Published on August 21, 2006
Every industry has its own version of the competitive analysis and its function is clear: to line up your product with other products and show where yours falls short and where yours is superior. Each industry brings a different spin to this old favorite and user experience design has its own set of criteria by which to judge competitors.
Take the simple competitive analysis shown in Table 5.1.
Simple Competitive Analysis
|Products||No specific products, links to product categories and specials||Home page features six products|
|Search box||Upper-left-hand corner, adjacent to primary navigation||Upper-right-hand corner, between primary navigation and account navigation|
|Navigation||Primary navigation organized by pet. Additional navigation for retail services||Primary navigation organized by pet. Additional navigation for account management|
|Contact information||Linked from left-hand navigation||Linked from top navigation|
|Shopping functions||Links to shopping cart, account management, and order status||Links to shopping cart and account management|
This tiny competitive analysis looks at the home-page contents of two prominent pet-related web sites. The table shows how each site’s home page stacks up in different categories. Despite the simplicity, this table includes the two main pieces of a competitive analysis: the competitors and the criteria to compare them by.
Competitive Analyses at a Glance
Because competitive analyses vary along only two dimensions—competitors and criteria—you’ll always see some mechanism for showing two or more sites side-by-side with the differences highlighted. The specific nature of those differences will vary depending on the criteria selected. At the same time, these documents can also vary by quantity—some are larger than others because they show more contexts or more competitors.
Figures 5.1 and 5.2 show two sample competitive analyses. The first is as simple as can be: a spreadsheet with the competitors along the top and the criteria down the side. The second is more complex, showing different kinds of comparisons for each competitor.
This simple spreadsheet can be the starting point for any competitive research. Whether it remains the ending point as well is up to you. Though it captures all the information adequately, it may not be appropriate for presenting results to stakeholders or other team members.
This page is from a more complex competitive analysis, where the research explored many different aspects of each site. Instead of presenting the data in one consolidated view, as in Figure 5.1, this report spreads the information over several pages.
Like the previous illustration, this shows a page from a report. This report, however, is organized differently, dedicating each page to a different aspect of the analysis, and illustrating this criterion for each competitor. The display of the data in this example is much like the table in Figure 5.1, with a couple of key differences. First, the data is displayed as a series of pictures—a technique called small multiples, which we’ll discuss further later in the chapter. Second, since this is a page from a larger report, it focuses on one issue, rather than all the issues.
In the entire process of creating a competitive analysis, preparing the document is the easy part. You’ll have some tough decisions to make about the document—for example, whether to present a simple table or something more elaborate—but ultimately, this is not what makes the process challenging. Once you’ve established the criteria and the range of competitors, gathering the data is also fairly straightforward, albeit time-consuming. Drawing a box around your analysis, however, and establishing boundaries to define what’s relevant to your project, is the more difficult task.
This will be a lot easier if you identify your purpose before you begin your analysis. A purpose statement can drive not only the types of information you collect about each competitor, but also how you present the data. That purpose may be as simple as, "We’re struggling with widget X in our site and we want to see how it is done on twenty prominent web sites." Or, it may be as complex as, "We’re building a system to support user group Y and we want to find out how this group has been supported elsewhere." If you’ve decided to do a competitive analysis, spend some time with the team brainstorming about what you want to get out of it. Articulate a purpose statement and compare notes with the rest of the team to make sure they have the same understanding of the purpose. In the end, you’ll define an agenda for the research and set expectations about how the information will be useful on the project.
Creating Competitive Analyses
The essential elements of a competitive analysis, described in layer one, are the purpose statement, the competitive framework (defined by the competitors and the criteria), and the data itself. You might find that your data is too extensive to fit into one table. Layer two describes the challenges of scaling the document from a simple table to a more in-depth analysis. Finally, to flesh out the document further, a third layer of information can provide more details about the overall project, the competitors themselves, or the method behind the analysis.
Layer 1: The Basics of Competition
Though it may be worthwhile to provide lots of detail about how you approached the competitive analysis or the rationale for the lineup of competitors, the brass tacks of the document are much simpler. To boil a competitive analysis down to its essentials, you’d find only the objective—the purpose of the analysis—and the data—the comparison. Of course, the data needs some kind of backdrop to make it meaningful; this is where the competitive framework comes in.
The competitive framework
Even the simplest competitive analysis displays two critical dimensions: the competitors and the criteria, or what we’ll call the competitive framework. The purpose of the competitive framework is to present the data in a way that makes it easy to compare the various sites across the different criteria.
When the competitive framework takes the form of a table, like the ones on the next page, the competitors run along the top of the table and the criteria along the side. The criteria can vary from the very general to the very specific. The first row of Table 5.2 offers a general comparison between navigation systems. Table 5.3 offers more specific comparisons.
Comparing Pet Web Sites with General Criteria
|Home Page Navigation||Primary navigation is different pet groups. Offers secondary navigation around account management and retail locations.||Primary navigation is different pet groups. Secondary navigation groups include additional pet product categories (like new items or sale items) and account management.|
Comparing Pet Web Sites with More Specific Criteria
|Additional item categories||None||Sale items, new items, clearance items|
|Account management links||My account, shopping cart, customer service||My account, sign in, checkout, shopping cart, shopping list|
|Offline shopping links||Store locator, in-store services||Catalog quick order, request a catalog, toll-free phone number|
|Other pet services links||Pet care guides, pet-related articles||Pet care newsletter, information center, pet pharmacy|
A different kind of competitive framework is known in MBA circles as the two-by-two. No, this isn’t the Noah’s Ark approach to comparing web sites. Instead, it plots competitors on a simple grid depicting only two criteria.
That right there is about twelve credit hours of most MBA programs. Notice that with a two-by-two, the number of criteria is shrunken down to two, so they tend to be broader. If you think these plots are based on actual numeric data, the graph has done an excellent job in turning subjective information (whether one pet-related web site is more specialized than another) into objective information. (In most MBA programs, this part is extra credit.) You could use real numbers and actually plot along the scale, but two-by-two presentations are ideal for very broad criteria that might not lend themselves to hard numbers. This type of graphic is useful to identify holes in a landscape. Competitors clustered around certain areas of the two-by-two can indicate that there are opportunities for your site to fill those holes.
In this two-by-two, different pet-related web sites are plotted on a simple graph. In this case, the axes of the graph represent the scope of the content (commercial vs. advice/information) and specialization (number of pet types supported).
There’s one other kind of competitive framework that appears in comparisons of different user experiences: the small multiples. This term belongs to visualization guru Edward Tufte. In Tufte’s The Visual Display of Quantitative Information, he writes, "Small multiples represent the frames of a movie: a series of graphics, showing the same combination of variables, indexed by changes in another variable." More plainly put, small multiples are a series of graphics that allow the viewer to easily compare similar sets of information. In the case of interface design for the web, this approach is most effective for comparing page layouts.
These small multiples show the general layout of the product page from three different pet web sites. The dark-gray boxes show navigation, while the black boxes show product description information. Promotional information is represented by light gray, shopping chart information by diagonal stripes, and related product by vertical stripes. (In color, you’d have more options.) Finally, the white boxes show related noncommercial content. From these small multiples, it’s easy to see that LucytheWonderDog.com uses a very simple layout and DrsFosterSmith.com prioritizes its advice content.
Data is where the rubber meets the road in a competitive analysis. The data can be as simple as yes-no values, indicating whether a site meets a particular criterion, or it can be descriptive, going into some detail for each criterion.
You’ve seen these kinds of competitive analyses on infomercials where the product in question is lined up with "other leading brands." For each feature, the product gets a check mark while its competitors get an X, to show you how versatile the product is. When it comes to web sites, the straight yes-or-no comparison is most effective for considering features, in other words, whether a set of web sites has a specific feature or not. In such a comparison, however, the subtle differences between the competitors may be lost.
Comparing Web Sites with Simple Yes-No Values
|Products on home page||Yes||Yes||Yes|
|Retail store locator||Yes||No||No|
|Expedited shipping options||Yes||Yes||Yes|
Some competitive analyses score the competitors in different criteria. You’ll see this approach in restaurant reviews where every place is scored on the quality of its food, the ambiance, the service, and the expense. For web sites, scores help give a little more substance to the comparison, though it may be difficult to generate the data. In this table, the sites are scored on a scale of one to five, where five indicates the site does a good job in the category, and one not so much.
Using Scores to Compare Web Sites
|Promoting products on home page||4||4||5|
|Alternate modes to find products||2||3||4|
|Promoting noncommercial content||3||1||4|
Notice that in this approach you’d have to define what it means to do well in each category.
Used more frequently than yes-no data or scores, descriptions specify how the competitors meet each criterion. Descriptions allow you to be more explicit about how the competitors stack up against each other, without resorting to potentially skewed numbers.
The two other formats for a competitive framework—the two-by-two and the small multiples—represent a different kind of data entirely. With plots on a graph or thumbnail images depicting screen layouts, the value is in the comparison. This isn’t to say that data presented in a table isn’t worthwhile as a comparison, but it can stand on its own. A plot on a graph, however, is meaningless unless it is lined up with other plots. The same is true for sketches of page layouts. Showing how much screen real estate is dedicated to navigation and content is pointless unless the reader has a basis by which to judge the amount of space. With two sketches next to each other, the reader can easily compare how one site uses its screen real estate with another.
Providing Further Descriptions for Comparisons
|Product catalog||Though the home page does not show the depth of the catalog, clicking into each pet category reveals a list of about two dozen product types, specific to the pet.||Extensive catalog of all kinds of pet supplies, somewhat more obscure than PetSmart.com. Clicking into a category reveals many different product types for each pet. There's a separate category for pet pharmacy, which is subcategorized by ailment.||The catalog is limited to dogs, cats, and horses, and focuses almost exclusively on medication. Nonpharmaceutical items are categorized under "accessories" for each type of pet.|
|Navigation||The catalog links on the home page are limited to pet types. Other links take users to store information and account information.||The catalog links on the home page are limited to pet types and sale categories.||Primary navigation is through a long list of ailments and needs, like "heartworm" or "grooming,"|
|Noncommercial content||Lots of noncommercial content but hidden behind two small links on the home page. This content does not offer any links into the catalog.||Extensive noncommercial content, with some linking into product catalog. The pet care articles reference each other, avoiding dead-ends in navigation.||Noncommercial content located in "Ask the Vet" section. Site contains extensive frequently asked questions, categorized by product category (though not linked to the products!) and a link to a separate pet care site.|
No matter how small your analysis, you’ll need to document the conclusions. Even if you’re doing a simple count of how many hits "log in" and "log on" get on Google, the data can’t stand on its own. You need to interpret the data in the context of your client and your project.
By putting your conclusion into words, you are establishing a partial direction for the design. The design team will take its cues for the design from these conclusions. When embarking on building a pet-related site, for example, the design team may seek out best practices from the competition. A cursory study of the landscape allows them to conclude that most pet-related sites use pet type (dog, cat, etc.) as the main navigation. A more in-depth study, however, leads to more detailed data—how sites specializing in one kind of pet categorize their content, how frequently sites use the same categories, the order of the categories (cats first or dogs first), how sites deal with uncommon pets (turtles). With data like this, the conclusions—and therefore design direction—can be better informed.
Layer 2: Tougher Competition
Competing in multiple events
In many cases, the simple bare-bones competitive analysis will suffice, but there will be some cases where your comparisons need to be deeper or broader: Perhaps the number of criteria makes a simple table unwieldy or you’re comparing your site to different sets of sites. For example, imagine you’re comparing the pet-related web site to other pet sites for navigation and design, but to major commercial sites for shopping cart and checkout functionality. As the scale of the analysis increases, you need a way to accommodate additional data in the document. There are two typical approaches: additional sections by criteria, and additional sections for each competitor.
Organizing report by criteria:
Organizing your competitive analysis by criteria creates a collection of smaller competitions in different events, so to speak. You might compare a handful of sites across five different groups of criteria: home page design, interior page design, search functionality, features offered, and navigation. Figure 5.3 shows the results of page layout comparisons across three fictitious pet web sites. For each of these categories, you’d create a separate competitive analysis but use the same competitors throughout. This approach allows you to identify the best players in specific categories and across all categories, but it can also create a disjointed picture of the user experience. In other words, because of the analysis, it is difficult to see the user experience of each competitor in its entirety.
Organizing report by competitor:
On the other hand, you could add data to your document by creating a separate section for each competitor. In this case, the competitive analysis reads as sort of a rogues’ gallery of sites, each one having a separate profile. Within each profile, the site is described with various criteria. This approach offers a holistic view of the user experience for each site, but it is more difficult to compare that site to others. Figure 5.2, at the beginning of this chapter, shows a page from a competitive analysis organized by competitor.
Regardless of how you organize your competitive analysis, with size comes a potential lack of focus. As you expand the reach of your analysis, either by adding more criteria or more competitors, you don’t want to lose sight of why you’re doing the analysis. Including a statement of purpose in the document itself is a great reminder, and also helps the document’s audience understand the context for the analysis.
Layer 3: Adding Further Detail
Competitive analyses with just elements from layer one usually stand well on their own, but you may want to add some flesh to the bones. Perhaps the most interesting information you can add is a description of your method for analyzing the competition.
Spelling out your process can help address any possible methodological inadequacies, especially for stakeholders who take these things seriously. What’s most worthwhile is rationalizing the selection of competitors and criteria.
The range of competitive web sites out there will vary, of course, depending on the site you’re building. The number may seem finite because your site is in a niche category and all the competitors are known players. The number may seem infinite because the web is vast and there are lots of sites competing for attention. You’ll have to narrow it down some way, and whatever way you choose makes for worthwhile content for your document.
At the same time, there are infinite criteria by which to compare sites. They can be as broad as the main navigation categories, or as narrow as the label on a button in a particular area of the site. You’ve made conscious decisions to include certain criteria—maybe they are a standard set used by your client or your company, maybe they were defined ahead of time by the stakeholders, or maybe you devised a special list just for this project. Whatever your methodology behind determining the criteria, this is excellent fodder for your competitive analysis.
Excerpted from Communicating Design: Developing Web Site Documentation for Design and Planning by Dan Brown. Copyright © 2006. Used with permission of Pearson Education, Inc. and New Riders.
You can learn more about the book at CommunicatingDesign.
Dan Brown recently wrote Communicating Design, published by New Riders, a book about documentation for web design. Dan is a founding partner at EightShapes, LCC, a user experience consultancy in Washington, DC. He plays mandolin poorly and devours comic books religiously.