
X (Twitter) Attribution for Crypto Apps When Links Fail

Key Takeaways
Crypto teams report that X posts with external links get less reach, so many post without links. The visits still come, but they often arrive with no UTM and no referrer.
Formo classifies t.co and x.com referrers as Organic Social, and visits with an X ads click ID (twclid) as Paid Social, even when the X in-app browser strips the referrer.
When a post has no link, use referral codes, the X in-app browser signal, and behavior-based attribution: what new users did within hours of first being seen.
Judge X by funded wallets and onchain actions, not clicks. X Ads click counts and analytics sessions measure different things and will not match.
X is where many crypto users first hear about a new app, but it is also the channel that analytics tools see least clearly. To attribute X traffic, combine three signals: links where you can use them, referral codes where you cannot, and the behavior of new wallets in the hours after a post. Then judge X by the wallets and onchain actions it brings in, not by clicks.
This guide explains why X traffic goes missing, how Formo classifies the X traffic it can see, and six practical ways to recover attribution when posts have no link.
Why X traffic is hard to attribute
Growth teams at DeFi and trading apps describe the same four problems:
Teams avoid links. Crypto teams report that X posts with external links get less reach, so many post without them, or add one link only in a reply. Affiliates and KOLs often do the same.
The X app strips referrers. When a user taps a link inside the X mobile app, the page often opens without a referrer. With no UTM either, the visit is recorded as Direct.
Threads and replies drive search, not clicks. A user reads a thread, then searches for the app or types its name later. That visit looks like Organic Search or Direct.
Click counts do not match sessions. X Ads counts clicks. Analytics counts sessions that loaded and ran the tracking script, so the two numbers always differ.
None of this means X does not work. It means the default signals undercount it, so you need a few deliberate ones.
How Formo classifies X traffic
Formo assigns every session a channel from the signals it carries. These are the rules that matter for X:
What the visit carries | Channel in Formo |
|---|---|
Referrer from t.co or x.com | Organic Social |
X ads click ID (twclid), even with no referrer | Paid Social |
A paid UTM medium with an X source or referrer | Paid Social |
A referral code (ref, referral, refcode, af, or referrer), or utm_medium=referral | Referrals |
A social UTM medium (for example utm_medium=social) | Organic Social |
No referrer, no UTM, and no code | Direct |
Formo also stores the first and last referrer, UTM values, and referral code on each wallet profile, and it records the browser of each session. Visits opened inside the X in-app browser are detected from the user agent, which gives you one more signal when the referrer is gone. See attribution key metrics in the docs for the full list.
Six ways to attribute X traffic without relying on links
1. Use one link, in the right place
If you post a link at all, put it where it costs the least reach, such as the first reply or your profile, and tag it with UTM parameters. A consistent naming scheme (for example utm_source=x, utm_medium=social, utm_campaign=launch-week) keeps the reports clean. You can build tagged links with the UTM generator.
2. Give every KOL and affiliate a referral code
A referral code survives when a link does not. People can type it, paste it, or reach it through a short path such as /r/alice. The Formo Web SDK detects the common referral parameters automatically and can also extract a code from a URL path that you define. Each code then appears as its own row in the Referrals channel, so you can compare KOLs by the wallets and deposits they bring, not by impressions. For program design and payouts, see DeFi referral program analytics.
3. Keep the X ads click ID on paid campaigns
For paid X campaigns, keep X's click ID enabled on your landing URLs. Formo reads twclid and classifies the session as Paid Social even when the in-app browser removes the referrer, so paid X traffic does not fall into Direct.
4. Use the X in-app browser as a signal
Because Formo records the browser of each session, you can filter for sessions opened inside the X in-app browser. It is not a full attribution source, since some users copy the link into another browser, but it shows how much of your Direct traffic actually started on X.
5. Attribute by behavior after a post
When a post has no link at all, look at what new users did right after it went out. In Formo, the Behavior filter has a Within first seen option: for example, users who connected a wallet or made a first deposit within 24 hours of first being seen. Compare that cohort across the days around your X posts, spaces, and KOL threads.
This is correlation, not proof. It works best when you post on a schedule and compare against a baseline period with no X activity.
6. Ask new users where they heard about you
Add one optional question at onboarding or in your waitlist form: where did you hear about us? Self-reported answers catch the X threads, group chats, and podcasts that no tracker can see. Store the answer as a user property so you can segment by it next to the tracked channels.
Measure X by onchain outcomes, not clicks
Once the signals are in place, build one funnel and break it down by channel:
Visit the landing page or app.
Connect a wallet.
Complete the first meaningful onchain action, such as a deposit, swap, or trade.
Return and act again in a later week.
Compare Organic Social, Paid Social, and Referrals on funded wallets, first-action conversion, and retention, not on visits. X often brings fewer visits than other channels but more committed wallets, and the reverse also happens. Check both first-touch and last-touch views, because X is often the first touch and rarely the last one. For the full method, read Onchain Attribution: The Complete Guide for DeFi Teams or see Formo attribution.
Why X Ads numbers and analytics numbers differ
X Ads reports | Your analytics reports | Why they differ |
|---|---|---|
Link clicks | Sessions | A click does not always become a loaded page with a running script |
Repeat clicks by one user | One session per visitor and day | Analytics groups repeat visits |
Clicks from automated traffic | Sessions from real browsers | Some automated clicks never load the page |
Conversions defined in X | Wallet connects and onchain events | The two tools count different conversion events |
Use X Ads numbers to manage spend inside X, and use your analytics to decide which campaigns produced funded, retained wallets.
Setup checklist
Install the Formo SDK on your site and app, and confirm wallet connects are captured.
Agree on one UTM naming scheme for any X links.
Create a referral code for each KOL, affiliate, and partner, and set a custom path pattern if you use short paths.
Keep the X ads click ID on paid campaign URLs.
Build a funnel from visit to first onchain action, broken down by channel and referral code.
Save a Within first seen segment and check it after each campaign.
Frequently Asked Questions
Why does my X traffic show up as Direct?
Two common causes. Many posts on X have no link, so visitors type your URL or search for you. And the X in-app browser often opens links without a referrer. With no referrer, no UTM, and no referral code, any analytics tool must record the visit as Direct.
Can Formo attribute X ads when the referrer is missing?
Yes, when the ad click carries X's click ID. Formo classifies a visit with a twclid parameter as Paid Social even if the referrer is empty, then links that session to the wallet that connects and to its onchain events.
How do I track KOL posts on X that have no link?
Give each KOL a referral code that people can type or that sits in a short URL path, compare new users by code, and use the Within first seen behavior filter to see what users did in the hours after each post went out.
Should I use first-touch or last-touch attribution for X?
Use both. X is often the first place a user hears about a crypto app, so first-touch shows its role in discovery. Last-touch shows which channel closed the conversion. Formo stores the first and last referrer, UTM values, and referral code on each profile, so you can compare the two views.
Why do X Ads clicks not match my analytics sessions?
X Ads counts clicks on the ad. Analytics counts sessions where the page loaded and the tracking script ran. Quick exits before the page loads, blocked scripts, repeat clicks, and bot clicks all widen the gap, so compare trends and onchain outcomes, not raw click totals.


