One: who owns the address?
If your page lives at someone-else.com/yourname, the recognisable part of that address belongs to the platform. Every mention of it, every backlink, every printed poster builds their name alongside yours.
The alternative is a domain you choose, whether that is one you register yourself or a ready made one the tool offers. Either way the address reads as yours, and if you ever move tools the address can move with you. This is the single decision that is hardest to reverse later.
Two: can you change where a link points after sharing it?
This sounds like a small feature and it is the reason short links exist at all. A link you can redirect is an asset. A link you cannot is a one time announcement.
It matters most for anything printed or embedded somewhere you cannot edit: a QR code on packaging, an address in a published article, a link in a video description on a channel you no longer control. Check that the tool lets you edit the destination without changing the link itself.
Three: do the statistics answer a question?
A total click count is not analysis. It tells you that something happened without telling you what to do differently.
Useful statistics break the number down by referring site, country, device and by individual link. That is enough to answer the questions people actually have: which platform sends people who act, whether the audience is on mobile, and which of your links nobody has ever tapped.
Also check how long the data is kept. A tool that shows the last seven days cannot show you whether last quarter was better than this one.
Four: what else does the account do?
A bio page is rarely the only link you need. There is usually a short link for a campaign, a QR code for something printed, and a file or contact card you keep sending by hand.
If those live in three different accounts, they get three different addresses and three sets of statistics that never line up. Look for one account that covers the page, the short links, the QR codes and the measurement, so a campaign can be read as one thing.
Five: is the free tier actually usable?
Free tiers exist on a spectrum. Some are genuinely enough for a working page. Others cap you at a handful of links, strip out statistics or put someone else's branding on your page, so the free tier is really a demonstration.
Read the limits before you build. The question is not whether there is a free plan but whether the free plan would still work for you in six months.
Six: what does your visitor actually meet?
Your visitor came from a profile, tapped once and expects a link. Anything between them and that link is friction you are paying for.
Check whether the page loads a cookie banner, an interstitial advert or a sign up prompt. Statistics can be collected from the click itself without asking anyone to consent to tracking, and tools that do it that way give you your numbers without spending your visitor's patience.
The guide on measuring clicks without a cookie banner goes into what you can and cannot learn that way.