Gozi.to vs the competition: an honest comparison

I run Gozi.to. Here is where it wins, where it loses, and which of its competitors you should use instead.

I run Gozi.to. I built it on Cloudflare's free plan, and the design goal was never “beat Bitly” — it was “never pay for hosting, and never break a link that already exists.” That one constraint decides almost everything about the product, good and bad.

If you are choosing a shortener, you are really choosing which trade-off you can live with. So here is the trade-off Gozi.to made, an honest comparison with the services people actually use, and a straight answer about when you should pick one of them instead.

The one-paragraph version

Gozi.to is the right choice if you want free click analytics with no account, no cookies on your visitors, and an API you do not have to pay for — and you can accept that anonymous links expire after 90 days, that there are no custom domains, and that one person maintains it with no SLA. If you need a branded domain, a team, or an uptime guarantee with a contract behind it, stop reading and use Short.io or Dub.co.

What Gozi.to actually does

Paste a URL, get gozi.to/aB3dK9x and a QR code. The specifics, because the rest of this post depends on them:

How it compares

Gozi.toBitlyTinyURLShort.ioDub.coRebrandlyis.gdYOURLS / Shlink
Account requiredNoYesNoYesYesYesNoYou run it
Custom code, freeSigned inPaidYesYesYesLimitedYesYes
Custom domainNoPaidPaidYesYesYesNoYes
Click analytics, freeYesLimitedBasicYesYesLimitedNoSelf-hosted
API on the free tierYesLimitedYesYesYesLimitedYesYes
Edit destination laterSigned-in linksPaidNoYesYesYesNoYes
Links expireAnonymous: 90 daysNoNoNoNoNoNoYou decide
301 permanent redirectNo, always 302Configurable301ConfigurableConfigurableConfigurable301Configurable
Cookies on redirectNoneYesYesVariesVariesVariesNoneNone
Export your dataCSV, 10k rowsPaid tiersNoYesYesYesNoFull database
Who is accountableOne person, no SLAA companyA companyA companyA companyA companyA small teamYou
Open sourceNoNoNoNoYesNoNoYes

Competitor columns describe each service's free tier as advertised on 22 September 2026. Free tiers change constantly — check before you commit to one.

Where Gozi.to genuinely wins

The free tier is the whole product, not a trial

There is no paid plan to graduate to, so there is no feature being held back to create one. The API, custom codes, QR codes and analytics are all free. Most competitors' free tiers exist to generate the friction that sells the paid tier; Gozi.to's exists because the running cost is genuinely close to zero.

It is the most private mainstream option

No visitor IP address is stored on a click — not hashed, not truncated, simply not recorded. No cookie is set on a redirect. The referrer is reduced to a domain before it is written down. The creator's IP is kept only as a keyed hash for rate limiting, and deleted after 30 days. If you are sharing links with people who would rather not be profiled, that is a real difference and not a privacy-policy flourish.

Anonymous links are genuinely anonymous, and still have stats

The management secret lives in the URL fragment, so it never travels to the server. Nobody at Gozi.to can look up who owns an anonymous link, because nobody stored it. is.gd and TinyURL give you anonymity without statistics; Bitly gives you statistics in exchange for an account. Gozi.to is the combination.

Moderation is aggressive, which protects you

Rescanning every 15 minutes for the first three days catches the classic trick of shortening a clean URL and swapping the destination once it has been shared. A shortener domain that gets blacklisted by mail filters is worthless to everyone using it, so strictness here is a feature for legitimate users, not a tax on them.

Where Gozi.to loses

No custom domains

Every link is gozi.to/…. For a brand, go.yourcompany.com measurably improves click-through, and — more importantly — you own it, so your links survive the shortener disappearing. Gozi.to cannot do this, and because the domain is a single configuration value baked through the service, it is not a small change. This alone disqualifies it for most marketing teams.

No teams

There are exactly two roles: user and admin. No shared workspaces, no per-seat access, no audit log of who changed what. Two people managing the same links means sharing one login.

One maintainer, no SLA, no support contract

The only support channel is an email address. If I am on a plane, nothing happens until I land. There is no status page, no on-call rotation, and no contractual remedy if your campaign link breaks. Competitors charge money partly because money buys somebody being awake.

The free plan has a hard ceiling, and it is shared

Cloudflare's free plan allows 100,000 Worker requests a day across my entire Cloudflare account, not just this project. Gozi.to degrades gracefully up to a point: at 90% it pauses anonymous link creation until midnight UTC while signed-in users carry on. But at 100% the Worker stops and every gozi.to link fails until 00:00 UTC. That has not happened, and the fix is a few dollars a month, but I am not going to pretend the ceiling is not there. Storage behaves the same way: under pressure, anonymous links start expiring after 30 days instead of 90.

Anonymous links expire, cannot be edited, and cannot be recovered

Ninety days, then gone. You cannot change an anonymous link's destination — the management token only lets you read statistics or delete it. And if you lose the gozi.to/m#m_… URL it is gone permanently, because only a hash of it is stored and there is no email on file to send a reset to. Signing in fixes all three, which is exactly the trade I designed, but it is still a limitation.

Always 302, never 301

Browsers cache permanent redirects, which means repeat clicks would never reach the service and would never be counted. Gozi.to chose analytics over cacheability. If you are consolidating link equity for SEO and want a genuinely permanent redirect, use something configurable.

The analytics are estimates

They come from Cloudflare Workers Analytics Engine, which samples under load and scales the result back up. At volume that is close; at low volume it is rough. There is no per-click event log, no unique-visitor count, no UTM builder and no QR-scan attribution. Also missing: bulk create, password-protected links, link-in-bio pages, and CSV export stops at 10,000 rows.

Some legitimate slugs get refused

Reserved words, and anything containing strings like login, verify, wallet or paypal, are rejected, because that is what phishing links look like. Occasionally that blocks something innocent. Other shorteners are also blocked as destinations, so you cannot chain one into another.

So which should you use?

What I did not measure

Saying this plainly is what makes the rest of the post worth anything. I have not benchmarked redirect latency across providers. I have no uptime figures for anyone, including Gozi.to — there is no monitoring behind that claim because there is no monitoring. And I have not audited what any competitor does with click data beyond what their own privacy policies say. The table reflects what the published pricing pages said on 22 September 2026, and free tiers move constantly. If something here is wrong or out of date, email hello@gozi.to and I will correct the post and say that I did.

Gozi.to is free, and you don't need an account to try it.

Shorten a link