The redesigned CeraVe home page on a laptop: a banner reading Find your routine in 4 questions, with a Find my routine button.
On a phone: the retailer list for Daily Moisturizing Lotion, sorted by price, with stock and a coupon shown for each retailer.

CeraVe

CeraVe’s website doesn’t sell anything. Its job is to help shoppers pick the right product and hand them to a retailer, sure of both. In our usability test, most shoppers lost that confidence before they reached a store.

My role
UX Researcher. I rotated between moderator, observer and note-taker across sessions.
Team
Vedant Atram, Valerie Garcia, Dhwani Parekh, Tory White
Method
Moderated usability test · 5 shoppers · 3 tasks · Oct–Nov 2025
Outcome
12 recommendations, all built into a coded prototype in 2026

The brief: a store that doesn’t sell

CeraVe is one of the best-known skincare brands in the US, but you can’t check out on cerave.com. Every “Buy” sends you to Amazon, Target, Walmart or another retailer. That makes the website a decision tool: it earns its keep only if shoppers leave it knowing what to buy and where.

An earlier heuristic evaluation and cognitive walkthrough had flagged problems with product discovery, the skincare quiz and comparison. Our team ran a usability test to see how easily people could get through real shopping tasks, and how they felt doing it.

For shoppers

Find a product that fits their skin, understand how it differs from the alternatives, and know the price before they leave.

For CeraVe

Hand every visitor to a retailer with confidence, so the visit ends in a purchase and in trust in the brand.

The shopping trip we tested

Five women aged 19 to 39 who buy skincare took part in 30-minute remote sessions on Microsoft Teams. Each worked through three tasks while thinking aloud, rated each one on the After-Scenario Questionnaire, and finished with the System Usability Scale.

TaskFinished on their ownEase, average rating
Find where to buy Daily Moisturizing Lotion online2 of 53.0 of 5
Build a routine with the Skincare Solution Finder1 of 52.6 of 5
Compare the Hydrating and Foaming Facial Cleansers0 of 53.6 of 5

Comparing scored highest on ease even though no one finished it alone. We think participants were rating it against the two harder tasks before it.

Where confidence dropped

Getting to a product was never the problem: everyone found it through the menu or search. Confidence fell at the moments a shopper had to decide something, and it fell at a different step in each task.

Buy · find where to buy

  1. Went wellFind the product

    Everyone got there, through the menu or search.

  2. Stalled or gave upPick a retailer

    A list of logos with no prices, sizes or stock. Three couldn’t decide.

  3. Stalled or gave upLeave for checkout

    Every participant was unsure about being sent to another site.

Choose · build a routine

  1. HesitatedFind the quiz

    Most looked for their concern in the menus instead.

  2. HesitatedAnswer it

    Those who started got through the questions.

  3. Stalled or gave upTrust the results

    Generic picks, no reasons. Puffy eyes returned no eye cream.

Compare · two cleansers

  1. Went wellOpen a cleanser

    Finding each product was easy.

  2. Stalled or gave upFind Compare

    Four of five never saw the button at the top of the page.

  3. Stalled or gave upUse the chart

    A fixed chart. No one could pick what was in it.

Each step summarises what participants said and did in the sessions. It isn’t a measured score.

What shoppers expected, and what they got

Each drop came from a mismatch between how participants already shop online and how the site asked them to shop.

Moment 1

Where should I buy this, and for how much?

Shoppers expected

To compare retailers on price, size and deals before choosing one, as they would in a shopping app.

The site gave them

Six logos and six identical Buy buttons. The price only appeared on the retailer’s own site.

“I didn’t realize that I’m on some other website.” Participant 2

Moment 2

Which of these is right for me?

Shoppers expected

To start from their concern, like dryness or dark circles, and see options to weigh up.

The site gave them

A quiz that started from skin type and returned one product per step, with no reason for any of them.

“By concern I don’t see anything for dark circles or puffiness.” Participant 3

Moment 3

How are these two different?

Shoppers expected

To tick the products they cared about and see them side by side. One called it a “checkbox situation”.

The site gave them

One small button at the top of the page, leading to a fixed chart of products they hadn’t chosen.

“It’s like read through every single thing.” Participant 2

The redesign, in shopping order

The redesign keeps CeraVe’s look and follows a shopper through the site: choose, then compare, then buy. Each chapter starts with the original screen. On the redesigned screens, hover or focus a numbered note to highlight it.

Moment 2 · Choose

Start from the concern, and explain every pick

Concerns now sit in the main menu and lead into a shorter finder. Its results say why each product is there and let shoppers swap or compare it.

Original quiz results: Hi! Here is your routine, with one cleanser and a Buy Online button.
BeforeOne product per step and no reason given.
Skincare menu open with By product, By concern and Tools columns, and a blue Find my routine panel.

Concerns sit in the Skincare menu and open the finder with that concern already chosen.

  1. Concerns lead straight into the finderMoment 2 · Choose
  2. Says what the finder does and how long it takesMoment 2 · Choose
  3. "Find my routine" in the main menuMoment 2 · Choose
Finder question 1 of 4, What would you like help with?, with six concern tiles and Dark circles and puffiness selected.

Four short questions. “Dark circles and puffiness” is a concern you can pick.

  1. 4 questions, with progress you can seeMoment 2 · Choose
  2. "Dark circles and puffiness" is now a concernMoment 2 · Choose
  3. Up to two concerns, in everyday wordsMoment 2 · Choose
Routine results for dryness and puffiness: the answers as tags, then Cleanse with Hydrating Facial Cleanser and Treat with Eye Repair Cream, each with reasons, a Swap menu and buttons.

Results keep your answers on screen, explain every product, and let you swap or compare it. Puffy eyes now return an eye cream.

  1. Your answers stay visible and editableMoment 2 · Choose
  2. "Why this fits you", tied to each answerMoment 2 · Choose
  3. Swap any product for an alternativeMoment 2 · Choose
  4. Compare the options for that stepMoment 3 · Compare
  5. Puffy eyes now return an eye creamMoment 2 · Choose

Moment 3 · Compare

Pick what to compare, then compare it

Compare now works the way participants expected: choose products as you browse, then see them side by side.

Original cleansers page: one Compare facial cleansers button above the filters and products.
BeforeOne Compare button at the top of the page, easy to miss.
Cleansers page with filters, product cards each with a Compare checkbox, two ticked, and a tray reading 2 products selected, Compare 2.

Tick Compare on any product. A tray keeps count and opens the table.

  1. A Compare checkbox on every productMoment 3 · Compare
  2. A tray counts your picks and opens the tableMoment 3 · Compare
  3. Filter by skin type and concernMoment 2 · Choose
Compare table for two cleansers with Add a product, a Show only differences switch, Remove links and highlighted rows where they differ.

You choose what’s in the table, and it highlights what’s different.

  1. Add a third product of your choiceMoment 3 · Compare
  2. Hide the rows that are the sameMoment 3 · Compare
  3. Differences highlightedMoment 3 · Compare
  4. Remove any productMoment 3 · Compare

Moment 1 · Buy

Show the price before shoppers leave

Shoppers choose a retailer by price and discount first, so the product page and the retailer list now lead with both, and say plainly where checkout happens.

Original retailer pop-up: six retailer logos, each with a black Buy button, and no prices, sizes or stock.
BeforeLogos and Buy buttons, with no price, size or stock until you leave.
Product page for Daily Moisturizing Lotion: size options each with a from-price, a live lowest price with stock count, Compare retailer prices and Find nearby stores buttons, and a note that checkout happens at the retailer.

The product page shows the lowest price for the size you pick, and says plainly that you’ll check out with a retailer.

  1. Every size shows its lowest priceMoment 1 · Buy
  2. Price and stock up front, before you leaveMoment 1 · Buy
  3. "Compare retailer prices", not a vague "Buy online"Moment 1 · Buy
  4. "Where to Buy" renamed "Find nearby stores"Moment 1 · Buy
  5. Says plainly that checkout happens at the retailerMoment 1 · Buy
Retailer list for Daily Moisturizing Lotion, 12 oz: Walmart lowest at $15.49, then Amazon, Ulta Beauty, CVS with a coupon, Walgreens, and Target out of stock.

The retailer list is sorted by price, shows stock and coupons, and every button says where it goes.

  1. Sorted by price, with the lowest labelledMoment 1 · Buy
  2. Coupons shown before you chooseMoment 1 · Buy
  3. Names the retailer and says it opens their siteMoment 1 · Buy
  4. Out-of-stock retailers stay visible, but lastMoment 1 · Buy
Find nearby stores page: a ZIP search with three store results, then products grouped by category under What you can find in stores.

“Find nearby stores” now means stores, and groups products by category so the page doesn’t overwhelm.

  1. The page says what it is: stores, not online retailersMoment 1 · Buy
  2. Each store shows distance, hours and what it carriesMoment 1 · Buy
  3. Products grouped by category, not one long listMoment 1 · Buy

From recommendation to build

After the sessions our team wrote 12 recommendations and rated each from 1 (low) to 5 (high) for severity. In the redesign I built every one of them, in CeraVe’s existing visual style, so the change is in how the site works rather than how it looks.

#SeverityWhat we recommendedWhat I built
15 of 5Show current prices for each online retailer, so shoppers don’t click back and forth between sites.Prices for every retailer, sorted lowest first
24 of 5Show which sizes each retailer sells.Every size shows its own lowest price
33 of 5Show which retailers have coupons, since shoppers choose by price and discount first.Coupon tags on the retailer list
44 of 5Rename “Where to Buy” to “Find nearby stores” so it isn’t confused with buying online.“Find nearby stores” for stores, “Compare retailer prices” for online
53 of 5List products under categories on the Where to Buy page.The stores page groups products by category
64 of 5Label the quiz clearly so shoppers know what results to expect.The finder says what it does and that it takes about a minute
74 of 5Put the quiz under Skincare, by concern, to match shoppers’ goals.Concerns in the Skincare menu open the finder
84 of 5Let shoppers see and swap in other products in their results.A Swap menu and Compare options on every step
95 of 5Explain why each recommended product suits the shopper.“Why this fits you” on every pick, tied to their answers
105 of 5Make the Compare button far more visible.A Compare checkbox on every product card, with a tray
115 of 5Make the comparison chart interactive, so shoppers choose what to compare.A table you build: add, remove, show only differences
123 of 5Make the sticky secondary menu clearly look like navigation.A navy “On this page” bar with clear labels

I added one change of my own: “Dark circles and puffiness” as a concern in the finder, after the participant who asked for help with puffy eyes got no eye cream.

How we’ll know it worked

The prototype is built and a scripted run completes all three tasks, but it hasn’t been tested with shoppers yet. The plan is to repeat the same three tasks with five new participants and measure against this study’s results. The targets are mine.

MeasureThis studyTarget
Find where to buy, on their own2 of 54 of 5
Build a routine with the finder, on their own1 of 54 of 5
Compare two cleansers, on their own0 of 54 of 5
Ease rating for each task2.6 to 3.6 of 54 of 5 or higher
System Usability Scale54.5Above the 68 average

Live prices depend on data from each retailer. Before building this for real, I’d check with engineering and retail partners which retailers can share price and stock, and how often it updates.

Limitations