Common Mistakes When Designing a Sequence-Building Exercise¶

Building a Sequence Flow section takes a few minutes - which is exactly why it's easy to publish one that hasn't been thought through as carefully as the rest of a survey often gets. None of the mistakes below break the section outright; they just quietly produce results that are harder to read, or that answer a slightly different question than the one you actually meant to ask.

Defaulting to a Full Sequence When You Only Need the Top Few¶

The single most common mistake: leaving a Sequence Flow section set to require every item placed, when your actual decision only ever involved a handful of them. Forcing a full ranking of fifteen items when you only care about the top three adds real fatigue - and fatigue late in a long sequencing task tends to show up as careless placement of exactly the items you cared least about ranking accurately in the first place, which quietly degrades data you didn't even need. If you'd be equally satisfied with "here are the top four, in order," use the partial-ranking option instead. See our building guide for exactly how to configure it.

Including Near-Duplicate Items¶

Two items that mean almost the same thing to a respondent split their own placement across a handful of near-identical options, which muddies the item-to-item transition data both Route Flow and Route Network depend on to find a clear consensus. Review your item list specifically for near-duplicates before publishing - not just similar wording, but similar underlying meaning, since respondents will treat them as interchangeable even if the labels look distinct to you.

Writing Items the Way an Internal Doc Labels Them¶

"Step 3: Profile Config" reads slower and less naturally than "Set up your profile." Items should read the way a respondent would actually describe the thing, not the way it's labeled in an internal spec or roadmap document - the gap between internal shorthand and how a real person would phrase it is easy to miss when you've been staring at the internal version for weeks.

Skipping the On-Screen Reminder on a Long Sequence¶

The intro message is shown once, before the task starts. On a sequence of ten or more items, a respondent partway through can genuinely forget the specific framing of what they're being asked to do - and without a persistent on-screen reminder, they're left reconstructing the task from memory rather than reading it again. This is free to add and costs nothing for a short sequence, so there's little downside to using it once a list gets long enough to plausibly need it.

Mismatching Full vs. Partial When Comparing Two Sequence Flow Sections¶

Running one Sequence Flow section as a full sequence and a related one as a partial top-few, then comparing their results as if they answer the same kind of question, produces a comparison that looks cleaner than it actually is - a dominant path computed from a full ranking and one computed from a capped top-4 selection are answering meaningfully different questions, even when several of the same items appear in both. Keep the ranking rule consistent across sections you intend to compare directly.

Treating the Buffer Setting as an Afterthought¶

The "Get ready" buffer defaults to three seconds, and that default is genuinely the right call for a standalone Sequence Flow section - but leaving it at the default inside a longer survey, where it's one of several similar interactive sections in a row, adds friction that a respondent notices the second and third time around. Set it to zero for anything past the first occurrence in a survey with multiple interactive sections.

FAQ¶

Is there a "safe" default I should use if I'm not sure which of these applies to me?
Full sequence with the default 3-second buffer is a reasonable starting point for a short (five-to-eight item), standalone Sequence Flow section - most of the mistakes above only start to matter once a list gets longer, gets embedded alongside similar sections, or the underlying decision genuinely only needed a partial ranking.

Can I fix a live Sequence Flow section's item list after respondents have already started answering?
You can, but changing the pool mid-flight makes results harder to compare across respondents who saw different versions of the task - treat it the same way you'd treat editing any other live question, and avoid it once real responses are already coming in.

Do these mistakes apply to a plain Ranking question too?
Several do - near-duplicate items and full-vs-partial mismatches cause the same problems in a plain Ranking question. The buffer and on-screen reminder settings are specific to Sequence Flow's own interface.

See our full building guide, How to Set Up a Sequence Flow Section (Full vs. Partial Ranking), for the complete walkthrough these mistakes are drawn from.