TapCub is live — analytics, insights and live chat, free on one platform
Growth & Marketing

Sign up flow: measure each step before you cut a field

A sign up flow runs from the sign up page to a working account. The trade-off behind each element, a six-event tracking plan, and a worked funnel showing why the step that leaks most is often not the form.

Sophia Bennett Published Nov 15, 2026 14 min read Growth & Marketing
Sample data: lifting verification from 65% to 80% adds 234 active users; deleting the company size field adds 42.Sample data

A sign up flow is every step between clicking "Sign up" and holding a working account: the page, the form, verification and the handoff into the product. Each required field or check adds a drop-off point, but measure before you cut: record one event per step and fix the step that loses the most people first.

What you will learn

  • Track one event per step, from the sign up page to the first key action
  • In the sample, fixing verification is worth 234 active users; deleting a field, 42
  • Field error events show which rule stops people; never log what they type

What is a sign up flow?

A sign up flow is the sequence of screens between a visitor clicking "Sign up" and that person holding a working account: the sign up page, the form, any verification step, and the handoff into the product. It ends where onboarding begins. What happens after the account exists, such as checklists, tours and first-use guidance, is covered in app onboarding.

The sign up page is one screen; the flow is everything that screen sets in motion. That difference matters because most teams redesign the page and never look at the steps after it. A form can convert well and still feed a verification step that quietly loses a third of the people who completed it.

The argument of this article: every required field, rule and confirmation adds a place to drop out, but you should measure before you delete anything. Record one event per step, find the step that loses the most people, and fix that one first. Guessing usually picks the wrong step.

The elements of a sign up flow, and what each one costs

Every element in a sign up flow buys you something: data, security, fewer fake accounts, a better first session. Each one also costs some share of the people who started. The table lists the common elements with that trade-off and a sensible default.

Two principles sit behind the defaults:

  • Ask for what you need now, not what you might want later. Company size, job title and "how did you hear about us" can be asked inside the product once the person has seen some value. Nielsen Norman Group's form usability guidelines start with "keep it short": cut fields that can be derived, collected later or left out.
  • Never make a security step harder than the risk requires. NIST's digital identity guidelines (SP 800-63B-4) tell services not to impose composition rules such as mixing character types, not to force periodic password changes, and to allow password managers and autofill. A password rule that rejects "correct horse battery staple" because it lacks a digit adds friction without adding security.
ElementWhat it costsWhat it buysSensible default
Extra fields (company, role)Each one is another reason to stopSegmentation dataAsk inside the product later
Password rulesErrors and retriesAccount securityLength rule plus blocklist, no composition rules
Email verificationAn inbox between the user and the productReal addresses, less spamLet people in first, confirm within days
Social login or SSOSome users refuse to link accountsNo password, no verification stepOffer it first if it activates better
Bot checkExtra step for real peopleFewer fake accountsOnly when abuse is measurable
Credit card up frontMany visitors leaveFewer, more committed trialsOnly for paid-only products
Onboarding questionsDelay before first valuePersonalised first sessionOne or two, skippable

Defaults are starting points; your funnel decides which element to change first.

Instrument the flow before you change it

You cannot find the leaky step without one event per step. A minimal plan for a web sign up flow:

EventFires whenProperties worth keeping
signup_page_viewThe sign up page loads (a pageview is enough)Source, landing page
signup_startedThe first field receives inputMethod shown (email, Google, SSO)
signup_field_errorValidation rejects a fieldField name, rule that failed
signup_submittedThe account is createdMethod used
email_verifiedThe verification link or code is acceptedMinutes since submit
first_key_actionThe person does the thing your product is forWhich action

Rules that keep the data trustworthy:

  1. Never send the values people type. The field name and the rule that failed ("password", "too short") are enough to find the problem. Email addresses and passwords do not belong in analytics events.
  2. Send signup_submitted from the server when the account is actually created, so a double click or a failed request does not count as a sign up.
  3. Decide what "first key action" means before launch. It is the step that turns a sign up into a user, sometimes called the aha moment. Without it you will optimise for accounts that never get used.
  4. Write the plan down with owners and required properties, so the events survive the next redesign. A tracking plan is the usual place.

A worked funnel: 5,000 visits to the sign up page

Sample data: one month for a SaaS product with email and password sign up and email verification required before first use.

StepPeopleRate from previous stepLost at this step
Visited sign up page5,000——
Started the form3,10062.0%1,900
Submitted2,23272.0%868
Verified email1,45165.0%781
First key action1,01670.0%435

Overall, 1,016 ÷ 5,000 = 20.3% of visitors become active users.

Suppose the team's first idea was to delete the optional-looking "company size" field, and that this lifts the submit rate from 72% to 75%. Then 3,100 × 75% = 2,325 submit, 2,325 × 65% ≈ 1,511 verify and 1,511 × 70% ≈ 1,058 reach the key action: 42 more active users.

Now take the step the funnel points to. Verification loses 781 people who had already handed over an email address. If letting people into the product before verifying (and asking them to confirm within a few days) raises that step from 65% to 80%, then 2,232 × 80% ≈ 1,786 verify and 1,786 × 70% ≈ 1,250 reach the key action: 234 more active users, more than five times the field deletion.

The 1,900 people who never start the form look like the biggest loss, and they matter. But many of them are reading the page to decide whether to sign up at all; that is a question for the page's message and the pricing, not the form. The people lost after submitting are different. They wanted an account and the flow stopped them.

app.tapcub.com

Sign up page → first key action

One month

Page

5,000

Started

3,100

Submitted

2,232

Verified

1,451

Key action

1,016

Verification loses 781 people who had already submitted an email; deleting one field would recover far fewer.
Sample data
Sample data: lifting verification from 65% to 80% adds 234 active users; deleting the company size field adds 42.

Find the field that loses people

The funnel shows which step leaks. Field error events show why. Sample data from the same month:

  • 744 of the 3,100 people who started the form (24.0%) hit at least one password error. Of those, 387 submitted (52.0%).
  • The other 2,356 never saw a password error, and 1,845 of them submitted (78.3%).

If the 744 had submitted at the same 78.3%, about 583 would have finished instead of 387, so 196 more submissions. Carried through verification (65%) and the key action (70%), that is about 89 more active users. A password rule is costing twice as much as the company size field.

What to do with a finding like this:

  1. Show the rule before people type, not after they fail it.
  2. Replace composition rules with a length rule and a blocklist check, in line with NIST SP 800-63B-4, which also asks services to tell users why a password was rejected.
  3. Allow paste and password managers. NIST requires verifiers to allow password managers and autofill. WCAG 2.2 success criterion 3.3.8 (Accessible Authentication, Minimum) says a step that relies on remembering or transcribing a password needs a mechanism such as paste or a password manager to be accessible.
  4. Keep the error message specific. "Use at least 15 characters" beats "Invalid password". Nielsen Norman Group lists highly visible, specific error messages among its top form recommendations.

Email verification: before or after access?

Verification protects you from typos, fake accounts and spam. It also puts an inbox between the person and your product, which is where the sample funnel loses 35% of its submissions. Three designs, from most to least friction:

DesignWhat happensUse it when
Verify before accessThe account is locked until the link is clickedRegulated products, or when abuse is expensive
Access first, verify within N daysThe person uses the product now; a banner asks for confirmationMost SaaS and consumer products
Code in the same screenA six-digit code is typed in without leaving the flowMobile apps and short flows

Whichever you choose, track the minutes between signup_submitted and email_verified. If most verifications happen within five minutes and almost none after an hour, people who do not verify quickly are lost, and a reminder after a day will not bring many back. Check the resend rate too; a high one often means the email lands in spam.

Social login and single sign-on

"Continue with Google" removes the password field and the verification step at once, which is why it often converts better. It also has costs: some people do not want to link accounts, work products need SSO for company domains, and you get less control over the email address you receive.

Measure instead of assuming. Add the method as a property on signup_submitted, then compare the share of each method that reaches the first key action. Rules for reading it:

  • If social sign up converts better to the key action, make it the first button and keep email as the alternative.
  • If social sign ups submit more often but activate less, you may be getting curious clicks; check whether those accounts come back.
  • If most of your traffic is from work email domains, SSO matters more than consumer social buttons.

Test a redesign without fooling yourself

Sign up rates move with campaigns, seasons and launches. To know whether a change worked:

  1. Compare the step you changed. A new password rule should move the started-to-submitted rate. If the total moved and that step did not, something else did it.
  2. Compare equal periods with similar traffic. Four weeks against four weeks, with no big launch in one of them, and mark the change date on your charts.
  3. Count enough people. With about 2,000 submissions a month, a two-point change in a step rate is a few dozen people. A randomised A/B test answers faster and more reliably; run one in your testing tool if traffic allows.
  4. Follow the cohort past sign up. A change that lifts sign ups but lowers the share who reach the first key action has not helped. The free trial conversion rate and onboarding metrics tell you whether the new users stay.

Common sign up flow mistakes

  • Optimising the page and ignoring the steps after it. In the sample, verification lost almost as many people as the form.
  • Deleting fields by gut feeling. Field error events show which field actually stops people.
  • Counting accounts instead of active users. A sign up that never reaches the first key action is not a win.
  • Logging what people type. Field names and rule names are enough, and safer.
  • Changing several steps at once. You will not know which change did what.
  • Ignoring the first session. The first-time user experience decides whether the new account is used again; general sources of friction are covered in user friction.

How to measure your sign up flow in TapCub

The 5,000-visit example above, step by step in TapCub.

  1. Register the events first. In All-platform analytics › Data › Tracking plan, add the six events with their properties, mark the required ones and set the status to Developing. After release, turn on Live verification (24 h) to see how many times each event arrived and how complete its required properties are, then move the status to Accepting and Live. Guide: create a tracking plan.
  2. Send the steps. On the website, call tapcub.track('signup_started', { method: 'email' }) and the other events from your code. If you only need to know that the form was submitted, data-forms="true" on the script records form submissions with the form's id, name and action, never the values typed (track form submissions).
  3. Build the funnel. In All-platform analytics › Behavior › Behavior analysis, add a funnel from the sign up page view to first_key_action. Each step shows its count, the rate versus the previous step, the drop-off and the median time from the previous step, and the largest drop, verification in the sample, is marked. Break it down by the method property to compare email and Google, and click a step to see who dropped there.
  4. Know who signed up. Once the account exists, call tapcub.login with your user ID (the site needs identified mode). The person then appears in All-platform analytics › Users › People with a day-by-day timeline of what they did before and after signing up (identify users).
  5. Watch a handful of drop-offs. Session replay in All-platform analytics › Behavior › Session replay is off by default. When turned on, input values are masked by default and password fields are never recorded, so you can watch how people move through the form without seeing what they typed.

TapCub measures the flow. It does not send verification emails or run A/B tests; those stay in your authentication system and testing tool.

FAQ

What is a sign up flow?

A sign up flow is the sequence of steps from clicking "Sign up" to holding a working account: the sign up page, the form, any verification and the handoff into the product. It ends where onboarding starts. Measuring it means recording one event per step so you can see where people drop out.

How many fields should a sign up form have?

As few as you need to create the account, often just an email and a password, or one social button. Everything else can be asked later inside the product. Before deleting fields, check field error events: one strict rule can cost more sign ups than several optional fields combined.

Should you require email verification before people can use the product?

Only when abuse or regulation makes it necessary. For most products, letting people in first and asking them to confirm within a few days keeps more of them. Track the time between submitting and verifying: if almost nobody verifies after the first hour, a locked account loses those people for good.

Does social login improve sign up conversion?

It often does, because it removes the password and verification steps. It is not guaranteed. Record the sign up method as a property, then compare how many of each method reach the first key action. If social sign ups activate less, the extra accounts may not be worth much.

How do you measure sign up flow conversion?

Track one event per step, from the sign up page view to the first key action, and build a funnel with them. Read each step's rate, drop-off and median time. Overall conversion is active users ÷ sign up page visitors; in the sample month it was 1,016 ÷ 5,000 = 20.3%.

Sources

Was this article helpful?

We use your answers to fix and expand this page.

Sophia Bennett

Product Adoption Consultant

Explores customer onboarding, live conversations, and self-service, turning common usage questions into practical guidance and improvements.

See which sign up step loses people

Tracking plan with live verification, step funnels with median time, and user profiles from the moment people sign up.

See your data clearly. Find your growth.

Every click, backed by data. Install one line of code and see your first numbers in a minute.

No credit card · Free plans for analytics and chat · Cookieless analytics