Onboarding is the path to first value
Onboarding is everything between a person deciding to try your product and that person getting something useful out of it. It includes the sign-up form, the first screen, the empty dashboard, the import of their data, the email that arrives the next morning and the invitation they send to a colleague. It is finished when they have done the thing the product exists for, and seen the result.
It is the stage where most trial users are lost, and where improvement is cheapest, because the people involved have already shown interest. A review asks one question at every step: does this help a new user reach first value sooner, or does it stand in the way?
Define first value and the steps to it
First value is a specific event, not a feeling. In an invoicing product it might be sending a first invoice. In a scheduling tool, a first booking received. In an analytics product, a first report showing the user's own data. It should be something the user came for and can recognize as progress.
- 01
Name the event
One sentence, observable in your data. If the team cannot agree on it, that disagreement is the first finding.
- 02
List the necessary steps
Everything a new user must do to get there: create an account, verify an email, connect a data source, create a first item.
- 03
Mark the steps you added
Tours, surveys, profile questions, plan selection. These are your choices, not necessities.
- 04
Distinguish roles
The person who signs up and sets things up has a different path from a colleague who is invited later. Each needs its own first value.
- 05
Write down the time it takes
Go through it yourself with a stopwatch, as a new user with nothing prepared.
Measure the path step by step
| Stage | What to measure | A warning sign |
|---|---|---|
| Sign-up started to completed | Share of visitors who finish the form | A steep drop on one field, or on phones |
| Email verification | Share who verify, and how long it takes | Many who never return from their inbox |
| First session | Share who complete a first key action before leaving | Most leave from the first screen |
| Data in | Share who import, connect or enter enough to use the product | Imports started and abandoned |
| First value | Share who reach the defined event, and time taken | Reached by a small minority, days later |
| Return | Share who come back in the following week | First value reached once and never again |
| Team | Share of accounts that invite a second user, where the product is collaborative | Single-user accounts that fade |
| Conversion | Share of trials that become paying | Conversion that tracks first value almost exactly |
Split each measure by acquisition source, device and plan. Users from one channel may fail at a different step from the rest. Then watch recordings or run sessions with new users at the worst stage to learn why.
Review the sign-up
- 01
Is every field needed now?
What to look forCompany size, phone number and role can be asked later, once the user has seen value, or not at all.
- 02
Is there single sign-on?
What to look forSigning up with an existing work account removes a password and often the verification step.
- 03
Does email verification block progress?
What to look forLetting people start and verify later keeps them in the product. Where verification must come first, make the email arrive at once and bring the user straight back to where they were.
- 04
Is a card required?
What to look forRequiring payment details before a trial reduces sign-ups and raises the commitment of those who continue. Decide deliberately, and measure both ends.
- 05
What happens with a personal email address?
What to look forIf you require a work address, say so before the form is submitted, with a reason.
- 06
Are errors clear?
What to look forPassword rules stated in advance, messages beside the field, nothing cleared on failure.
Review the first session
- The first screen. A new account is empty. An empty dashboard with zeros and blank charts tells the user nothing. Replace it with one clear first action.
- Doing over watching. Product tours that point at buttons are skipped and forgotten. Guide the user through a real first task, with their own content, and let the interface teach as they go.
- Sample data. Where real data takes effort to bring in, a realistic example lets people explore the result first. Make it obviously sample data and easy to remove.
- A short checklist. Three to five setup steps with progress shown, each leading somewhere useful. Ticking the first one automatically helps.
- Sensible defaults. Pre-filled settings and templates so that the user edits instead of creating from nothing.
- Questions that pay back. If you ask what the user wants to achieve, change what they see as a result. Otherwise do not ask.
- Help within reach. A way to ask a question or see a short guide without leaving the task.
The hardest step in many products is getting the user's own data in: an import, an integration or a first batch of entries. It deserves more design attention than any welcome screen, and it is where a person on your side offering help can make the difference.
A worked example
Reviewing onboarding for a staff rota product
This example is invented to show a review from definition to change. A product for planning staff rotas converts few of its free trials.
- First valueThe team agrees on: a published rota that at least one staff member has opened. Before the review, nobody had written this down.
- MeasuresOf every 100 sign-ups, 81 complete the form, 64 verify their email, 40 add staff, 19 publish a rota and 11 have it opened by a staff member.
- Finding oneEmail verification blocks entry. A third of those lost never return from their inbox. Verification is moved to the moment of publishing.
- Finding twoAdding staff means typing each person one at a time. A paste-from-spreadsheet option and a sample team are added, so that a rota can be drafted before the real list is ready.
- Finding threePublishing sends staff an email with a link that requires them to create a password before seeing their shifts. Staff are now shown the rota first, from a secure link, and invited to set up an account afterwards.
No visual redesign took place. Three changes to the order and effort of existing steps roughly doubled the share of trials reaching first value in the following two months, and paid conversion rose with it.
Emails, invitations and the upgrade moment
Follow-up that helps
- Emails triggered by what the user has and has not done
- One suggested next step per message, linking straight to it
- Invited colleagues land on something useful, not on a setup wizard
- Limits and prices visible before they are hit
- A person available to help during the trial
Follow-up that harms
- The same email sequence for everyone, whatever they did
- Feature announcements to people who have not started
- Invitees asked for details the owner already gave
- A paywall that appears without warning in the middle of a task
- Trial expiry that locks users out of their own data
The upgrade request works best just after the user has received value, and at a natural boundary such as adding another colleague. Asked earlier, it reads as a toll.
The onboarding review checklist
SaaS onboarding review checklist
0 of 11 checkedWhat to do next
Create a new account in your own product today with a fresh email address, on a phone, and time how long it takes to reach the result a customer signs up for. Write down every step. Then ask your team what first value is, and see whether the answers match.
Our SaaS UX design page explains how we review and redesign onboarding, and a UX audit covers it alongside the rest of the product. Tell us what your trial numbers look like.
Common questions
What is a good activation rate for a SaaS product?
There is no universal figure, because first value is defined differently in every product and trial users vary by channel and price. Compare your own rate over time and between user groups. The useful question is where people stop before reaching first value, since that shows what to change, whatever the overall number.
Should we use a product tour?
Sparingly. Tours that highlight interface elements one after another are skipped by most people and remembered by few. Guiding a user through one real task, with their own content, teaches more and delivers value at once. Short contextual hints, shown when a feature becomes relevant, work better than an introduction to everything at the start.
Should the trial require a credit card?
It is a trade. Asking for a card reduces sign-ups and filters for more serious users, who convert at a higher rate. Not asking widens the top and puts more weight on onboarding. Which earns more depends on your product, price and sales model. If you can, test both and compare paying customers, not sign-ups.
How long should onboarding take?
As short as reaching first value allows. Some products can deliver it in minutes, and others need data, integrations or colleagues first. Where the path is long, give an early partial result, such as a view with sample data, and show progress clearly. Measure the time taken by real users, not the time you estimate.
Who should own onboarding in a SaaS company?
One named person or team, with responsibility across sign-up, product, emails and trial support, and with the activation measure as their target. Onboarding fails when marketing controls the form, product controls the first screen and nobody controls the journey between them. Shared ownership usually means that no one is accountable.