01 — E-commerce

Why users abandon checkout — and what to do about it

A moderated usability study with 12 participants that uncovered five critical trust gaps in a retail checkout flow. Findings directly shaped a redesign that reduced abandonment by 28%.

My role

Lead researcher

Duration

6 weeks

Methods

Usability testing, interviews

Participants

12 (moderated)

The problem

The client, a mid-size fashion retailer, had seen checkout abandonment climb to 74% over 18 months following a platform migration. Conversion rate analysis pointed to the payment step as the biggest drop-off point — but analytics alone couldn't explain why.

I was brought in to find out what was actually happening for users at this stage, and to give the product team concrete, evidence-backed direction for a redesign.

Research question: What are the specific barriers that prevent users from completing a purchase once they've reached the checkout flow?

Research approach

I chose moderated usability testing over unmoderated because I expected the barriers to be emotional and contextual — things like anxiety, confusion, and trust — which are better explored through live conversation than remote click-tracking.

1

Screener and recruitment

Recruited 12 participants who had made an online clothing purchase in the past 3 months but had also abandoned at least one checkout in the same period. Split evenly by gender and age range (25–55).

2

Session design

60-minute sessions, each beginning with a warm-up interview about online shopping habits, followed by a task-based usability test using a staging environment loaded with real product data. Participants were asked to purchase a specific item using a test card.

3

Observation and note-taking

Sessions were recorded with consent. A product manager and designer observed 4 sessions each as silent observers, which later helped align on findings without needing lengthy presentations.

4

Synthesis

Used affinity mapping with the observing stakeholders to cluster observations. Coded transcript excerpts against a framework of trust, clarity, and control barriers. Severity ratings applied to each issue using a 4-point scale.

What we found

Five recurring themes emerged. All 12 participants encountered at least three of them during their session. The most impactful — and the one that surprised the team most — was the near-complete absence of trust signals at the payment stage.

"I just don't know what's going to happen after I click pay. There's nothing there that makes me feel like it's safe."

— Participant 7, 38, returning online shopper
01

No trust signals at payment

11 of 12 participants hesitated or commented on the absence of security badges, accepted payment logos, or reassurance copy at the payment step.

02

Unclear total until the final step

Delivery cost only appeared on the final confirmation screen. 8 participants expressed frustration or distrust when the total changed from what they'd expected.

03

Guest checkout buried

The guest checkout option was de-emphasised to drive account creation. 7 participants either missed it entirely or assumed they had to register to buy.

04

Return policy not visible

For first-time buyers especially, the return policy link was either missing from the checkout flow or buried. This was consistently cited as a confidence barrier.

What changed

The product team used the findings to brief a checkout redesign, prioritising the top three severity issues. I presented findings directly to the CPO and CTO, alongside the video highlights reel, which removed the need for further convincing.

28%

Reduction in abandonment rate post-redesign

5

Issues surfaced, all actioned in next sprint

3 wks

From findings to design brief

A follow-up round of usability testing (5 participants, unmoderated) was run 8 weeks after launch to validate that the trust signal changes had resolved the core concerns. All five participants completed checkout without hesitation at the payment step.

What I'd do differently

The study was narrowly scoped to the checkout flow, which was the right call given the timeline. But I'd have benefited from one or two sessions that started from product discovery — browsing to basket — to understand how trust is built (or eroded) before users even reach checkout.

I'd also include a short exit survey alongside the moderated sessions to gather quantitative confidence ratings at each step, which would have given us a cleaner severity ranking to present to leadership.

Usability testing Think-aloud protocol Affinity mapping Thematic analysis Stakeholder communication
← Back to all projects Next project GP referral research →