Unit 11.04: When interactivity is a substitute for a decision

An interactive chart with no default finding is a data set with a user interface.

Interactivity as indecision

Four situations, sorted by whether the interaction serves the reader or the author.

The code lists them.

SITUATIONS = [
    ("we could not decide what to show, so we made it filterable",
     "interactivity as indecision"),
    ("readers genuinely need their own store's view",
     "interactivity as a real requirement"),
    ("the dashboard has 14 filters and no default view",
     "the reader does the analysis you were paid for"),
    ("hover shows the underlying record for outliers",
     "interactivity as depth on demand"),
]
print(f"{'situation':56} verdict")
for situation, verdict in SITUATIONS:
    print(f"{situation:56} {verdict}")

print("""
The first and third are the same failure: an interactive chart with no default
finding is a data set with a user interface.

A dashboard should answer its question before anyone touches a control.
Filters refine that answer; they do not replace it.
""")

The first and third are the same failure. "We could not decide what to show, so we made it filterable" hands the analysis back to the reader, who has less context than you and less time.

The second and fourth are genuine: readers who need their own store's view, and detail available on demand for the points that warrant it.

The mistake this prevents

The mistake is measuring a dashboard by how much it can show. The measure is whether a reader who touches nothing leaves with the finding - everything else is refinement on top of that.

Takeaway

A dashboard should answer its question before anyone touches a control. Filters refine the answer; they do not substitute for having one.

Mark complete