At Baymard, we’ve just released a new UX benchmark with 10 “Mass Merchant” UX case studies.

This follows from our large-scale user testing of Mass Merchant UX and adds to our existing ecommerce UX benchmark.

In this article, we give you a snapshot of the overall UX performance.

10 Mass Merchant UX Case Studies and the Overall Performance

These are the 10 in-depth Mass Merchant UX case studies.

The 10 sites have been manually assessed across 500+ research-based UX parameters relevant to Mass Merchant ecommerce, resulting in 5,600+ weighted UX performance scores and 4,600+ best practice examples from these desktop and mobile sites.

Each of the 5,600+ UX performance scores from the 10 Mass Merchant and Department Store case studies is summarized in the interactive scatterplot below — showing you how they perform collectively and individually:

{{ scatterplot-graph: size=big + habitat=public + base-sites=collection:mass-merchant + view-structure-id=gemini-st_k5dbh6l4}}

The overall UX performance of the Mass Merchant sites ranges from “poor” to “good”, with most sites in the “mediocre” to “decent” range.

Here are some of the overarching UX issues.

The industry performance highlights user experience challenges across the industry, particularly regarding product discoverability, cost transparency, and the consistency of product data.

While homepages presented the main navigation categories and a broad range of product types, ads occupied prime positions, and intermediary category pages gave no route to a full product list.

Search often struggled with synonyms, misspellings, and sorting for search results that disregarded relevance.

Product lists omitted unit prices, stock status, and the attributes that distinguish similar items, which made product evaluation more difficult.

Furthermore, order cost and delivery timing stayed unstated until late in the flow.

Specifically, multiple product page “Buy” sections provided no estimate of the order total or a date it would arrive, and in checkout, shipping options were frequently labeled by speed rather than by a delivery date.

These friction points, combined with address entry that was largely unaided and account menus whose paths were incomplete and ambiguously named, created a shopping environment that hindered product comparison and efficient checkout.

In summary, with only 1 site delivering a “good” overall user experience, there is plenty of room for improvement.

A publicly available overview of the research and benchmark can be found on our Mass Merchant UX research overview page.

Getting access: all 5,600+ UX performance scores, 4,600+ best practice examples, and the UX insights from researching the Mass Merchant ecommerce industry are available immediately and in full within Baymard. (If you already have access through an account, open the Mass Merchant UX Benchmark collection.)

If you want to know how your Mass Merchant or Department Store desktop and mobile sites perform and compare, then learn more about getting Baymard to conduct a Mass Merchant & Department Store UX audit of your site.

Post this article on X

Related articles

• 4 Ways to Improve UX for Ecommerce Mass Merchant Sites
• Display “Price Per Unit” For Multiquantity Items (86% Don’t)
• Use “Delivery Date” Not “Shipping Speed” (41% Don’t) — From UX Research to Implementation Roadmap
• Provide a “Fully Automatic Address Lookup” Feature (55% Don’t)

At Baymard, we’ve just released a new UX benchmark with 7 “E-cigarette & Vaping” UX case studies.

This follows from our large-scale UX user testing and adds to our existing ecommerce UX benchmark.

In this article, we give you a snapshot of the overall UX performance.

7 E-cigarette & Vaping UX Case Studies and the Overall Performance

These are the 7 in-depth E-cigarette & Vaping UX case studies.

The 7 sites have been manually assessed across 500+ research-based UX parameters relevant to E-cigarette & Vaping ecommerce, resulting in 3,600+ weighted UX performance scores and 2,500+ best practice examples from these desktop and mobile sites.

Each of the 3,600+ UX performance scores from the 7 E-cigarette & Vaping case studies is summarized in the interactive scatterplot below — showing you how they perform collectively and individually:

{{ scatterplot-graph: size=big + habitat=public + base-sites=collection:e-cigarette-and-vaping + view-structure-id=gemini-st_ugn77krh}}

The overall UX performance of the E-cigarette & Vaping sites is “mediocre,” with 5 sites in the “mediocre” and 2 in the “poor” range.

Here are some of the overarching UX issues.

The benchmarked E-Cigarette & Vaping sites highlight significant user experience challenges across the industry, particularly regarding product discoverability, transparency, and data consistency.

While category navigation is plentiful, homepages frequently prioritize promotional content over direct product access, and on-site search systems often struggle with synonyms, misspellings.

Additionally, filters omit essential purchase criteria.

Critical filters like price, ratings, and brand are often absent, while industry-specific options — such as nicotine strength and blend ratio — lack explanations and number of product matches.

Furthermore, users face difficulties comparing products due to incomplete or inconsistent list item information and comparison features, while product pages and carts often fail to provide crucial transparency regarding total order costs (see guidelines #825 and #2320) and delivery timelines (see #802 and #543).

These friction points, combined with fragmented product information across variations and opaque, poorly labeled customer account management (see #2937 and #876), create a challenging shopping environment that hinders efficient product selection and checkout.

In summary, with no sites delivering even a “decent” or better overall user experience, there is plenty of room for improvement.

A publicly available overview of the research and benchmark can be found on our E-cigarette & Vaping UX research overview page.

Getting access: all 3,600+ UX performance scores, 2,500+ best practice examples, and the UX insights from researching the E-cigarette & Vaping ecommerce industry are available immediately and in full within Baymard. (If you already have access through an account, open the E-cigarette & Vaping UX Benchmark collection.)

If you want to know how your e-cigarette or vaping desktop and mobile sites perform and compare, then learn more about getting Baymard to conduct a E-cigarette & Vaping UX audit of your site.

Post this article on X

Related articles

• 5 Essential Filter Types Users Need on Product Listing Pages (57% Don’t Offer All 5)
• 2 Key Design Principles for Product Listing Information (64% Get at Least 1 Wrong)
• Always Explain Industry-Specific Filters (62% Don’t)

At Baymard, we’ve just released a new UX benchmark with 6 new “B2B Electronic Components & Machinery” UX case studies.

This follows from our large-scale user testing on B2B Electronic Components & Machinery UX and adds to our existing ecommerce UX benchmark.

In this article, we give you a snapshot of the overall UX performance.

6 B2B Electronic Components & Machinery UX Case Studies and the Overall Performance

These are the 6 in-depth B2B Electronic Components & Machinery UX case studies.

The 6 sites have been manually assessed across 420+ research-based UX parameters relevant to B2B Electronic Components & Machinery, resulting in 2,600+ weighted UX performance scores and 2,000+ best practice examples from these desktop and mobile sites.

Each of the 2,600+ UX performance scores from the 6 B2B Electronic Components & Machinery case studies is summarized in the interactive scatterplot below — showing you how they perform collectively and individually:

{{ scatterplot-graph: size=big + habitat=public + base-sites=collection:components-machinery + view-structure-id=gemini-st_omiypisn }}

The overall UX performance of the B2B Electronic Components & Machinery sites ranges from “poor” to “decent”, with most sites performing poorly and only one site within the “decent” range.

Here are just a few of the overarching UX issues we observed.

The B2B Electronic Components & Machinery sites largely fell short of the UX standards the procurement context demanded.

Compatible parts, substitutes, and alternatives were almost entirely unsupported across the industry.

In a domain where out-of-stock and end-of-life parts were routine, the inability to reach validated alternatives from the product page left a significant sourcing gap.

B2B ecommerce signals were frequently missing at the point where purchasing decisions needed to be made.

Minimum order quantities and volume price breaks (see guideline #2432), shipping timeframes, and in-stock quantity were inconsistently surfaced across both the Product Table and the buy section.

In a procurement context defined by fixed budgets and project deadlines, their absence left the most critical purchasing inputs unavailable at the point where decisions needed to be made.

Cart information and B2B-specific functionality fell short of procurement expectations

Around a third of sites fail to provide key B2B shipping and delivery details in the cart, omitting information directly relevant to procurement deadlines at the first point where order costs are evaluated.

More than half of the sites provide no “Share,” “Download,” or “Save” features for cart contents.

In a B2B context where orders frequently require internal review or approval before purchase, the absence of these features forces reliance on workarounds such as screenshots or manual transcription.

In summary, across the moments that mattered most in the procurement workflow, evaluation, comparison, and ordering, the sites provided limited support.

As this benchmark shows, with only 1 site delivering a “decent” overall user experience, there is plenty of room for improvement.

A publicly available overview of the research and benchmark can be found on our B2B Electronic Components & Machinery UX research overview page.

Getting access: all 2,600+ UX performance scores, 2,000+ best practice examples, and the UX insights from researching the B2B Electronic Components & Machinery ecommerce industry are available immediately and in full within Baymard. (If you already have access through an account, open the B2B Electronic Components & Machinery UX Benchmark collection.)

If you want to know how your electronic components or machinery desktop and mobile sites perform and compare, then learn more about getting Baymard to conduct a B2B Electronic Components & Machinery UX audit of your site.

Post this article on X

Related articles

• B2B Electronics Sites: 2 High-Level Takeaways from 2,500+ Hours Testing
• B2B Electronics: Always Allow Users to Search Long Lists of Filter Options
• B2B Electronics: Use “Product Tables” to Display Product Listings on Desktop
• B2B Electronic Components and Machinery Sites: New UX Benchmark with Over 4,800 Performance Scores and 3,600 Best Practice Examples

Today at Baymard, we’re announcing the launch of our new Event Ticketing UX research.

This research study focused on primary and secondary ticketing marketplaces, venue platforms, and performing arts centers offering a wide variety of event tickets, including concerts, theatrical performances, sporting events, comedy shows, and live venue entertainment across the USA and UK:

• Major Ticketing Marketplaces & Secondary Vendors: TicketMaster, StubHub, SeatGeek, AXS, Vivid Seats, TickPick, Gametime, Event Tickets Center, TicketWeb, Eventbrite, and Broadway.com

• Venues & Performing Arts Centers: Royal Albert Hall, National Theatre, Barbican, LA Phil, and Fox Theatre

During testing, the participants encountered 2,800+ medium-to-severe usability issues as they attempted to use Event Ticketing sites to explore options and eventually purchase tickets.

Our analysis of these issues has resulted in 370+ guidelines on Event Ticketing industry UX, available in our research findings.

In this article, we’ll provide 3 high-level observations on where users on Event Ticketing sites struggle the most:

• Opaque pricing and late-stage surcharges
• Delayed constraint filtering and hidden unavailability for ticket options
• High interaction cost for spatial and sightline evaluation of individual seats

Unclear Pricing and Ticket Availability Communication Drives Site Abandonment

Shopping for event tickets differs fundamentally from traditional B2C ecommerce because users are not purchasing physical, returnable objects — they are committing to irreversible, future experiences.

While product shopping is conversion-driven and relies on spec comparisons, experience shopping is confidence-driven and must be designed for hesitation.

Because a live event cannot be returned, any poor outcome results in permanently lost time, money, and emotional energy.

”They told me that’s sold out…How do you know what’s available?” While trying to find tickets for “Nurse John” on Ticketmaster.com (first image), this participant became increasingly annoyed at not being able to see which shows still had tickets available. ”This is really frustrating to me. Why would they show me tickets that aren’t even available? That’s dumb”, she remarked, after clicking through to a third event only to discover on the partner page that there were “No more tickets” (second image). Scarcity is a regular challenge on Event Ticketing sites, and many sites fail to handle “sold out” or “low inventory” events well.

Consequently, users approach ticketing platforms with low trust and high anxiety, seeking maximum reassurance before committing.

Despite this need for early clarity, the central UX failure across Event Ticketing sites is hiding essential decision-making data until after users have committed significant effort.

Forcing users into deep-funnel interactions to reveal basic facts about total costs, contiguous seat availability, or physical sightlines multiplies interaction costs and triggers continuous backtracking (“pogo-sticking”).

Across our usability testing, this lack of upfront transparency consistently manifested in three foundational UX issues that severely impede user evaluation and drive site abandonment.

1) Opaque Pricing and Late-Stage Surcharges

”It doesn’t tell you immediately on here how much the tickets are. So why would I hit ‘Buy Now’ if I don’t know how much the tickets are? The last thing I’m going to do is hit ‘Buy Now’ if I don’t know how much it costs.” While searching for tickets to “Haïm Leads Handel & Bach” at the LA Phil (Android), this participant was both perplexed and annoyed by the absence of pricing information in the showtime listings on the “Event Details” page (first image). To discover the cost of attendance, she was forced to navigate to the “Seat Selection” page (second image) — a deep funnel level that many users are unwilling to reach without first confirming a price range. This demonstrates how withholding costs risks immediate funnel abandonment, as users are unlikely to commit to a specific performance without basic price transparency.

”The one thing this one’s lacking is the price. It’s not telling me”, one participant complained while scanning the showtime listings for “PAW Patrol Live!” on StubHub.com (iOS) (first image). Wanting to purchase 5 tickets for his family, price was a critical factor. ”Cause that’s the thing…my kids may watch this for 20 minutes and then they don’t want to see it anymore. So I want to know how much [it is].” To find out, he had to select a date, enter his ticket quantity in a blocking overlay (second image), wait for the seat map to load, and scroll through the ticket listings where the prices were finally revealed (third image). By withholding costs until late in the funnel, the site left a poor impression on the participant and, ultimately, risked abandonment from a price-conscious customer.

When evaluating tickets across live events, users need upfront price transparency so they can quickly compare options against their budget with confidence.

Yet ticketing platforms routinely obstruct this process by hiding total costs until late in the booking funnel.

In testing, participants experienced severe friction when showtime listings and section maps omitted ticket prices, forcing labor-intensive, cross-page price tracking just to uncover basic costs (see guidelines #3117 and #3123).

”$75 in fees, really?…You know they’re just nickeling and diming me. I am not that interested in committing to this.” After selecting 2 seats for a $360 total to “& Juliet” at the Fox Theatre (iOS) (first image), this participant navigated through multiple subsequent pages, investing time and effort into evaluating add-ons and booking parking tickets. However, upon finally arriving at the cart, she was shocked to discover that the final total was heavily inflated by a wave of unexpected surcharges — including $75 in ticket fees and a $3 parking fee (second image), as well as a $5 “Order Charge” (third image). “That’s garbage. You know what? Bake it into the cake. If you want to make the tickets cost more, then I can choose from there, but don’t suddenly add on my fees.” Disgusted by the total lack of upfront pricing transparency, she immediately abandoned her cart to look for tickets elsewhere.

Furthermore, withholding mandatory service fees until the final cart or order review pages triggered intense sticker shock and eroded platform credibility, driving anxious buyers to abandon checkout entirely in favor of transparent competitors (see guideline #3135).

2) Delayed Constraint Filtering and Hidden Unavailability

When searching for tickets, users need immediate clarity on availability and an easy way to specify their group size right from the start.

Yet ticketing platforms routinely turn ticket discovery into an exhaustive trial-and-error process by hiding these essential constraints.

”So this is where I would start to get very frustrated”, complained a participant shopping for tickets on the “Event Details” page (iOS) to “A Midsummer Night’s Dream” at the Barbican (first image). After finding a suitable date on the “Showtime Listing” page (second image) she navigated to the seat map — only to find a few single seats available (third image). Spending almost 4 minutes repeating this process for 2 separate performances and finding insufficient seating on both, she became visibly annoyed. ”I think at this point I would just look for a different show”, she continued and returned to the event list, discouraged by the effort to manually check every date for seats.

In testing, participants repeatedly hit frustrating dead ends deep in the funnel simply because event listings failed to signal a “Sold Out” status early in the flow.

Moreover, because sitting together is a often non-negotiable requirement for group buyers, failing to prompt for ticket quantity prior to showtime, section, or seat-selection pages forced continuous backtracking as users repeatedly loaded seat maps only to find scattered single seats (see guidelines #3114, #3120, and #3126).

3) High Interaction Cost for Spatial and Sightline Evaluation

When assessing ticket value, in addition to weighing price, users will also need to carefully weigh seat location and sightline quality — a detailed evaluation process where they naturally invest considerable time.

Yet ticketing interfaces routinely undermine this process by imposing heavy interaction friction.

”I’m trying to click into it. It doesn’t want to click.” While searching for 2 tickets to a “Lakers vs. Heat” game on TickPick.com, this participant attempted to use the zone map to view seats in section 110. However, when clicking on a section failed to filter and update the ticket listings, she assumed the feature was broken. ”I would say this is a bit faulty in terms of the navigation, because it’s not letting me click”, she explained, repeatedly attempting to interact with the map. This lack of responsiveness prevented her from exploring specific sections and evaluating her ticket options as intended. Combined with higher prices than those seen on a previous site, the broken interaction became a tipping point for abandonment. Frustrated, she chose to return to a competitor offering better deals, prioritizing both price and functional reliability over TickPick’s “faulty” experience.

Multiplying these unnecessary clicks across multiple candidate events, dates, and venue sections quickly causes fatigue, frustration, and site abandonment.

In testing, participants struggled when venue maps acted as static graphics rather than dynamic section filters or failed to highlight a selected listing’s exact physical location.

”I don’t know how this works…It doesn’t seem to give me that information…Yeah, I would not proceed.” This participant was looking for 5 tickets to a classical music concert on AXS.com (iOS) and arrived on the “Ticket Details” page after selecting a ticket within his budget. Wanting to verify his exact seating location and evaluate the stage view, he was disappointed to find only a static map image. Left feeling highly uncertain about the seat’s proximity and perspective, he criticized the lack of information: ”I prefer websites…that kind of tells you like, how it views…There’s a great website, where once you click on a specific seat, it gives you a picture from that seat.” Refusing to complete the transaction blindly, he abandoned the funnel to search for the same ticket on alternative platforms that offer superior visual clarity and interactive seat views.

Furthermore, when sites omitted embedded “Seat-View” previews on the “Ticket Selection” and “Ticket Details” pages, anxious buyers worried about sightline obstructions were frequently driven completely out of the booking funnel to search for venue photos externally (see guidelines #3138, #3452, and #3455).

Provide Transparency, Early Constraint Filtering, and Frictionless Evaluation for Event Ticketing Users

”Wow, this has the seat view too. Oh, this has everything…this is extremely helpful.” While browsing tickets to an ice hockey game on Gametime.co, this participant hovered over a “Price Pin” on the venue map, which activated a tooltip with a “Seat-View” image, and was instantly impressed. “I like this combination,” she continued, referring to the map interactivity and the “Seat-View” thumbnails. ”That really allows me to stay on the page the entire time.” Because the “Ticket Selection” page presented these views both in the listing panel and as interactive tooltips on the venue map, she was able to evaluate seat quality, location, and cost simultaneously without having to navigate deeper into the funnel — empowering her to make an efficient, informed selection.

”Oh, that’s cool. Okay, so this is showing me the view from those seats, which is pretty good. So I’m going to select that.” While evaluating her options on the interactive seat map for a concert at the LA Phil (Android), this participant tapped on a seat she was interested in (first image). This triggered an overlay featuring a photograph of the actual seat view, providing immediate visual context regarding the stage perspective (second image). Satisfied with the clear visibility, she proceeded to confidently buy her tickets.

To resolve these friction points and foster purchasing confidence, event ticketing platforms must shift from deep-funnel data disclosure to an upfront, user-centered evaluation flow.

Platforms can resolve user friction by aligning the booking journey with users’ core needs for early clarity, financial predictability, and visual reassurance.

This requires satisfying users’ need for transparent pricing from their very first interaction — eliminating the anxiety of hidden fees and late-stage sticker shock.

It also means respecting group buyers’ nonnegotiable requirements by validating capacity upfront, sparing them from tedious trial-and-error searches.

Finally, platforms must support users’ need to evaluate ticket value by providing seamless spatial context around seat location and sightlines, giving them the confidence required to commit without leaving the sales funnel.

Users with access to the Event Ticketing industry research can start with the 37 “Essential” UX guidelines, before moving on to consider the rest of the 370+ guidelines on improving Event Ticketing UX.

Implementing the recommendations provided within will help to ensure users have what they need to evaluate ticket options transparently, increasing their confidence when making a purchase decision.

Getting access: all 370+ Event Ticketing UX guidelines are available today within Baymard. (If you already have access through your account, open the Event Ticketing study.)

If you want to know how your desktop site, mobile site, or app performs and compares, then learn more about getting Baymard to conduct a UX Audit of your site or app.

Post this article on X

At Baymard, we’ve just released a new UX benchmark with 10 “B2B Medical & Pharma” UX case studies.

This follows from our large-scale user testing and adds to our existing ecommerce UX benchmark.

In this article, we give you a snapshot of the overall UX performance.

10 B2B Medical & Pharma UX Case Studies and the Overall Performance

These are the 10 in-depth B2B Medical & Pharma UX case studies.

The 10 sites have been manually assessed across 400+ research-based UX parameters relevant to B2B Medical & Pharma, resulting in 3,400+ weighted UX performance scores and 2,600+ best practice examples from these desktop and mobile sites.

Each of the 3,400+ UX performance scores from the 10 B2B Medical & Pharma case studies is summarized in the interactive scatterplot below — showing you how they perform collectively and individually:

{{ scatterplot-graph: size=big + habitat=public + base-sites=collection:medical-pharma + view-structure-id=gemini-st_i06psz9k }}

The overall UX performance of the B2B Medical & Pharma sites ranges from “poor” to “mediocre”.

Here are just a few of the overarching UX issues we observed.

Product documentation and specification details were often incomplete or inconsistent between products, complicating comparison.

While the specifications that were present read clearly, the documents behind them were frequently missing, and the same attribute was often labeled differently from one product to the next (see guidelines #2542 and #839).

Furthermore, product prices were often unclear (see guidelines #792 and #817), and full order cost, and delivery timing were commonly absent from the product page buy section.

Search offered little help when a query was imperfect.

Sites handled exact product titles well, but did much less once a query or an entry was slightly off (see guidelines #355 and #358).

Carts couldn’t be saved, shared, or handed to another person to complete.

B2B users often need to collaborate with others on their orders, but nothing in the cart supported an order assembled by one person and placed by another.

As this benchmark shows, with no sites delivering even a “decent” or better overall user experience, there is plenty of room for improvement.

A publicly available overview of the research and benchmark can be found on our B2B Medical & Pharma UX research overview page.

Getting access: all 3,400+ UX performance scores, 2,600+ best practice examples, and the UX insights from researching the B2B Medical & Pharma ecommerce industry are available immediately and in full within Baymard. (If you already have access through an account, open the B2B Medical & Pharma UX Benchmark collection.)

If you want to know how your Medical & Pharma desktop and mobile sites perform and compare, then learn more about getting Baymard to conduct a B2B Medical & Pharma UX audit of your site.

Post this article on X

Today at Baymard, we’re announcing the launch of our new updated and expanded Accounts & Self-Service UX research.

We first investigated UX issues in Accounts & Self-Service in 2018.

In this update, we’ve retested all the UX issues observed in 2018, verified all our UX solutions, wrote new guidelines based on newly observed Accounts & Self-Service UX issues, and rewritten the bulk of our verified guidelines for improved usefulness and clarity.

This work has resulted in 57 new and updated Accounts & Self-Service guidelines based on 1,400+ usability issues observed during testing.

The 57 guidelines are a key foundation for ensuring a high-performing Accounts and Self-Service UX.

Additionally, we’ve included mobile test observations for the first time.

As a result, this study represents our most comprehensive findings on Accounts & Self Service ecommerce UX.

In this article, we’ll highlight 3 high-level insights from our new Accounts & Self-Service UX research findings.

New Insights for Accounts & Self-Service UX

Compared to 2018, the landscape of Accounts & Self-Service has changed dramatically, particularly in 3 areas.

1) Account Security and Sign In

At J.Crew, participants updating their passwords were provided no guidance regarding the site’s password requirements, and they were unable to view their input, further increasing the likelihood of errors (first image). It is only after a password that doesn’t meet the requirements is submitted that the site discloses the requirements (second image).

The standard username and password combination required to create and sign in to an account is quickly being replaced by passkeys (e.g., fingerprint, Face ID, or device PIN) and one-time passcodes (OTPs), which are considered by many users to be more convenient — and more secure.

Indeed, account security is also top of mind for users — especially when personal information and saved payment methods are involved.

“I use two-factor authentication — it seems just to be standard now to get a code sent to either my phone number or email address. It seems like a lot of websites are doing that now for safety reasons”, remarked a participant viewing the sign in preferences at Walmart. Participants generally appreciated having the option to enable two-factor authentication (2FA).

“Two-factor authentication — not all the time, I’m going to be really honest. It’s kind of site-dependent, but do I use it — especially if it’s a large account or a large website — something I’m doing a lot on. It’s almost always on, but not always”, explained a participant at Home Depot, noting the two-factor authentication (2FA) banner at the top of the “Security & Password” page.

“I feel good that whoever may or may not have the password will not be able to log in anymore, but to ensure that doesn’t happen again, I would set up 2-factor authentication”, remarked a participant at Target. After updating her password she took the extra step to enable two-factor authentication (2FA). “It is enabled. I’m assuming the next time I sign in, I would need to receive a text message or an email code to be able to log in.” Offering optional account security features is a low-friction way for sites to allow security-conscious users to opt in without burdening other users who are less risk-averse.

During recent testing, 81% of participants reported saving credit card details in their account at sites they shop at frequently, and 60% appreciated when two-factor authentication (2FA), which requires confirmation from a user’s verified account email or phone number before being able to sign in, was available to activate.

It’s therefore key that ecommerce sites move beyond traditional username and password combinations to support users signing in to their accounts.

Indeed, as observed during testing, strong visible and optional security features such as two-factor authentication (2FA) and device management build confidence and trust with users.

However, these features must be implemented carefully to avoid becoming a roadblock to users checking out or managing their account details.

“Oh, okay. Maybe I typed it wrong. Um, it doesn’t like me”, laughed a participant (iOS) attempting to sign in at American Eagle (first image). She proceeded to reset the password, but was unable to verify the new password because it was also only displayed as masked (second image). “That was a little bit of a pain”, she remarked negatively after resetting the password.

Next, account sign in is viewed by many users as simply a hurdle to overcome to proceed with more productive tasks, such as checking out, tracking orders, managing account details, or initiating returns.

However, when account sign in is problematic, users’ frustration will mount, as they haven’t even reached the account feature they’re interested in yet.

2) Order Management and Tracking

Order management and tracking continue to be top priorities for users.

.

Whether making a one-time purchase or subscribing to recurring shipments (e.g., every 3 weeks, every 4 months), recent testing reveals that users’ expectations regarding self-service order management and tracking features and information have only intensified since 2018.

“I often go to my previously ordered items to buy them again, rather than searching the site to try to find them. If I want to make sure I get that same thing again”, explained a participant at Bloomingdale’s as she scrolled the “Order List” page looking for a previous order containing a whisk. However, she couldn’t easily locate the desired order because only the order number and order date were visible by default (first image). Consequently, she opened 4 orders before stumbling across the whisk order (second image). Hiding basic item and order details in collapsed sections performed as poorly in testing as not providing these details at all (see guidelines #3218 and #925). Indeed, this design places the burden on users to interact with the interface simply to discover the contents of each order.

The “Order List” and “Order Details” pages serve as critical starting points for users when they are looking for specific order-related information, and when they want to modify or cancel an order, start a return, review a purchased item, or revisit a purchased item to buy it again.

Across multiple large-scale studies, we’ve observed that users need easy access to certain basic information and actions to effectively identify and manage their orders.

For example, without a clear understanding of order contents and status, and adequate features to identify desired orders on the “Order List” page, users will be needlessly delayed in beginning their order-related tasks, which can result in unwanted additional orders and place more strain on customer service.

Likewise, when key details for items on the “Subscriptions List” page are missing or unclear, users will be unable to easily identify and manage recurring orders.

“Right now it’s in Opa Locka, which is here in Miami. So, it’s arrived there, but it doesn’t give me an estimated delivery date. So that’s a little disappointing. I would hope that that information would be there”, complained a participant at Bloomingdale’s when he discovered the “Order Tracking” overlay lacked an estimated delivery date (first image). “Okay, so here it’s saying, the expected delivery date is today, the 23rd, by 9:00 PM.” He searched the tracking number at the carrier site and confirmed the expected delivery date and time (second image). “Yeah, see, it would be helpful if the information from USPS about the expected delivery date would also be included. I’m just not seeing it anywhere. I’m not seeing it anywhere on the website”, he lamented after he returned to his tracking information at Bloomingdale’s.

When it comes to order tracking, for most users the key details needed to track an order are the estimated delivery date, the carrier name, and the subevents or stages of the order delivery process.

Some tracking details will be contained in the order shipment notification email, but users who need more detail on order tracking than is in the shipping notification email will visit the “Order Tracking” page on the ecommerce site.

Ideally, the destination page on the ecommerce site will contain all the tracking information users need, including photo proof of the delivery.

3) Order Returns

“Let’s see, ‘View Details’. That’s not very clear. I’m looking for where I can start my return. I don’t see it here — the return part. So let me see, I’ll click ‘Return Policy’.” This participant at Wayfair searched for information about returning an item in the order listing (first image) and in the “Order Details” overlay (second image). She then resorted to opening the return policy link in the footer (third image). Simply removing the “Start Return” button when the return window has closed, as was the case at Wayfair, created needless uncertainty for participants, and often resulted in wasted time and effort.

“Um, this is the tricky part — this is where Amazon gets you — so sneaky! They generally don’t give you the option to put it back on your credit card, unless you’re very — there’s not a bubble here, right? If you’re being very quick about it, you’re going to click ‘Amazon account’, blah, blah, and you’re going to click ‘Return’. They still have your money! I actually want my money back. I don’t want to add this to my gift card amount or whatever. So I think this is a bit sneaky on Amazon’s part.” This participant at Amazon, having been duped by the “Refund Method” interface in the past, carefully located and clicked the “Refund to Visa” link (first image), which revealed the radio button for her preferred refund option (second image).

While for many sites returns are an undesirable aspect of ecommerce, how sites perform during returns is critical for end users.

A high-performing returns experience can, perhaps counterintuitively, become the basis of a loyal customer relationship.

On the other hand, a returns experience fraught with confusion, friction, and unnecessary obstacles will cause some users to avoid shopping at a site in the future.

Yet during large-scale testing, participants frequently struggled to determine an order’s eligibility for returns and to confidently navigate the return flow.

Moreover, returns has perhaps undergone the largest transformation since 2018.

Indeed, in 2018 returning an item typically required finding a box, printing a paper return shipping label, repackaging the item, dropping it off with a carrier, and waiting several weeks for a refund.

“’Label-Free at the UPS Store, no printer needed’. I will do that option. I don’t have a printer, so that’s generally how I do returns”, explained a participant (iOS) at Bloomingdale’s, as she chose the “Label-Free” drop-off at the UPS store (see guidelines #3279 and #3411).

“Okay, it’s giving me all the options. I’m going to do the ‘schedule for pickup’ option. ‘It’s free, choose a time and place, and we’ll pick it up.’ I’ve got to pack the item, that’s fine.” This participant (Android) returning an item at Walmart opted for the free FedEx pickup method.

On the contrary, during recent testing, participants benefited from label-free — and sometimes box-free — return shipping methods, as well as “Return Pickup”.

Additionally, for some participants, refunds were initiated as soon as the returned item was scanned during drop-off or pickup.

Overall, users expect a much more responsive returns experience with clear, easy-to-locate return details, multiple return options (including at least one free one), and straightforward refund options and confirmation.

Support Key Customers with High-Performing Accounts UX

“It looks like it’s only giving me an option to delete it, not actually update it. Okay. Let me go back — maybe I missed something.” A participant (iOS) at Wayfair wanted to update the expiration date on a saved credit card (first image). “Nope. Looks like that might be the only way. Let me see if maybe there’s a different option on this. Nope. Well, it looks like I will have to delete the debit card and then reenter it.” When she tapped to select the card, she discovered she could only delete it (second image). She spent over 2 minutes searching the “My Account” section for a method to edit the expiration date; she ultimately resorted to deleting the card and re-adding it with the updated expiration date.

Accounts & Self-Service is a unique aspect of the ecommerce user experience.

Unlike the other ecommerce themes we’ve researched, Accounts & Self-Service is often not part of the “direct purchase funnel” — but rather comes post-purchase.

However, this isn’t to indicate that Accounts & Self-Service is less important than the other ecommerce themes.

Returning users — those most likely to use self-service features most often — are critically important to a successful ecommerce business.

This means the Accounts & Self-Service UX performance can be just as important as the UX performance of the Product Page, Homepage, Category Taxonomy & Main Navigation, or any other ecommerce areas when it comes to a site’s overall long-term sales.

It’s therefore key that users’ experience in the Accounts area of a site is as smooth as possible to support an ongoing site-user relationship.

This article presents the research findings from just a few of the 700+ UX guidelines in Baymard – get full access to learn how to create a “State of the Art” ecommerce user experience.

If you already have access through your account, go explore the updated Accounts & Self-Service UX research.

If you want to know how your desktop site, mobile site, or app performs and compares, then learn more about getting Baymard to conduct a UX Audit of your site or app.

Post this article on X

Related articles

• Use a Fake “Editing” Flow When Updating Credit Card Details (78% Don’t)
• Always Provide 6 Key Order-Tracking Details on the Ecommerce Site
• Accounts & Self-Service UX 2025: 5 Common Pitfalls & Best Practices
• New 2024 Order Tracking & Returns UX Benchmark with 950+ Performance Scores and 850+ Best Practice Examples
• New 2023 Mobile Customer Accounts UX Benchmark with 1,600+ Performance Scores and 1,100+ Best Practice Examples

At Baymard, we’ve just released a new UX benchmark with 7 “Telco” UX case studies.

This follows from our large-scale user testing on Telco UX and adds to our existing ecommerce UX benchmark.

In this article, we give you a snapshot of the overall UX performance.

7 Telco UX Case Studies and the Overall Performance

These are the 7 in-depth Telco UX case studies.

The 7 Telco sites have been manually assessed across 500+ research-based UX parameters relevant to Telco providers, resulting in 3,100+ weighted UX performance scores and 2,400+ best practice examples from these desktop and mobile sites.

Each of the 3,100+ UX performance scores from the 7 Telco case studies is summarized in the interactive scatterplot below — showing you how they perform collectively and individually:

{{ scatterplot-graph: size=big + habitat=public + base-sites=collection:telco + view-structure-id=gemini-st_47eezi8l }}

The overall UX performance of the Telco sites is “poor,” with only 1 site in the “mediocre” range.

Here are some of the overarching UX issues.

On the benchmarked sites, proprietary plan names and technical specifications were presented as labels without explanation.

No site explained its industry-specific filters or provided inline help for technical specifications within comparison features, and plan features and specification labels were also often undefined.

Comparison was obstructed at every level, from plans to handsets to network coverage.

Contract length was missing from plan listings, comparison features were largely absent for handsets, every site listed specifications in inconsistent units and labels, and 4G and 5G coverage were displayed on separate maps.

The commercial terms of a financed device and ongoing service were routinely unstated.

Warranty upgrades were missing on all product pages in the benchmark, and returns, shipping cost, delivery timing, and the context around discounts were each missing from a substantial share.

As this benchmark shows, with no sites delivering even a “decent” or better overall user experience, there is plenty of room for improvement.

A publicly available overview of the research and benchmark can be found on our Telco UX research overview page.

Getting access: all 3,100+ UX performance scores, 2,400+ best practice examples, and the UX insights from researching the Telco ecommerce industry are available immediately and in full within Baymard. (If you already have access through an account, open the Telco UX Benchmark collection.)

If you want to know how your Telco desktop and mobile sites perform and compare, then learn more about getting Baymard to conduct a Telco UX audit of your site.

Post this article on X

Related articles

• New Telco UX Research: 3 High-Level Takeaways from 3,500 Hours of Testing
• 17 Common UX Pitfalls Telco Websites Suffer From
• Internet Service Providers UX Benchmark: 1,700+ Performance Scores and 1,100+ Best Practice Examples

Today at Baymard, we’re announcing the launch of our new Insurance UX research.

8 US and 12 UK Insurance sites were tested, including providers (some of whom are also banks) and aggregators (where test participants could compare Insurance quotes across many providers).

The 9 UK Insurance sites tested were Halifax, Aviva, Direct Line, LV/Liverpool Victoria, TSB, Insure and Go, Allianz, Avanti, and Santander, while the 3 aggregators were Compare the Market, MoneySuperMarket, and Go Compare.

The 8 US Insurance sites tested were Liberty Mutual, Travelers, Progressive, GEICO, Amica Mutual, Lemonade, Root Insurance, and Allstate.

This primary research is supplemented by our ecommerce UX benchmark, which includes AXA UK, Generali, “Progressive”, “Liberty Mutual”, “Travellers”, “Allstate”, and State Farm.

During testing, the participants encountered 1,700+ medium-to-severe usability issues as they attempted to use Insurance sites to explore options and eventually add coverage.

Our analysis of these issues has resulted in 190+ guidelines on Insurance industry UX, available in our research findings.

In this article, we’ll provide 3 high-level observations on where users on Insurance sites struggle:

• Understanding unfamiliar Insurance concepts and descriptions
• Progressing through the Insurance quote flow
• Establishing trust with Insurance brands

Users Struggle with Insurance Concepts

Shopping for insurance, while having some similarities to traditional ecommerce shopping, poses unique challenges to users.

A key differentiator of Insurance sites is that users seeking coverage are often asked for information that is more complex and less familiar than the address and payment details requested in general ecommerce.

One reason for this is that getting insurance cover (or coverage) is an occasional task that happens only very infrequently for auto or home insurance.

“’What is the current market value of your home.’ Oh blimey! Crikey!…Err…Ohh..I think it was somewhere around £575,000.” This participant at Money Supermarket (iOS) reacted with surprise and then hesitated at the home value question. She made her best guess and moved on, though she wasn’t confident of the estimate.

“Now, I wouldn’t know what the rebuild cost is. I’ve got no idea…I’d be a bit worried about putting my own figure in because I’ve got to be way out!” This participant at Go Compare struggled to enter an accurate estimate of the “Property’s Rebuild Cost”, and proceeded to skip the question. His response reflects the risk of asking users questions they do not know: poor answers, skipped answers, and even site abandonments. As the participant said, “I think what I’d probably do at this stage is just try and continue without filling it in, see if it insists on me entering something in there. In which case, I might have to think again about using somewhere else for a quote.”

”There’s so many different [roof options] here!… Like ‘Shingle’, I have no idea what that is.” This participant at Money Supermarket slowly scrolled through the list of options for the “What is your roof made of?” question, eventually answered “Tile” as the material of her roof, and moved on.

As a result, there is a heightened risk of users being unfamiliar with concepts or physical items that are a key part of the quote process, such as “5-level mortice deadlock” in a home insurance quote.

Furthermore, when faced with unknown values, participants would frequently hesitate for extended periods, bypass fields, or resort to unguided guesswork.

“’Do you need more…?’ I’ll put in the amount that they suggest I need.” After seeing the site’s recommendation, this participant at Direct Line confidently adopted the exact suggested figure. This default almost certainly gave her an answer more accurate than those she had provided on other sites, where she announced that she was guessing wildly, since she did not know what cover level to choose. On one such site, she estimated her contents at less than 15% of this value.

“Roof is 50% flat? I’ll just adjust it.” Here at Compare the Market (iOS), this participant said that images made the process “quick and easy”, and answered the flatness percentage of her roof, the material of her external walls, and the material of her roof in under 8 seconds. Further, the visuals created a measurable time savings for the participants who struggled to answer these questions on other sites.

Thus, when users are asked to describe a component or attribute of a house or a car, a clear explanation should be provided, along with visuals if they are needed to fully understand the option.

It’s also important to help users to estimate values (such as the value of their home or its contents) by providing tools that allow them to reach an accurate estimation and default values that reflect; for example, the average of similar homes in that area.

Similarly, helping users with estimating their car mileage by being flexible about the relevant time period (e.g., daily, annually) will make the process that bit less troublesome.

Often suggesting default values that reflect the typical value of a home, car, or insurable component is a good idea.

Default values are typically fine for most users, and they prevent users from making what are often widely inaccurate guesses about the value of some insurable component (e.g., a roof).

Additionally, visual aids in Insurance were observed to benefit users — for example, showing types of roofs rather than only listing them in text (e.g., “tile”, “slate”, “concrete”, etc.).

Providing users with the information they need to make informed inputs during the quote flow will help achieve a more accurate estimate, benefiting both the site and the user.

The Interaction UX of the Insurance Quote Flow Can Bring Users to a Halt

The insurance quote flow is the centerpiece of most users’ journeys on Insurance sites.

The quote flow is typically a form that includes a series of questions users must respond to to determine the coverage they should purchase (or even if they’re eligible at all for purchasing coverage).

“It just keeps taking me back to this.” This participant on Allstate saw that her name had incorrectly populated as “Hannah Hannah [Last Name]”, so she clicked on the browser’s “Back” button to correct it (first image). The quote page did not respond to her “Back” button click, and so she clicked the “Allstate” logo, which took her back to the homepage and restarted the quote process (second image).

During testing, participants were observed to stumble often during the flow, with some disruptions so severe they led directly to a participant abandoning the site.

For example, if a user has invested 20 minutes in filling out an insurance form — only to get stuck entering data into a particular field, experience trouble going back in the flow, or run into an error that clears some previously entered data — the user is likely to simply give up in frustration.

In short, unlike in general ecommerce, where users may be more forgiving of interaction UX issues or glitches in relatively short checkouts, issues experienced by users during the quote flow process quickly compound to the degree that users will abandon the process entirely.

Moreover, users rely heavily on navigation and being guided through the quote flow in such a way that they can understand each step of the process and feel oriented while progressing through it.

Thus, it’s important to support users within this flow.

“Oh, too late now…I guess I can go back. Can I go back? I can go back.” This participant at Liberty Mutual was able to move to the previous step using the site’s on-page “Back” button, and the button shows several of the qualities that allow for findability. It appears in parallel with the site’s “Save & Continue” button, it contrasts effectively with the content around it in size and color, and it was consistently placed at the bottom of Liberty Mutual’s quote flow pages.

Likewise, Liberty Mutual’s “Add” and “Edit” links were easy for participants to find at the right margin of each coverage type (first image). Note the large, prominent button and the inclusion of “Edit” or “Add”, which immediately signifies to users that this is where they click to change their coverage. Activating the links opened a modal (second image).

This would include, for example, ensuring the “Back” button works as they would expect, that they are allowed to edit any details they provide at the point that they supply that information, and that there is an updated cost summary visible at each stage.

Furthermore, to save users wasting time and effort if they are not eligible for the cover they need, they should be informed of this blocker at the earliest opportunity, so they don’t even bother entering a quote flow for an insurance product they ultimately can’t purchase.

Users Struggle to Trust Insurance Brands

“Wow! None of these companies I’ve ever heard of…I’ll tell you, I wouldn’t call any of these people…Here’s your bottom of the barrel…’National General’. The ugly stepchild [of Allstate], so to speak…It’s the same with this ‘Bristol [West]’ company, ‘Dairyland’…I don’t know who they are…” This participant on Travelers’ Insuramatch aggregator system (iOS) equated unfamiliarity of insurance companies with low quality.

“’227’ seems a bit high for a site that’s not in the forefront of my mind. I’m not too familiar with it. I don’t know how popular it is with others, but for me, it’s a step below the others, just because I don’t have an experience or a history with them. I want something that’s been established.” As this participant showed on Lemonade (iOS), users are often unlikely to pay premium prices for an unfamiliar company’s insurance plan. As the participant noted, unfamiliar insurers are automatically seen as “a step below the others”.

Finally, trust and transparency are especially important for insurance customers.

Insurance can be costly, and the repercussions of buying a policy that doesn’t provide the expected coverage could be severe in the event of having to make a claim.

“So, for me that would stand out quite quickly…If a company or a brand have a good Defaqto score, then that would give me more confidence to want to shop with them. Knowing they’re 5-star rated, and just being reviewed by Defacto. I think that means it’s quite a reputable company.” This participant read about Aviva’s 5-star rating from Defaqto, which was featured front and center on the homepage. This gave him confidence in the company as he went on to shop Aviva’s home insurance.

Thus, external validation for both the company and its products is crucial, especially since users are likely to consider several providers before settling on one and could be swayed by another provider that seems to be more trustworthy.

This participant’s Insure and Go quote journey was 20 minutes, and four of those were spent learning the cost of one add-on and returning to her place. The participant saw the “Hazardous Activities” add-on and was interested that, according to her, “They actually say it’s covered for free”. She selected the add-on (first image) and in the next step saw her updated cost (second image) and recognized, “It’s actually £115 now, whereas it was like £56, £58? Something like that”. She decided to remove it from her quote. Getting from the add-on to the updated cost took 54 seconds, and the participant then spent another three minutes backtracking through loading screens and review steps to the page where she could remove the add-on.

Trust must also be earned in the process of getting a quote.

Across general ecommerce sites, 40% of users abandoned checkout because of extra costs that surfaced late in the journey.

Unexpected price changes have the same effect on users seeking insurance cover, so every cost should be disclosed upfront before users actually choose it so that they aren’t forced to recalculate premiums, or even worse, don’t know why the price suddenly jumped as they proceed through the quote process (see guidelines #3108 and #3105).

Provide Insurance Users with a Solid UX Foundation

“I don’t want to create an account…I’m gonna get constantly hounded…Also having another account out there that can potentially get hacked. This one has my driver’s license number, it has more proprietary information.” Some participants who abandoned Root referenced their concerns about data security as a reason for leaving the site when they were forced to “Create an account to save your progress”.

Insurance sites, perhaps uniquely in ecommerce, may have one shot to get it right with users.

Users are less likely to return to an Insurance site where they’ve struggled to achieve their goals.

And once users leave, if they find a competing insurance product, they’re more likely to stick with that product for longer, compared to general ecommerce, given the “buy it and forget it” mentality some users will adopt when shopping for insurance.

Therefore, it’s key that the Insurance site has high-performing UX that supports users navigating unfamiliar concepts, challenging quote flows, and brand-trust issues.

Our Insurance UX research topic contains the critical UX research that helps ensure your Insurance site is meeting users’ needs.

Users with access to the Insurance industry can start with the 41 “Essential” UX guidelines, before moving on to consider the other 150+ guidelines on improving Insurance UX.

Additionally, consider viewing our Insurance benchmark, which contains case studies on 7 top Insurance sites.

Getting access: all 190+ Insurance UX guidelines are available today within Baymard. (If you already have access through your account, open the Insurance study.)

If you want to know how your Insurance desktop and mobile site performs and compares, then learn more about getting Baymard to conduct an Insurance UX Audit of your site.

Post this article on X

Related articles

• New Insurance UX Benchmark with 1,700+ Performance Scores and 1,200+ Best Practice Examples

How much revenue is your site losing each month because of unnecessary UX friction?

If you are like most ecommerce leaders or UX managers, you know that friction points across checkout, product pages, and navigation cause shoppers to abandon.

Yet, when competing for roadmap resources or budget approvals, UX research is often categorized as an internal software or subscription expense rather than a top-line revenue driver.

Without concrete financial estimates to back up your design priorities, critical usability improvements often get pushed aside.

To help you translate UX friction into clear commercial impact, Baymard has developed an Opportunity Sizing Calculator.

This tool converts Baymard’s actionable user research into quantifiable projections, showing you the exact revenue you can unlock by systematically applying proven UX best practices to your site.

Moving from Opinions to Research-Backed Forecasting

Prioritizing a product roadmap often turns into a debate of subjective opinions, where the loudest voice in the room or the most senior person’s viewpoint takes priority.

UX designers and product managers must frequently advocate for critical usability fixes against competing internal requests, relying on qualitative arguments that struggle to secure executive support.

Baymard’s Opportunity Sizing Calculator bridges the gap between design priorities and executive decision-making, helping you secure buy-in.

This means you’re better able to:

• Connect specific usability insights directly to top-line revenue
• Build an unshakeable business case for leadership
• Give directors and VPs clear line-of-sight into the ROI of resolving site friction

Get your personalized opportunity metrics: Book a short call with our team and find out how much revenue you could recover with better UX.

Book a demo call now

What the Calculator Outputs Reveal

Once your inputs are applied, the calculator provides clear projections broken down by total impact and site area.

• Annual recoverable revenue range: You will see a range (low, midpoint, and high) of annualized dollar value currently suppressed by UX friction
• Breakdown by UX theme: You can pinpoint where the largest financial gains are (for example, in checkout design, mobile navigation, or product detail pages)
• Plan tier impact: The output illustrates how expanding your access to additional Baymard guidelines unlocks cumulative revenue potential across your roadmap

How the Opportunity Sizing Tool Calculates Recoverable Revenue

The calculator projects potential conversion lifts by applying parameters derived from 200,000+ hours of UX testing across 40+ ecommerce industries.

Business Performance Inputs

To generate a baseline projection, the calculator evaluates your site’s monthly traffic alongside one of two standard performance metrics:

• Option A: Revenue per Visitor (RPV)
• Option B: Average Order Value (AOV) combined with Conversion Rate (CVR)

The tool also takes into account your industry to anchor the calculations to relevant benchmark data.

Standardized Guideline Scoring

Every UX best practice guideline in Baymard’s catalog carries an estimated lift based on two core factors:

1. Frequency of encounter: The percentage of users who interact with the problematic UI component
2. Severity of friction: The conversion impact observed during user testing when a site violates the guideline

Quantifying the individual guideline allows us to quantify the opportunity size for each UX theme, and for each Baymard subscription plan.

That means we can help you decide on the best plan for your team and organization, and show the ROI you can expect from implementing best practices available within that plan.

Compounding Fixes with Diminishing Returns

Fixing two separate UX issues on a website does not produce simple additive results.

For example, resolving an issue in the cart improves the baseline performance of the site, and any subsequent fix in the checkout flow operates on that newly improved baseline.

The calculator applies a compounding model to reflect diminishing returns, ensuring that combined projections remain mathematically grounded rather than artificially stacked.

Applied Revenue Caps

Even a complete overhaul of every UX issue will not double a site’s conversion rate overnight. There is a practical ceiling to the impact of usability improvements alone.

To maintain realistic financial projections, the methodology applies standard cap models to total revenue lift, preventing the opportunity size from being overinflated:

• Conservative (low): 15% revenue lift cap
• Baseline (midpoint): 25% revenue lift cap
• Aggressive (high): 35% revenue lift cap

Find Out How Much Revenue Your Site Is Leaving Behind

Leaving standard UX best practices unaddressed directly suppresses your revenue performance.

Stop guessing what drives conversion, and get the dollar values to back up your design recommendations.

Book a short call with our team to run your metrics through the Opportunity Sizing Calculator and get a personalized report for your site.

Post this article on X

Related articles

• Time is Money: How Pre-Validated UX Research Protects Your Resources & Budget
• Generalizability of UX E-Commerce Research
• UX-Ray for Figma: Align Your Designs with Research-Backed UX Best Practices
• AI Heuristic UX Evaluations with a 95% Accuracy Rate (Human-Level Accuracy)
• 2026 Progress: Instant UX Analysis of Your Site Now Across 209 UX Parameters (Still 95% Accuracy Rate)
• UX-Ray 2.0: Research-Backed UX To-Do List in Minutes
• UX-Query - Get Instant Answers to Any UX Question
• Product Page UX 2026: 10 Pitfalls and Best Practices
• Accounts & Self-Service UX 2025: 5 Common Pitfalls & Best Practices
• Checkout UX 2025: 10 Pitfalls and Best Practices
• Homepage and Category Navigation UX 2026: 10 Pitfalls and Best Practices (64% of Sites Have Mediocre-to-Poor Performance)
• Product List UX 2025: 8 Common Pitfalls & Best Practices (80% Have Serious Issues)
• Mobile UX Trends 2026: 10 Best Practices
• Desktop UX Trends: 10 Common Pitfalls & Best Practices

At Baymard, we’ve just released a new UX benchmark with 8 “Food Delivery & Takeout” UX case studies.

This follows our large-scale user testing on Food Delivery & Takeout UX and adds to our existing ecommerce UX benchmark.

In this article, we give you a snapshot of the overall UX performance.

8 Food Delivery & Takeout UX Case Studies and the Overall Performance

These are the 8 in-depth Food Delivery & Takeout UX case studies.

The 8 mobile sites and apps have been manually assessed across 390+ research-based UX parameters relevant to Food Delivery & Takeout, resulting in 3,100+ weighted UX performance scores and 1,900+ best practice examples from these mobile sites and apps.

Each of the 3,100+ UX performance scores from the 8 case studies is summarized in the interactive scatterplot below — showing you how they perform collectively and individually:

{{ scatterplot-graph: size=big + habitat=public + base-sites=collection:online-food-delivery + view-structure-id=gemini-st_l9jw5tjp }}

The overall UX performance of the Food Delivery & Takeout sites and apps ranges from “poor” to “mediocre”.

What is missing is almost everything specific to this industry, rather than borrowed from general retail.

Repeat ordering was undersupported despite its importance within the industry.

With no route to past orders from the homepage, and order histories too sparse to identify them by, one of the most common things a customer does here was also one of the least accommodated.

Dietary and customization needs had no supported path at any decision point.

Category-specific filters were absent throughout and customization options went undescribed on the item page, leaving allergy and restriction constraints needlessly difficult to resolve.

Orientation and return paths failed at every step of the ordering flow.

Scope markers, breadcrumbs, and back-paths were missing throughout, so going one level deeper meant losing any reliable way back out.

Furthermore, fulfillment timing and tipping were presented without the detail needed to commit.

For an ecommerce industry built on habitual reordering from a phone, often under time pressure and dietary constraint, these are not refinements but the substance of the task.

As this benchmark shows, with no sites delivering even a “decent” or better overall user experience, there is plenty of room for improvement.

A publicly available overview of the research and benchmark can be found on our Food Delivery & Takeout UX research overview page.

Getting access: all 3,100+ UX performance scores, 1,900+ best practice examples, and the UX insights from researching the Food Delivery & Takeout ecommerce industry are available immediately and in full within Baymard. (If you already have access through an account, open the Food Delivery & Takeout.)

If you want to know how your food delivery or takeout mobile site and app performs and compare, then learn more about getting Baymard to conduct a Food Delivery & Takeout UX audit of your site.

Post this article on X

Related articles

• 3 High-Level UX Takeaways from 1100+ Hours of Testing Leading Food Delivery and Takeout Sites
• Grocery and Food Delivery Site UX: Allow Users to Add “Past Purchases” to the Cart from the Homepage