing it, stop updating it and stop thinking about it.
Put the saved hour into a place that did send people. If two sources account for most of your visits, a second launch, a better screenshot or a reply in the comments there is worth more than a fifteenth listing. In the Favors.dev founder marketing co-op, members trade verified favors in the points economy, and the launches page is where a launch you already tagged can get extra eyes on it.
Frequently asked questions
What UTM parameters should I use for directory submissions?
Use three: utm_source for the specific directory (producthunt, saashub), utm_medium for the kind of listing (launch, directory, community), and utm_campaign for the batch (launch-2026-11). Google's own URL builder guidance treats source, medium and campaign as the core set, and utm_content and utm_term add little for a one-time listing.
How do I track which startup directory sent traffic?
Give each directory its own utm_source value, paste the tagged link into the submission form, and read the source column in your analytics after a few weeks. Without a tag, many visits from a directory show up as generic referral traffic or as direct, and you cannot tell which listing earned them.
Do UTM parameters hurt SEO or my backlink?
Not in a way a founder needs to worry about. The link still points at your page, and a page that is reachable at a clean address and a tagged one should declare the clean address as its canonical version. Do not use UTM tags on links between pages of your own site, because that overwrites the original source of a visit.
Can I use UTM links on Product Hunt or Hacker News?
Check each form. Some platforms accept a tagged URL in the website field, some strip query strings, and some rewrite outbound links. Test with your own click after the listing is live and record any row that did not carry the tag as untracked.
How long should I wait before deciding a directory does not work?
Thirty days is a fair window for a permanent directory listing, and a week is enough for a launch platform, where traffic comes in the first days. Record the number, decide keep, refresh or drop, and move on.