← All Examples
🛍️Retail / E-commerceQuick

Customer Returns & Satisfaction

Turn returns into insights — understand why products come back and what would prevent it.

Sender: CX Manager at DTC furniture brand
Participant: Customer — Returned a $900 sofa after 2 weeks
This is an illustrative example showing how Willit could be used in retail / e-commerce. All names, quotes, and data are fictional. We never use real customer interviews for marketing purposes.
1

The Brief

The sender described what they wanted to learn. Willit's AI refined these instructions into a natural interview flow.

We've had a significant uptick in sofa returns over the past quarter and I need to understand whether this is a product issue, a photography issue, a description issue, or something else entirely. Each return costs us roughly $400 in logistics and lost margin. This customer returned a 3-seater sofa in the Slate fabric after owning it for 14 days. I want to understand their experience — not to fight the return, which we've already processed, but to genuinely understand what happened. **What specifically didn't meet expectations:** What was wrong? Color, size, feel, quality, smell, firmness? I want the specific thing that made them pull the trigger on a return. Not a general 'didn't work out.' **Website vs reality gap:** How did the product page set their expectations? Were the photos accurate? Was the written description useful? Was there information they needed that wasn't there? **Did they try to make it work:** Did they live with it for a few days and try to like it? Or did they know immediately? Understanding the resolve to return tells me whether this was a marginal mismatch or a clear problem. **Return process experience:** How was the return process itself? Easy, frustrating, confusing? Even if the product failed, the return experience shapes whether they'd shop with us again. **What would have prevented the return:** If we'd done one thing differently — better photos, a swatch sample, a clear size guide, a different product recommendation — would they still have it? **Future purchase intent:** Are they done with us or would they consider buying again, maybe a different product?
2

The Interview

Willit's AI detective conducted a quick interview with a Customer. The conversation explored 6 topic areas through natural follow-up questions, adapting in real-time based on the participant's responses.

Identify what specifically didn't meet expectationsIdentify the website vs reality gapUnderstand the decision process to returnAssess the return process experienceDetermine what would have prevented the returnGauge future purchase intent
3

The Report

Willit automatically extracted structured insights from the conversation — scores, goal coverage, key quotes, and red flags.

Interview Scorecard

EngagementSentimentDepthQualityCoverageCoherence

Metric Averages

Engagement
72
Sentiment
52
Depth / Accuracy
76
Info Quality
80
Goal Coverage
88
Coherence
84

Summary

The customer returned the sofa because the fabric color was significantly warmer in person than on the website — the Slate colorway photographed as cool grey but arrived as warm greige. They lived with it for 10 days before deciding it wasn't going to work with their existing furniture. The return process itself was smooth and they remain open to purchasing again if they can see the real color first.

Goal Coverage

Covered

Identify what specifically didn't meet expectations

  • The Slate colorway was expected to be a cool grey based on product photography — the actual piece was noticeably warm with beige undertones
  • Size and comfort were exactly as expected — color was the sole driver of the return
Covered

Identify the website vs reality gap

  • Product photos were shot in a bright white studio with cool-toned lighting — this appears to have shifted the color significantly
  • Written description did not include any warm undertone language — customer read 'sophisticated grey' and expected precisely that
Covered

Understand the decision process to return

  • Kept the sofa for 10 days hoping they'd adjust to it — returned only after confirming it clashed with their rug in different lighting conditions
Covered

Assess the return process experience

  • Return process was described as 'surprisingly easy' — scheduled pickup online, no questions asked, refund arrived within 5 days
Covered

Determine what would have prevented the return

  • A fabric swatch program would have prevented the return — customer would have seen the real color before committing to a $900 purchase
  • Color description needs to acknowledge warm undertones — 'warm greige' or 'greige-grey' rather than 'sophisticated grey'
Partial

Gauge future purchase intent

  • Would consider buying from the brand again — specifically mentioned interest in the Charcoal colorway if swatches were available

Gap: Did not explore whether the return experience changed their perception of the brand's trustworthiness for a future high-value purchase

Key Quotes

“It wasn't grey. It was beige with aspirations. The photo must have been taken in a room with no windows.”
“I kept it for a week and a half thinking I'd come around to it. I didn't. It just didn't go with anything I own.”
“If I could have seen a swatch first, none of this happens. I've done it with other brands. It's worth whatever it costs.”

Red Flags

  • Studio lighting in product photography is systematically misrepresenting warm-toned fabrics as cool-toned — likely affecting multiple colorways beyond Slate
  • Product copy does not disclose undertone information — customers are making $900+ decisions without the color context they need

Follow-up Suggestions

  • Audit all warm-toned fabric colorways for photography accuracy — reshoot in natural or mixed lighting if necessary
  • Update product copy for all fabric colorways to include explicit undertone language (warm, cool, neutral)
  • Evaluate swatch program feasibility — competitor offering swatches is named as a direct factor in their confidence to purchase there vs here

Ready to run your own AI interviews?

Set up your first interview in under 5 minutes.

Get StartedMore Examples