Cross Rhythms Direct - Usability Report

Gervase Markham - 11th May 2003


This document is a series of comments on the general usability of the Cross Rhythms Direct (CRD) web shop. It concentrates mostly on the CD side of the business, because that's where the greatest potential for growth is, but it also has some things to say about the integration of the different types of merchandise. I've also taken the opportunity to record thoughts that came to me which were not strictly usability-focussed.

This document is not primarily focussed on the look and aesthetics of the site. The referenced mockups are for illustrative purposes only and, while indicating the sort of simplicity and usability that is the aim of the exercise, should not be taken to mean that, for example, highlighted CDs should have a red border, or the search box absolutely has to be in the top middle of the page.

Why Cross Rhythms?

The first question to ask is: why would anyone ever buy CDs from Cross Rhythms as opposed to any other CD shop, online or real-world? What does CRD have that the others do not?

You can't compete on price, and probably wouldn't want to given that part of the point is to fund CR. You can't compete on breadth of catalogue at the moment – and probably never will be able to, given that your catalogue is probably defined by that of your supplier, CPO. You could encourage people to buy from CRD because it supports the ministry, but sadly only a limited market care enough for that reason alone to be sufficient.

However, I do think Cross Rhythms Direct's has a number of unique features which could make it a compelling site from which to purchase CDs. They are:

1. Honest Reviews

CRD's most important hook, in my view, is its reviews - a large databank of honest data about the quality of each CD. On any other site, because the focus is on selling you stuff, every CD is generously reviewed (see Juice for the magazine equivalent of this.) CR can attract and keep punters because they know what they are getting when they buy something.

So, the reviews and the ratings need to be prominent in the interface. Changing the squares into a gold and silver stars is therefore a retrograde step, because it minimises this distinctive. Cross Rhythms has never been ashamed to say when something's mediocre, and should not be afraid to do so even in the shop interface. People respect honesty. Taking this idea further, a “Hall of Fame” or “Our Reviewers Recommend” highlighting all the (available) CDs to have ever got a 9 or 10-square review, organised by genre, would be useful e.g. for someone buying a present.

2. Independent Catalogue

The second distinctive CR has, or could have, is the independent CDs in its catalogue. Providing facilities for unsigned and up-and-coming bands to distribute their music both helps them in their ministry and provides a unique visitation reason for CRD. On the Net, there is no reason why bands should need a record deal to appear in web stores; CR should make it as easy, as automatic and as attractive as possible for bands to sign up to distribute their stuff through CRD. The way to play this is to have both independent and mainstream CCM CDs look exactly the same in the interface (“Wow! My CD's there alongside The Newsboys!”), and to encourage the independent bands to link to CRD from their websites as the authoritative web purchasing source.

3. The Experience CD

A third distinctive is the Experience CD – which is a great marketing tool for CRD. The track listing on the back of the CD should have the CRD ID numbers on it (and, on another usability note, should use colour coding to distinguish tracks from other content, rather than in the current useless alternating way.) The listings from the current and past CDs with details of each track should be available in the interface, to make it as easy as possible for users to buy the music they've heard. One idea might be, if licensing permits, to make copies of just the CD from each issue available for free or a modest fee – people taking them up would be more likely both to buy music from CRD, and to subscribe to the mag.

4. Usability

The fourth way in which is could stand out, of course, is that it could have the easiest-to-use interface that made it a pleasure to browse through and buy from :-)

Current Site Usability

1. Reviews vs. CDs

The first usability issue with the site is the false dichotomy between the “Reviews” part of the site, branded in blue, and the “Shop” part, branded in red. Immediately on visiting CRD at (archived copy), one is asked to choose between the two sides, and the split is present throughout the site - everything seems duplicated. There are

  • two pages for every CD (1), (2)
  • two search forms (1), (2)
  • two formats of search results (1), (2)

The two halves are different enough (different formats, icons etc.) that it's jarring to move from one to the other. The two halves are crosslinked at random spots, but the effect is to keep the user unsure of quite where they are.

Currently, on the main URL, the user is presented with two options, depending on whether they “know what they want” or not. The two options take them to either the red or blue sections - but the two options are basically the same from a user's point of view. The underlying idea of having a task-oriented interface is a good one, but the point of divergence for the two tasks is too early.

This can be seen because clicking either link presents the user with a choice of searching or browsing - it's just that in one case, they are searching CDs, and in the other, they are searching reviews. There's no real difference here – in a more usable interface, the user would have the option of searching for an artist or CD they know the name of, getting a list of search results, and only then have to choose between buying the listed items immediately or reading reviews.

Recommendation: The site should be unified - there should be a single search, with an option to search either “CDs available to buy” or “all reviews”, a single format of results, and a single page for each CD.

Currently, an option is available to browse alphabetically by Artist or Title. However, users do not arrive at the shop and think "Hmm. I'm going to buy a CD whose title begins with the letter T today." If they know the name of the artist in question, they'll type their name into the search box.

Recommendation: Permit instead browsing by genre, sorted by number of squares. CRD should not be afraid to plug better music more prominently – selling more good stuff means happier customers, and happier customers come back.

2. Retail/Independent

The second major issue is that CRD has two separate parts - the independent CDs and the CDs from artists signed to major labels. This produces an unfortunate usability hurdle to cross because CDs can be sent out in one of two ways, which currently require different payment and checkout mechanisms. This is harder to solve, because it involves elements not under your control. The best solution would be to eliminate the distinction and have a single point of payment – this could be done by having a retail partner which agreed to take independent orders as any other CD order but then pass the details back to Cross Rhythms to pass them on to the artist. However, if this is not possible, then the differences between the two routes should be minimised.

Recommendation: the shop should treat all purchases as a single basket up until the moment of checkout. At the point of clicking "Checkout", if the user has selected all items from a single shop, the user would be taken to the checkout for that shop. Only if the user had selected items from both shops would something else happen. In this case, the user would go to an intermediate screen, where all the information common to both checkouts would be entered. Then, the user would be led through both checkouts in turn, with as many fields as possible pre-filled from the first screen. This way, the inconvenience is kept to a minimum. The two checkouts would look and feel as identical as possible.

In implementation terms, this would probably be done by maintaining the Cart on the Cross Rhythms site throughout, and then transferring the item numbers and information for the Retail CDs in one big chunk to your credit-card-taking partner site just before you take the user to their checkout.

3. Navigation

Websites, unlike e.g. books or catalogues, have no physicality to give you a spatial frame of reference, and so navigation has to be carefully designed so the user never feels lost, and can always get where they want to go from where they are.

Navigation should also be omnipresent and consistent. Consistency makes the user feel like they know what's going on, and means that when they want something, it's always in the same place.

The current site has very little navigation; you can view (one of your) carts, and check out, but otherwise each page doesn't fit itself into a user's mental model of how the site is constructed. It's also inconsistent – it's sometimes the sidebar, sometimes text links, sometimes little green buttons.

Recommendation: I have prepared a mockup of what the front page of the CD shop might look like.

This obviously doesn't contain all the elements that the real thing might contain, particularly in the main content area, but it embodies many of the recommendations in this report, particularly those on navigation, and strives for the sort of easy simplicity that makes a site usable and understandable.

The new front page design has the following features:

  • Less space wasted on logos; space more efficiently used
  • Simple and uncluttered visual design
  • Consistent and omnipresent tabbed interface for switching between departments.
  • Prominent search interface, with single search box (see 4.)
  • Breakdown into categories by genre rather than alphabetically
  • The Experience CD track listings available
  • Hall of Fame available for recommending particular items
  • Single cart (see 2.)
  • More informative pullouts (see 6.)

A page like the mockup above should be what you see when visiting . The page currently at that URL wastes a great deal of space; it uses one third of the screen to display a logo, and only has two links “above the fold”. The one at is a little better, but it doesn't seem to be linked from anywhere.

4. Search

The heart of any web shop is its search interface. In an ideal web shop, every page should have a single search box in a consistent location, which should Just Work - that is to say, the ideal search interface consists of a single text box and a single button which, when pressed, produces exactly the result you want. If it's not as easy to search as Google, which usually gives you the right result from all the pages on the web using only a single textbox and button, then it needs to be made easier.

The ideal presented above may not be totally achievable in the real world; it is acceptable to have one or or perhaps two other widgets, as long as their default is the most sensible option. For example, on the mockup, to deal with the problem that CRD does not sell all the CDs which CR has reviewed (which, in an ideal world, it would do), we have a dropdown to choose whether the search should be restricted to CDs available to buy. The default for this dropdown would be different depending on which part of the site the search box was situated.

Recommendation: there should be a single search box, which should search for the given text in both artist and album (or, if all digits, as a reference number). If there are hits in multiple categories, it should present them all.

Additionally, it shouldn't place any restrictions on how the user formats the search - so a search for "Sara Groves" or “SARA GROVES” should produce exactly the same results as a search for "Groves, Sara". Removing the current restrictions means that you don't need three paragraphs explaining how the Search function works. Users don't read instructions anyway.

If you search on a number which is a CD ID number, there would be only a single result, and you should be taken straight to the page for that CD, rather than to the search results page.

5. Search Results

The current search results (the ones you get from the shop search rather than the review search) are not too bad but, as discussed above, they minimise the CR distinctive of reviews and ratings, and also present artist names in a “computer” format rather than a human one.

Recommendation: I've done a redesigned version of the search results (without the surrounding navigation, which is mocked up at the other URL).

The new search results design has the following features:

  • Removal of unnecessary info on which shop the CD is available from
  • Search results are across both artist and title
  • Lists artists' names as people expect them
  • Gives a short extract from the review
  • Gives the full rating information (see “Why Cross Rhythms?”)

6. Pullouts

The current site has the web equivalent of pullout quotes in a magazine – boxes highlighting particular CDs. These are a good idea. However, each CD should have its own potted synopsis which is not just the first sentence and a half of the review. There's no way you can automatically summarise a review using a computer; but it's not a hard task for a human. It may take a little while, but it'll sell each CD so much better if the two sentences which summarise it are a proper summary. E.g. for Sara Groves, “Past The Wishing”:

"Catchy pop one moment, gentle piano-driven balladry the next... a good listening experience."

rather than:

"After the success of Sara's last album 'Conversations', this is somewhat of a retrospective in that this 2000 US release contains nine songs that Sara wrote between 1990 and 1997."

Recommendation: See the mockup; it gives some examples of improved pullouts. Note that they don't have a “Buy” button – this question is open for debate, but I think not having one fits with the ethos of the user making an informed decision about each purchase – the usage scenario would involve them clicking the “more...” link to go to the page for that CD, and then clicking Buy from there.

7. Checkouts

These comments relate to the CPO checkout, not the independent one (which I haven't used), but would also apply to the combined checkout suggested above. Apologies for the lack of a mockup of a better interface; there wasn't time. This analysis is not exhaustive.

Basket: I came back to my basket after a week or so, and the contents were still there (good.) However, it only gave the title of the album and not the artist. “The Life and Times of Absolute Truth”? Who's that buy? Why did I add it?

Recommendation: add the artist to the album name in the “Title” field. Make the title a link back to the page for that CD.

It's very rare that a customer will want more than one of any CD. Therefore, an interface which assumes that buying 27 is as likely as buying 1 will be unnecessarily complicated.

Recommendation: dispense with “Quantity”, “Unit Price” and “Subtotal”, and just have:

Quench: Afterglow £13.99 Another One | Remove
Andy Hunter: Exodus £12.99 Another One | Remove

Then, you can ditch the “Update” and “Reset” buttons (no text boxes any more), and just have “Empty” and “Checkout”, and the whole interface gets a lot simpler.

Login: It asks for a username. That makes the user think “what on earth did I choose for my username?” What it actually wants, and should say it wants, is an email address. For most users, who only have one email address, that removes all uncertainty.

Checkout: The CD names are now linked – but the links take you to a complicated and seemingly pointless screen about that CD, which has links like “catalog” which take you to the CPO catalog, and not the CRD one. Disaster! The user shouldn't be able to wander off like this into the CPO site and get horribly confused. As far as they are concerned, they are buying from CRD, not CPO.

Recommendation: if CRD is to continue partnering with CPO, it needs to obtain more control over the user experience its customers have – to style the site more like its own, and to prevent them being able to wander off into the bowels of CPO.


Moving further away from usability, the current CRD shop has two major deficiencies which need to be rectified or minimised.

1. Small Catalogue Size

The small size of the catalogue could be rectified, in the short term, by linking through to for CDs you don't sell. You aren't going to get the sale, so you might as well pick up the referral commission. It also makes CR more of an authoritative source for CDs. If someone comes to buy something and finds they can't, they are less likely to go via CRD in the future. Conversely, even if they buy it from another merchant, they will hopefully still have appreciated the CR reviews and will therefore use CR as an access portal for their next purchase, which might be something you do stock.

Hopefully, as the catalogue expands, you can replace Wesley Owen links with CRD/CPO links transparently.

2. Lack of Music Samples

Being able to listen before purchasing is one of the key distinctives of a web shop as opposed to a bricks-and-mortar one. This is one of the reasons why “listening posts” have become so much more common in record shops recently. Therefore, a web store needs this in order to compete.

However, it's quite possible that this is an absolute nightmare to negotiate, organise and administer. On the other hand, you could have the mechanism already in place for the magazine side of things – I don't know. It's arguable whether 30-second excerpts could fall under the definition of “fair use for review” under copyright law. Maybe you could consult a lawyer on that point. But, otherwise, the lack of samples could be rectified by linking to bands' websites, or deep-linking direct to their samples, or offering Cross Rhythms Experience CDs by post.


CRD has the potential to become an authoritative source for Christian music CDs in the UK. However, to attain that position, it needs to leverage its unique selling points and take steps to minimise its weaknesses.

Time constraints have meant that this report is briefer, and the mockups less polished, than I would ideally like. It also contains what appear to be unsupported assertions (e..g “users don't read instructions”) which are actually usability principles well supported by research. Therefore, I would be happy to be contacted to discuss anything in this report further, and elaborate on my ideas and suggestions.

My hope is that this report will indirectly promote the growth of the Cross Rhythms ministry by helping CRD to become more successful and profitable.

Gervase Markham

Original URL: