… onboarding spans more than one sitting, and several of the setup and learning tasks that bring the actor to competent use can be taken in any order. Charted progress gives that onboarding a bounded path the actor can return to across sessions.
Problem
When reaching competent use requires several setup and learning tasks across multiple sessions, an actor given no view of the whole cannot tell what remains or which task matters next. A checklist made of trivial interface clicks rather than real outcomes makes this worse: the actor ticks every box and still has nothing working at the end.
A getting-started checklist works when an artifact cannot deliver its full value in a single first action—for example, when the actor has to connect a data source, configure a workspace, and run a first workflow before the tool becomes useful in daily work. Laying those milestones out in one visible list turns an ambiguous setup period into a bounded set of tasks the actor can finish at their own pace.
What decides whether the list helps or gets dismissed is how the steps are chosen and framed. Steps named after interface mechanics—such as opening a settings page or viewing a tour—teach the actor to click through for completion’s sake without making them any more capable. Steps named after real outcomes in the actor’s work—such as publishing a first page or answering a first request—leave the actor with a working setup when the list ends.
Starting a checklist at zero after the actor has already completed registration or initial setup is discouraging; crediting the steps they have already finished gives the list momentum from the start. Conversely, every extra item on the list is another point where the actor can abandon it, and a persistent checklist that stays on screen after the actor knows the tool becomes permanent visual clutter.
Solution
Present a multi-step first run as a short checklist of three to seven steps, naming each step by the outcome it produces for the actor rather than the control it exercises. Show what is completed, what to do next, and how much remains; credit any step the actor has already finished before the list appears; and remove or fold away the list once every step is done.
Do not start the progress counter at zero after the actor has already answered setup questions, and do not interleave finished and unfinished items in a way that hides what is left. Always let the actor dismiss the checklist early if they already know the tool or do not need the remaining milestones.
Continuation
Report each completed milestone through status feedback at the moment it finishes, and update the list so the actor always sees where they stand. Where a single milestone involves an ordered sequence of inputs of its own, let step by step carry the actor through that item before returning them to the list. Where the right next move depends on the actor's specific data or behaviour rather than a shared set of milestones, hand off to next-best action instead of a fixed checklist…
All connections
- Enables
- Onboarding — Onboarding spans more than one sitting, and several of the setup and learning tasks that bring the actor to competent use can be taken in any order.
- Enacts
- Learnability — lays out what competent use involves as a set of outcomes the actor can reach one at a time
- Complements
- Step by step — Where a single milestone involves an ordered sequence of inputs of its own, let step by step carry the actor through that item before returning them to the list.
- Alternatives
- Next-best action — Where the right next move depends on the actor's specific data or behaviour rather than a shared set of milestones, hand off to next-best action instead of a fixed checklist
- Enabled by
- Status feedback — Report each completed milestone through status feedback at the moment it finishes, and update the list so the actor always sees where they stand.
- Appears in
- Learning: Shaping the first encounter, step 5
In practice
Framing and marking milestones
Tie completion to outcomes that matter to the actor’s work rather than administrative events that matter only to the system. When a milestone finishes, confirm it in place and highlight the next uncompleted item. Reserve celebratory feedback for substantial milestones; praising every minor click quickly feels patronising.
Placement and lifecycle
Place the checklist where the actor can return to it between tasks—on the home surface, in a collapsible sidebar card, or behind a compact progress badge in the chrome. Keep it visually distinct from ongoing task lists so sample milestones are never confused with the actor’s own work, and retire the container once the list is completed or dismissed.
Research on this pattern
- Goal-gradient acceleration and endowed progress: in field experiments on reward programmes, Kivetz, Urminsky and Zheng (2006) found that people accelerate their effort as they get closer to completing a goal, and that participants given a ten-stamp card with two stamps already credited completed it at a higher rate and faster pace than participants given an eight-stamp blank card requiring the exact same effort. First-run checklists apply this by crediting account creation or initial setup as the first completed step.
- Outcome-centred vs activity-centred checklists: reviewing onboarding progression systems across production software, Hulick contrasts checklists whose steps culminate in the core domain outcome the actor came for against checklists that stop at account housekeeping or interface tours, observing that activity-only lists leave actors with a completed meter and an unconfigured tool.
Resources & references
- Samuel Hulick, UserOnboard. Progression systems and Success states
- Kivetz, R., Urminsky, O. & Zheng, Y. (2006). The goal-gradient hypothesis resurrected: purchase acceleration, illusionary goal progress, and customer retention. Journal of Marketing Research 43(1)