Unit 02.04: Writing the chart brief before opening a tool

The chart brief is written before any tool is opened, and it is one page.

Nine fields, all decided in advance

Decision, audience, what they know, question, comparison, measure, chart type, takeaway sentence, and known limits.

The code prints a completed brief.

import json

brief = {
    "decision": "whether to keep the Tuesday promotion after March",
    "audience": "the retail department head",
    "they_already_know": "the promotion exists and roughly what it costs",
    "question": "did Tuesday revenue rise relative to comparable days?",
    "comparison": "Tuesday against Wednesday and Thursday, 12 weeks either side",
    "measure": "revenue per open hour, to control for opening times",
    "chart_type": "slope chart, before and after, one line per day",
    "takeaway_sentence": "Tuesday revenue rose 11% while comparable days were flat.",
    "known_limits": ["12 weeks either side only", "one store",
                     "no control for weather"],
}
print(json.dumps(brief, indent=2))

print("\nEvery field is decided before a tool is opened. `takeaway_sentence`")
print("is written as a claim, so the chart either supports it or it does not.")

measure is the field that does quiet work: revenue *per open hour* rather than revenue, because opening times differ between days and would otherwise confound the comparison. That decision belongs in the brief, not in a footnote discovered later.

takeaway_sentence is written as a claim with a figure in it, so the finished chart either supports it or the brief was wrong.

The mistake this prevents

The mistake is writing the brief after the exploration, from what you found. Written then, it justifies whatever you happened to build. Written first, it is a specification you can fail to meet - which is the point.

Takeaway

Write the brief before opening a tool: decision, audience, comparison, measure, takeaway sentence and limits. The measure often needs normalising, and that decision belongs here.

Mark complete