Research & Personas Insight Report

What you'll produce

A one-to-two-page research & personas insight report for FreshCart's first-time-shopper drop-off: the user segment you're designing for, one realistic primary persona, 3–5 sharp insights (each labeled fact or assumption), and a set of neutral, past-behavior interview questions you'd ask real first-timers. This is the Empathize → Define foundation of your entire case study — every later deliverable (the IA, the flows, the wireframes, the test plan, the final UI) traces back to a problem you framed here. It proves the most important Topic 4 skill and the one hiring managers probe hardest: that you design from real user understanding instead of your own taste, and that you can tell a fact from an assumption and write a question that doesn't lead the witness. A designer who can say "I redesigned checkout because research showed X" beats one who says "I thought it looked better" every time.

Instructions

1. Name the segment, narrowly. Not "FreshCart users" — the people who churn. Write one sentence defining the exact segment (e.g., "first-time shoppers placing their very first order"). Add one line on why this segment (tie it to the 58% first-order abandonment — these are the users leadership hired you to save).
2. Separate what you know from what you assume. Make two quick lists from the brief: Facts (things stated or measured — the 58% drop-off, the "confusing"/"checkout is a maze" reviews, the under-3-minutes goal) and Assumptions (things you suspect but haven't verified — why they drop off). You'll label every insight against these. Never let an assumption masquerade as a fact; that's the single fastest way to fail this rubric and to mislead a design team.
3. Build one primary persona. Give them a name, age, and a one-line context that explains why they're new to grocery delivery (just moved, new baby, lockdown habit, etc.). Then write four fields: Goal (what they want from this single first order), Context (device, time pressure, where/when they shop), Frustrations (what makes them quit — drawn from the reviews, not invented), and A quote (one sentence in their voice that captures the pain). Keep it to a type of person, not a real individual. Avoid vanity details (favorite coffee) that don't change a design decision.
4. Write 3–5 insights, each labeled and each design-relevant. Format every one as: the insight → (fact / assumption) → so what for the redesign. An insight is not a feature ("add a progress bar") and not raw data ("58% abandon") — it's the interpretation that points at a design move ("first-timers can't tell how many steps checkout has, so they bail mid-flow — fact that 58% drop between browse and checkout; assumption that step-opacity is the cause; so the redesign needs a visible, finite progress indicator"). At least one insight must be an honest assumption you'd validate, not a fact dressed up as one.
5. Write 5 neutral, past-behavior interview questions. Re-read Topic 4 Lesson 4.2. Every question must ask about a real past event ("Tell me about the last time you…"), never a hypothetical ("Would you use…") and never a pitch ("Don't you think checkout should be faster?"). At least one should probe the moment of abandonment directly. These are the questions you'd run with ~5 real first-timers to confirm your assumptions before you design anything.
6. State the one assumption that, if wrong, breaks the redesign. Close with the single load-bearing belief — the thing that, if false, means you're solving the wrong problem beautifully (Topic 3). Name it and name how you'd test it.
7. Keep it to 1–2 pages. The deliverable is sharp insights everyone can act on, not a fat report (Topic 4 Lesson 4.3). Cut anything that doesn't change a design decision.

Rubric

The app's AI scores your submission against these criteria and gives feedback. Levels: Needs work (1) / Solid (2) / Excellent (3). Passing = every criterion at Solid or above; revise and resubmit until it passes.

• Narrow, justified user segment — 1: "all FreshCart users" or vague · 2: a clear segment (first-time shoppers) · 3: a sharp segment tied to the 58% first-order drop-off and why winning the first order matters.
• Realistic primary persona — 1: a list of vanity traits or a real named individual · 2: a believable user type with goal, context, and frustrations · 3: a persona whose context and frustrations come straight from the brief's evidence and clearly point at design decisions (incl. a quote in their voice).
• Insights labeled fact vs. assumption — 1: insights unlabeled, or assumptions stated as facts · 2: 3–5 insights each labeled and design-relevant · 3: every insight reads insight → fact/assumption → so-what, with at least one honest assumption to validate.
• Insights are interpretations, not data or features — 1: restates analytics ("58% abandon") or jumps to a feature ("add a progress bar") · 2: most insights interpret the behavior · 3: each insight explains why users behave that way and what it means for the redesign.
• Neutral, past-behavior interview questions — 1: leading, hypothetical, or pitchy · 2: mostly neutral and past-focused · 3: 5 sharp, non-leading, past-behavior questions, at least one probing the moment of abandonment.
• Load-bearing assumption named and testable — 1: missing, or a trivial assumption · 2: a real riskiest assumption · 3: the genuinely flow-breaking assumption, with a concrete way to test it before designing.
• Coherence with the case study — 1: a standalone report disconnected from the redesign · 2: insights reference the redesign · 3: explicitly hands off to later deliverables (IA/flow, usability test), framing this as the Empathize/Define foundation.