Jaguar Land Rover

28 pages of audit, owned end-to-end. Every finding, annotation and recommendation was mine. Three tiers of recommendations, all three shipped. Copy, structural and functional changes all landed inside the engagement.

Rebuilt. Six years on, JLR redid the checkout from the ground up - the deeper outcome the audit had implied.

At a glance

Rebuilt. Six years on, JLR redid the checkout from the ground up - the deeper outcome the audit had implied.

HYPNOTISES

A luxury brand can be undone by a checkout

Both Jaguar and Land Rover customers buy into something specific: appreciation for craft, attention to detail, robust technology, generational trust. Those expectations don't pause when the customer moves from the showroom to the website. What I found auditing the New Car Online Sales (NCOS) checkout was that the digital experience was doing the opposite. Not catastrophic failures - small ones. A button that quietly deleted user input. A reservation fee whose "fully refundable" framing was visually buried. A part-exchange flow that asked the same question twice in different language. A chatbox that covered the price the user was trying to read.

Each one survivable on its own; together, a slow erosion of the same trust the brand had spent decades building. Each small flaw is forgivable on its own. Stack a dozen of them across a single transaction and you have asked a luxury customer to do something a luxury customer should not have to do: tolerate friction.

A luxury brand can be undone by a checkout

Both Jaguar and Land Rover customers buy into something specific: appreciation for craft, attention to detail, robust technology, generational trust. Those expectations don’t pause when the customer moves from the showroom to the website. What I found auditing the New Car Online Sales (NCOS) checkout was that the digital experience was doing the opposite. Not catastrophic failures - small ones. A button that quietly deleted user input. A reservation fee whose “fully refundable” framing was visually buried. A part-exchange flow that asked the same question twice in different language. A chatbox that covered the price the user was trying to read.

brief

The brief

Audit the final stretch of the online vehicle purchase - order summary through to reservation payment - and return three tiers of recommendations separated by what it would take to ship them.

Tier 1 - copy only. No development. Words on screens, microcopy, button labels (adjustable with the CMS)

Tier 2 - small structural changes. Component re-arrangements, breakpoint fixes, field sizing, hierarchy.

Tier 3 - functional changes. Flow re-architecture, behavioural fixes, new states.

The constraints shaped the engagement: a third-party checkout we did not own, limited development bandwidth, a complex non-linear user flow, a brand whose tolerance for mis-step was - appropriately - very low. The 3-tier framework was how I bridged a tension the brief contained without naming: the brief asked for impact with minimal changes; the findings, in their depth, said redo the checkout entirely.

The brief

Audit the final stretch of the online vehicle purchase - order summary through to reservation payment - and return three tiers of recommendations separated by what it would take to ship them.

Tier 1 - copy only. No development. Words on screens, microcopy, button labels (adjustable with the CMS)

Tier 2 - small structural changes. Component re-arrangements, breakpoint fixes, field sizing, hierarchy.

Tier 3 - functional changes. Flow re-architecture, behavioural fixes, new states.

process

How I worked

Three methods, in order of confidence. Analytics first - funnel drop-offs by step, device and acquisition channel. The single biggest fall-off sat at Finance & Valuation. Field studies next - ten depth interviews in Old Street coffee shops with prospective JLR customers, at the saturation threshold for qualitative usability research. Stakeholder workshop last - journey mapping plus affinity mapping; the journey map became the artefact everyone in the room afterwards used to talk about the work.

How I worked

NCOS Checkout User Journey, January 2020. Five steps, conversion percentages per step for Land Rover and Jaguar, verbatim user feedback drawn directly from the interviews.

findings

What I found

Step 1 - Summary: trust signals buried.

The single most important phrase on the page - "fully refundable £500 reservation fee" - was set in the same weight as the surrounding copy. Participants who noticed the framing visibly relaxed; those who didn't stayed tense. Bolding it was the highest-leverage Tier 1 fix in the whole audit.

Step 2 - Finance & Valuation: the killing field.

Where the funnel collapsed. The flow asked the same question multiple times in different language, surfaced finance options to customers who had said they were paying cash, and offered a "Cancel and go back" CTA that silently deleted everything the user had entered.

What I found

Step 1 - Summary: trust signals buried.

The single most important phrase on the page - “fully refundable £500 reservation fee” - was set in the same weight as the surrounding copy. Participants who noticed the framing visibly relaxed; those who didn’t stayed tense. Bolding it was the highest-leverage Tier 1 fix in the whole audit.

Step 2 - Finance & Valuation: the killing field.

Where the funnel collapsed. The flow asked the same question multiple times in different language, surfaced finance options to customers who had said they were paying cash, and offered a “Cancel and go back” CTA that silently deleted everything the user had entered.

Why is this giving me finance options again, since I already stated I'm buying with cash? - Field study participant, Step 2

Why is this giving me finance options again, since I already stated I’m buying with cash? - Field study participant, Step 2

There's so many buttons and calls to action all at once. And all I wanted to do was to get a valuation… I feel overwhelmed. - Field study participant, Step 2 (who then abandoned the valuation entirely)

There’s so many buttons and calls to action all at once. And all I wanted to do was to get a valuation… I feel overwhelmed. - Field study participant, Step 2 (who then abandoned the valuation entirely)

Step 3 - Delivery: small frictions that read as carelessness.

The postcode field was set to a width that suggested a full address was expected; one participant typed their full address into it before realising. The retailer selection language was confusing. The chatbox obscured the part-exchange figure on smaller screens. Form fields could not be navigated with tab and enter. No single one is catastrophic. Cumulatively, they were the design moment that made a customer feel the brand wasn't paying attention.

Step 4 - Payment: trust at the moment of money.

The reservation page asked the customer to part with £500 and presented them with paragraphs of legal copy. Participants slowed down, got tense, re-read the "fully refundable" framing they hadn't noticed earlier.

Step 3 - Delivery: small frictions that read as carelessness.

Step 4 - Payment: trust at the moment of money.

The reservation page asked the customer to part with £500 and presented them with paragraphs of legal copy. Participants slowed down, got tense, re-read the “fully refundable” framing they hadn’t noticed earlier.

Why does it say it's fully refundable by your chosen retailer. Surely I'm buying from Jaguar… - Field study participant, Step 4

Why does it say it’s fully refundable by your chosen retailer. Surely I’m buying from Jaguar… - Field study participant, Step 4

recomMendations

The 3-tier recommendations

Tier

What is required

Examples from the audit

Tier 1

Copy changes only - no dev.

Bolding and surfacing the "fully refundable £500" framing. Shortened reservation copy. Microcopy aligned to user mental models.

Bolding and surfacing the “fully refundable £500” framing. Shortened reservation copy. Microcopy aligned to user mental models.

Tier 2

Small structural changes - minor dev.

Narrower postcode field. Sticky order summary. Chatbox collapsed by default. Tab/enter accessibility. Secondary CTA alignment fixed below 1440px.

Tier 2

Functional changes - full dev.

Removing the destructive "Cancel and go back". Allowing part-exchange valuations to be edited. Restructuring the Finance & Valuation sub-flow.

Removing the destructive “Cancel and go back”. Allowing part-exchange valuations to be edited. Restructuring the Finance & Valuation sub-flow.

outcomes

Two outcomes, six years apart

The 2020 outcome. All three tiers shipped inside the engagement. The reframed £500 reservation fee was the single largest commercial-impact change of the Tier 1 set; the destructive "Cancel and go back" behaviour was removed; part-exchange valuations became editable; the Finance & Valuation sub-flow was restructured.

The 2026 outcome. Some time after 2020, JLR rebuilt the online checkout from the ground up - live stock visibility, in-flow vehicle configuration, retailer-integrated pricing, a much richer experience overall. None of my specific recommendations are identifiable in it as discrete artefacts. They belong to the era of the page I audited, which no longer exists. But the deeper diagnosis - that the digital experience has to match the luxury brand promise, or the brand pays for the gap - is now visibly delivered.

The rebuild was, in effect, the recommendation the audit had implied but the engagement could not commission. A Tier 4 - full flow rebuild - sat outside the brief and outside what the 2020 timeline could carry. The audit pointed there anyway. Six years on, JLR went there.

The NCOS Checkout audit was one work-stream of the broader GAINS programme - which projected £1.17M of annualised additional profit and a +33% programme ROI within its first three months. The companion case study, Turning every pound twice, covers the wider programme economics.

Two outcomes, six years apart

The 2020 outcome. All three tiers shipped inside the engagement. The reframed £500 reservation fee was the single largest commercial-impact change of the Tier 1 set; the destructive “Cancel and go back” behaviour was removed; part-exchange valuations became editable; the Finance & Valuation sub-flow was restructured.

reflection

Cancel and go back

The title of this case study is also what I am doing with it.

I wrote the first version of this case study in 2020 and kept it on my portfolio. That version was honest about its constraints. It also sells the work short in ways I can now see, and would not allow my own designers to publish today.

Outcomes belong at the front. A six-years-on version can lead with what shipped and what stuck. The reader should know within ten seconds whether the work was worth their time.

"We" is a hiding place. The original is written in the first-person plural. Technically accurate; I was one of a five-person squad. But every finding, every annotation, every recommendation was mine. Owning it isn't arrogance - it's precision.

Process is not the story; findings are. The original is heavy on methodology and light on what the methodology found. The interesting moves a UX designer makes are not the workshops; they are the patterns the workshops surface - and the gap between brief and finding is itself a finding.

Cancel and go back

The title of this case study is also what I am doing with it.

Outcomes belong at the front. A six-years-on version can lead with what shipped and what stuck. The reader should know within ten seconds whether the work was worth their time.

“We” is a hiding place. The original is written in the first-person plural. Technically accurate; I was one of a five-person squad. But every finding, every annotation, every recommendation was mine. Owning it isn’t arrogance - it’s precision.

sources

NCOS Checkout Audit (28 pages, January 2020), Recommendations playback (January 2020), field-study notes from ten moderated user interviews (Old Street, late 2019), NCOS Checkout User Journey Map, GAINS Interim Playback and ROI summary (Dec 2019 / Jan 2020). Live-site verification: JLR UK, May 2026.

Any comments or questions?

I’d love to hear from you.

Any comments or questions?

I’d love to hear from you.

Contact