Published
9 MIN READ
SaaS onboarding: what goes into the first session
SaaS onboarding is the path the product builds between creating an account and the first value action, the useful thing the tool delivers on its own. It isn't the welcome tour or the customer success team's job: it's interface scope, decided before the first screen is designed.
Anyone searching for the term in Portuguese lands in neighboring territory. The pages that rank deal with customer success processes: a dedicated analyst, eight steps, a service playbook. None of them answers the question of someone specifying a product: what needs to be on screen for the user to get there without help.
The confusion is expensive because the decision is about scope. Either onboarding goes into the project as a designed flow, with screens and a success criterion, or it doesn't and happens improvised, with the effect showing up at renewal.
What is SaaS onboarding?
It's the set of screens and decisions that takes someone from newly signed up to a user who has already gotten something out of the product. The Nielsen Norman Group defines onboarding as the process of familiarizing users with a new interface, using dedicated flows and elements that are not part of the app's normal interface, in an article by Alita Kendrick published in June 2020 and revised in January 2021.
The definition carries an uncomfortable consequence: everything that is onboarding is an extra layer. The same group recommends avoiding it whenever possible and spending the effort on making the interface easier to use. A well-solved first session is the one that needs the least explanation, and it opens the four moments that decide the product experience.
Product onboarding is not customer onboarding
They're two disciplines with the same name in Portuguese. Customer onboarding is human and commercial: implementation, data migration, training, account follow-up. Product onboarding is the interface doing that work on its own, for people who will never talk to anyone at your company. Brazilian material covers the first well and ignores the second.
A marketing platform guide from November 2024 organizes the subject into eight steps and describes the role of the onboarding analyst. A product school, in February 2025, talks about stages and types. A software vendor, in June 2026, applies AI to the customer success team. None of them says what needs to exist on screen.
The boundary changes what you hire. Before adding more people to support customers, there's an earlier question: should this person need someone at all? A low-ticket product can't sustain human implementation, and the website that promises what the product needs to deliver has already done its part.
Why isn't a completed checklist activation?
Because completing tasks and getting value are different things. A metrics framework published in May 2026 by a digital consultancy brings together two numbers it attributes to industry benchmarks: average activation in B2B SaaS sits at 37.5%, in a 2024 survey of 62 companies, and more than 98% of users who don't reach any value milestone drop off within two weeks, according to Amplitude's 2025 product benchmark report.
The same material records a subscription app with over 90% completion of its onboarding flow and most users gone by the second day. The dashboard showed onboarding working. Retention showed the opposite.
A guided tour doesn't fix it either. In an article from February 2023, revised in 2024, Page Laubheimer notes that tutorials interrupt users, don't necessarily improve task performance and are quickly forgotten. In the shift from website to digital product, the metric stops being visits and becomes what the person does after getting in.
How to define the activation event before designing the screen
An activation event is the observable action that separates those who stay from those who disappear. The idea comes from the AARRR model, published by Dave McClure in 2007, which already defined activation as the first behavior capable of predicting long-term value, and not as a completed signup or a filled-in profile.
The May 2026 framework proposes three tests, attributed to a subscription infrastructure company. The event only works if it passes all three.
The retention curve diverges. Those who took the action retain better, and the gap holds. A difference that disappears within a week doesn't count.
The relationship holds everywhere. Across all segments, channels and platforms. An event that only predicts retention in one channel is an artifact of that channel.
Pushing people toward the action moves retention. It's the causal test, run as an experiment, and it separates vanity numbers from a metric you can make decisions with.
Public examples are useful for seeing the method, never for copying the number, like the messages-per-workspace threshold at Slack, disclosed around 2016. Defining the event is a product decision, and it comes before the design and before how the product is built behind the screen.
How much time is there until the first value action?
Less than most teams imagine. A benchmark report published in April 2026 and updated in July compiles 547 companies and arrives at a median of 1 day, 1 hour and 54 minutes between signup and first value. The average is 1 day, 12 hours and 23 minutes, and the source recommends reading the median, since the average is inflated by a long tail.
Time to first value action | Typical activation rate |
|---|---|
Under 2 minutes | 40% to 60% |
2 to 5 minutes | 25% to 40% |
5 to 15 minutes | 15% to 25% |
Over 15 minutes | Under 15% |
The table comes from the same report. There's also the first-week benchmark: the May 2026 framework cites Amplitude's 7% rule, under which a product reaches the top quartile when at least 7% of a cohort comes back on day seven. B2B products with heavy integration play on a scale of months, and there the mistake isn't taking time, it's taking time without delivering anything.
What goes into the first session and what stays out?
What leads to the first value action goes in. Everything else stays out. The 2026 benchmark report is specific about both sides.
Signup at the smallest possible size. Each extra field drops form completion by 10% to 15%: with email only, it stays between 80% and 90%; with six fields, it falls below 30%.
An environment populated with sample data, so the person sees the tool working before configuring anything.
A checklist with 3 to 5 critical items. Above 8, completion plummets.
A path segmented by the signal the signup already provides: company size, role, integration intent, stated use case.
What leaves the first session and comes back after the first result: advanced integrations, inviting teammates, profile customization and the tour of the rest of the product.
The 2026 version of this logic is the product setting itself up: the user writes what they want to do, and the product pulls in integrations, builds the workspace and delivers a draft. When the internal team doesn't have the bandwidth for this without stopping the roadmap, the way to go is to reinforce the design team that already exists.
What to ask for in the scope when hiring
Onboarding doesn't show up in a proposal on its own. It needs to be listed, or it becomes a leftover from implementation. In the SaaS and custom software projects that come through Wama, like Wiz Benefícios, Pepper, Visor and Milhas Pix, the question comes before the design, in the longest phase of the project: what is the action that proves the tool works, and how many screens separate signup from it.
The activation event defined, with how it's measured and who instruments it.
The first-session screens designed, with the empty states of the first access.
What was deliberately postponed until after the first result.
The success criterion that will say whether the flow worked.
It isn't brand rework or a full dashboard redesign. It's the shortest front of a product project and the one whose effect you can observe fastest, where it's easy to show how design pays for itself in business results.
If the product doesn't exist yet, the conversation starts earlier: the investment in the first version of the product and the website the company puts live before the product.
Common SaaS onboarding mistakes
Calling the end of the tour activation. A tour measures attention, not value delivered.
Asking for setup before delivering something. Someone who hasn't seen the tool work won't invest effort.
A single path for every profile. The signal to segment already came in at signup.
Treating the first access as a welcome. It's a work screen, and the work is the first result.
Measuring completion instead of retention. The pretty number and the useful one are rarely the same.
Frequently asked questions about SaaS onboarding
What is a good activation rate for a SaaS?
Available surveys put the B2B SaaS average at around 37%, so being above that is already reasonable. The number matters less than the definition behind it: a high rate for an event that doesn't predict retention decides nothing.
How long should it take a user to reach the first value action?
It depends on the product. In tools for individual use, 2026 benchmark data shows much higher activation when the first value arrives in under five minutes. In B2B products with integration, the scale is weeks, and what counts is delivering partial results along the way.
Does a guided tour work in SaaS onboarding?
Not much. Usability research shows that tutorials interrupt users, don't reliably improve task performance and are quickly forgotten. What works better is help that appears in context, inside the screen where the person is working, at the moment they need it.
Can you improve onboarding for a product that's already live?
Yes, and it's the most common scenario. The safe path is to define the activation event, measure where the current cohort gives up and tackle the stretch with the biggest drop. Redesigning the first session without that diagnosis swaps one unknown flow for another.
Who is responsible for onboarding within the company?
The design belongs to product and design, the instrumentation to engineering, and the follow-up to whoever owns retention. The common mistake is treating the subject as a single team's task, which produces a pretty flow without measurement or measurement with no one to act on it.
If you have a digital product live and want to discuss the first session with a studio that works with SaaS every day, talk to Wama about your project.
