Competitive Analysis, Part 2

In: Articles

By Dan Brown

Published on October 2, 2006

Excerpted from Chapter 5 of Communicating Design: Developing Web Site Documentation for Design and Planning by Dan Brown.

Analyzing the Competition: The Basics

Like any deliverable, a competitive analysis must start with a situation analysis: a hard look at the purpose, the timing, and the audience for the document. These aspects of your situation will drive the document’s contents and design.

The competitive drive

There are many reasons why you’d want to compare your site to others out there, but unless you make that purpose explicit in the competitive analysis, your data is meaningless for other people reading it. Even if you’re doing a competitive analysis that won’t see the light of day, establishing and documenting its purpose helps keep the competitive analysis in check and avoids analysis paralysis, too much thinking and not enough doing.

For the purpose, the best place to start is the motivation. Doing broader competitive analyses—comparing competitors across multiple contexts—usually comes from a need to get a sense of the landscape and industry best practices. The value here is knowing what the cost of entry is for a site like yours—knowing what the minimum expectations will be for users who are used to your competitors.

More narrow competitive analyses—those that look at one specific set of criteria—are usually motivated by a particular design problem. In these situations, you can treat the analysis almost as a science experiment, where the statement of purpose establishes a research question or a hypothesis. Labeling is a typical design problem that can be addressed with a quick competitive analysis. In this case, your purpose would be something like: “To identify what labels are used most frequently among our competitors for product navigation categories.”

A timeline for analysis

A competitive analysis is one of those funny documents that usually appears at the beginning of the overall process, but can easily appear at any time. Some methodologies call for a competitive analysis as a standard step at the beginning, whose purpose is simply to understand the competitive landscape before the design process begins.

If your team was tasked with building a new online pet store, you would likely look at existing online pet stores. The sites themselves vary: Some sites focus more on dogs and cats, or specialize in medications. Doing such a comprehensive analysis allows you to make some strategic decisions about how your site meets needs not already met—perhaps a site that specializes in pet reptiles or in plush toys for dogs. It may help you determine a competitive advantage, like having noncommercial content about pet health that can cross-sell items from the catalog.

An early competitive analysis can help set the stage, but it may also skew your other decisions at this stage—which features to include and which features will have the highest priority, for example. Though the competition can provide good information about these ideas, those decisions should also be driven by user research. A competitive analysis at the beginning of the design process may be incomplete. Until you do user research, you won’t know what features and issues are most salient.

Like concept models, which can serve as tools for understanding the problem at hand, a competitive analysis may never see the laminate top of a conference room table. It might only provide an overview of the landscape for you and your most trusted team members. Such informal analyses may be appropriate later in the project timeline, too, when the design is well under way and you have some specific issues in need of a benchmark. Of course, in doing a competitive analysis later in the process to clarify specific design issues, you can uncover other issues that would have been useful at the beginning of the project.

Suppose you’ve established a strategic direction for your site—specializing in plush toys for dogs—but now need to decide on a labeling scheme for the product categories. An informal analysis of the competition can give you a sense of what labels your audience may be used to. (Establishing whether your audience is the same as that of PETCO.com, for example, is a different exercise entirely.) Ultimately, the content of the analysis depends more on the need than where it appears in the process. Still, it’s safe to say that early competitive analyses tend to focus on strategic issues, while those late in the process generally try to shed light on specific design issues.

It’s about the fans

Conflict, tension, resolution: These are what draw spectators to a competition. Unfortunately, no web site competitive analysis will be as compelling as the Olympics or Tour de France. Still, there are people who do take a keen interest, and definitely want to know how the drama will end.

The audience for the competitive analysis will affect the presentation of its data. If your competitive analysis must be readable by people beyond the immediate team, you may need to create it as a self-contained package, requiring no specific knowledge to understand it. In this case, your document might need descriptions of the criteria, an explanation of why the analysis is being done, an overview of the conclusions, or all of the above.

On the other hand, a competitive analysis serving as an internal benchmark does not need to have as much context or description built in. If driven by the need to give the team an overview of the landscape, the document can forsake context for speed of delivery.

Prepping the data

The purpose of a competitive analysis is to show consistency between sites—thereby establishing industry practices (good, best, or otherwise)—or to show marked differences—identifying where some sites stand out. Inconsistencies can indicate that one particular site has developed an innovative solution to a common problem, or that the industry has not settled on a singular approach.

In the early days of e-commerce, for example, a consistent shopping cart model had not yet emerged. Eventually, the industry arrived at a de facto standard—nothing set in stone, but something that most users came to recognize.

Newer online features—such as the ability to add tags to content to make it easy to find later—are still trying to find their feet as far as design. Different sites do it differently, and showing these differences is the key to good competitive research.

Your data, therefore, needs to paint a clear picture and demonstrate these comparisons. When compiling it, you should think about visual mechanisms to show distinctions and similarities. With yes-no data and scores, this is pretty straightforward, but with descriptions, it becomes a little more complex. To add some depth to the information and make the data jump out a little more, you can assign keywords to the descriptions, or code them to indicate whether the site’s approach is good or bad.

Figure 5.6: Here are three rows from a competitive analysis. The descriptions are exactly the same throughout, but the presentation changes slightly: The first row just shows the raw data, while the second row adds a headline. This makes the competitive analysis easier to read. In the third row, the description includes a small icon to indicate a technique that would be appropriate for the author’s own project.

Site A	Site B	Site C
Content lives under two layers of categorization. Users must select a main category and then a sub-category before seeing a list of content.	Content is categorized with metadata and users can enter search terms to find content, or use a system of links to filter content.	To find content, users must enter relevant search terms. They can narrow down search results by clicking on metadata.
Navigation Categories Content lives under two layers of categorization. Users must select a main category and then a sub-category before seeing a list of content.	Metadata Filters Content is categorized with metadata and users can enter search terms to find content, or use a system of links to filter content.	Search and Filter To find content, users must enter relevant search terms. They can narrow down search results by clicking on metadata.
Navigation Categories Content lives under two layers of categorization. Users must select a main category and then a sub-category before seeing a list of content.	Metadata Filters Content is categorized with metadata and users can enter search terms to find content, or use a system of links to filter content.	Search and Filter To find content, users must enter relevant search terms. They can narrow down search results by clicking on metadata.

Tips for Effective Competition

A competitive analysis can generate a lot of raw data, and sifting through it can be daunting for team members and stakeholders who want to reach the bottom line. The best thing you can do for your competitive analysis is nail down your conclusions and spell them out early in the document. These conclusions will look different depending on whether you’re looking at a specific design problem or trying to get a sense of the overall competitive landscape.

For specific design problems

In the flurry of collecting data, you may get carried away with presenting all of it in an effective way. If you’ve set up the competitive analysis effectively and gathered a comprehensive set of data, you should have a problem statement that’s driving the analysis, and a response to the problem statement. Table 5.7 shows an example.

Table 5.7: Problem Statements and Responses

Good Problem Statement	Good Answer
Our site has a feature for allowing customers to add tags to our products. What interface elements do other sites use to support this feature?	Sites with tagging features overwhelmingly use a simple text field and ask users to separate tags with spaces. Since tagging is relatively new, nearly every competitor includes a “what is this?” link adjacent to the text field. Upon pressing Return (or Enter), the user gets immediate feedback from the interface that their tags have been added.

If your competitive analysis was motivated by several different design problems, be sure to have a conclusion for each one.

For landscapes and overviews

When you’re looking at the competition just to get a sense of what’s out there, you may not have a specific design problem in mind. This doesn’t mean you can’t come to conclusions, however. After nosing around the competition a bit, you should have a handful of key take-aways—essential messages that the team should keep in mind as they move forward with the design process. You might organize these take-aways by context, identifying one or two bullet points for each of the design areas you looked at.

Table 5.8: Key Take-Aways for Pet Store Web Sites

Category	Take-Away
Navigation	Pet type is the primary navigation, even on sites where the catalog is limited to certain products (such as medications).
Search	Search results tend to offer both product listings and category listings. There are no advanced search options.
Product Categories	Within each pet type, products are listed either by type (e.g., leashes) or need (e.g., training).
Checkout	Of the pet sites looked at, none offered any sort of special checkout.
Shipping	All shipping options available

With Every Competition Comes Risk

Table 5.9 shows the results of a competitive analysis. It lines up a conclusion (in the first column) with observations from the competitive sites. Without each other, the data and the conclusions are meaningless. Data provides support to the conclusions and conclusions give purpose to the data. In this example, the clients were wondering whether any pet supply sites have successfully moved away from using pet type as primary navigation, and whether product types (e.g., leashes) or needs (e.g., training) were most appropriate for product categories.

Table 5.9: Balancing Data and Conclusions in Competitive Analyses

Conclusion	PetSmart.com	PetMeds.com	LucytheWonderDog.com
Pet web sites always seem to use pet type as primary navigation, even on sites that offer a limited catalog. The exception to this rule is sites that specialize in one type of pet.	PetSmart’s navigation categories are as close to industry standard as they come, and their extensive catalog demands narrowing at this level.	PetMeds limits its catalog to medications for dogs, cats, and horses, but even this small group of pet types is used as the primary navigation. Since the catalog is smaller than a general store, however, PetMeds.com uses an additional layer of navigation on the home page.	Since LucytheWonderDog focuses exclusively on dogs, there are no pet type categories, and the primary catalog navigation includes product categories.
Even though it is semantically jarring, all sites freely mix product type with user need. Categories such as “training” and “flea and tick control” sit next to “dog bowls” and “dog beds.” Notably, sites tend to favor one type over another. So even though a site will use both product types and user needs as category, one type of category will appear more frequently.	Subcategories in PetSmart’s catalog tend to focus more on product types for conventional pets, though there are a few user-need categories. More unconventional pets such as reptiles, fish, and birds, however, tend to have more user-need categories. Overall product type categories dominate.	PetMeds is an interesting case study because two levels of navigation are spelled out on the home page, so it’s easy to see how they categorize their products. The highest level of navigation is based on need, such as “flea and tick control” or “grooming” or “bone and joint.” There are a couple of notable exceptions, such as “medications” and “accessories,” which serve more as miscellaneous categories than anything else. The second level of navigation consists mostly of product types.	Unlike PetSmart, Lucy’s categories for dogs include a few product types (such as “shampoo”) but mostly rely on user needs, such as “grooming” or “dental” or “joint care.”

This table demonstrates the balance between observations and conclusions—presumably, this table is a good balance. The risk in doing a competitive analysis is that you skew too much in one direction, missing the forest for the trees, or vice versa.

Start with your conclusions

If you have a lot of data—you spent a lot of time with many different competitors, looking at many different criteria for comparison—you might get caught up in trying to include all that information in your analysis, to the detriment of your conclusions. The solution here is methodological: Make sure you know your conclusions before starting to document your data. Then you can present your data in context of the conclusions.

Keep your data meaningful

You could also build a document with all generalizations and no data. In this case, you may be overly concerned about simplifying the data for your audience, and end up boiling it down to the point of meaninglessness. You may also find yourself unconsciously biased—wanting to prove a point but finding the data doesn’t stand up to your hypothesis. The solution here is also methodological: Before starting the document, identify a handful of data points that support each conclusion.

Presenting Competitive Analyses

When planning a presentation, you must first determine the purpose of your meeting, and then how you want to structure it.

Meeting purpose

In the context of user experience, there are two main purposes to a competitive analysis: justifying an overall strategy for the design and justifying a specific set of design decisions. Therefore, the purpose of your competitive analysis presentation will be one of these two things. To this end, you might combine the presentation of your competitive analysis with the presentation of other strategy documentation or design documentation.

Providing justification for design

Besides describing each of the criteria you looked at, there are a couple of other things you should include in your presentation.

Balance competitive research with user research. When presenting results from a competitive analysis, compare what you found against what users have asked for. Your presentation should show where there is a disconnect between a competitor’s site and your target audience, as well as instances where a competitor’s site does a good job meeting user needs.

Have an opinion. You’re the expert, and if your clients don’t put you on the spot explicitly, it doesn’t mean they don’t want to hear your opinion. Express what you like or dislike about each competitor. Talk about what works for you and what doesn’t work for you.

For designers, be explicit about design direction. If your presentation is strictly internal, your team members will want some kind of bottom line—a core set of take-aways they can bring back to their desks to help focus their efforts.

Providing justification for strategy

“Strategy” can mean a lot of things. When using competitive research to make overall decisions about the direction of the web site, the key to the presentation is identifying what role the research plays in your decision making.

Identify the focus of the research. Before diving into the specific differences between various competitors, you can provide some background by stating what it is you were looking for in the competitive analysis. Typical background questions for an overall strategy might relate to the industry baseline for a set of features, the scope and breadth of content available, or the balance between original content and advertising. By stating these issues up front, you’ve given the stakeholders some context for the presentation.

Identify the implications of strategic decisions. Since your pet-related web site wouldn’t jump off the Brooklyn Bridge just because Pets.com did, you need more than just a competitive analysis to justify a strategy. In considering the strategic decisions of your competitors, you need to identify what kind of impact the same decisions in the context of your business would have. By stating explicitly the implications of strategic decisions, you’ll go a long way toward putting the competitive analysis into context.

Table 5.10: Conclusions and Implications from a Competitive Analysis

Conclusions	Implications
Most pet-related commercial web sites feature a wide variety of products on the home page that rotate daily or weekly.	The site would need a system of business rules and a publishing strategy to determine which products would appear and how often they would change.
Most pet-related web sites use pet type as the main navigation categories.	We would need a publishing strategy unique to that pet type for each main navigation area.

Meeting structure

Once you have a sense of the purpose of your meeting, you can decide upon an agenda. Competitive analyses are the kind of story that can be hard to tell because there are two main dimensions—the characters and the moral. There are pros and cons to each. Focus too much on the competitors, and the main message (the moral) is lost. Spend too much time on the conclusions, and your stakeholders might wonder whether you’re just pushing your own agenda. Still, the meeting structure needs a spine, a central focus to keep conversation moving in the right direction, and this can be either one of these dimensions.

Competitor-driven story

It may seem counterintuitive, but since the competitors will form the basic structure of the meeting, you should actually start with an account of the criteria. Provide an overview of the criteria at the top of the meeting to set the stage and then dig into each competitor more specifically. Describe how each competitor measured up in each aspect.

This approach works well for laying the landscape and for addressing broader issues, like how each site serves the needs of its target audience. By taking the stakeholders on a tour of the competition, you give them a sense of what they’re up against. This walking-tour approach helps answer broad strategic questions, such as which features are available on the site, what appears most prominently on the home page, and how the site prioritizes content, but not specific design issues, such as comparing the treatment of “add to cart” buttons.

Though a competitor-driven approach is best for discussing larger strategic issues, it can also work well for specific design problems. If you’re looking at just one aspect of the design, your meeting can show how the particular design problem was solved for each competitor. When discussing a specific design problem, your meeting can end with the conclusions, drawing together lessons learned from all the competitors.

Moral-driven story

Instead of structuring your presentation in terms of the competitors and walking through all the issues for each competitor separately, this approach takes the opposite tack: For each issue, you talk about how each competitor stacks up against the others. To set the stage, provide a short overview of the competitors. You don’t need to get into comparing them at this point, but instead describe why they were included in the study.

The presentation then focuses on your conclusions. For each conclusion, you’ll first need to describe the criteria you looked at to arrive at the conclusion—in other words, what you analyzed on each site.

For example, you might conclude that the highest level navigation categories on a pet-related web site are usually pet-type, but that this isn’t the only system of categorization used on the site. To support this conclusion, you looked at three different criteria: the navigation categories on the home page, the metadata attached to products in the catalog and other content, and the structure of intermediate “gallery” pages (galleries are lists of products or content that appear on the user’s path between the home page and the product or content page). Finally, within each of these criteria, you make observations about each of the competitors. This structure is useful for both high-level strategy analyses and specific design problems: The content of the conclusions may be different, but the logic behind them is essentially the same. In this approach, the criteria do not disappear, but become a bridge between your conclusions and your observations.

Presentation risks

When presenting a competitive analysis, you might run into a few snags. You may have moments when the conversation is either derailed—a stakeholder brings up an issue that calls the methodology or the results into question—or gets way off topic. Here are a couple of ways your presentation might spin out of control, and how to reel it back in.

Maintain perspective

The worst way for a meeting to get off topic is to get too caught up in the competition. Although you’ve invited people to the meeting to discuss your site’s competitors, it can be easy to lose perspective on the purpose of the competitive analysis. Symptoms of this problem include getting stuck on one particular design element in a meeting about strategy, or spending too much time talking about one competitor over others. Even though your competitive analysis is meant to address one particular design problem, you might find conversation straying from that design problem into other areas of the site, or your participants might start talking about more strategic issues.

If the conversation is productive, you may not see this as a risk at all. However, if you have a specific agenda and certain goals for the meeting, these kinds of digressions may not be productive regardless of how interesting they are. To get back on track, jump in and remind the participants of the purpose of the meeting. One way to help stop this problem before it happens is to write the purpose of the meeting on a whiteboard or flipchart at the very beginning. If someone attempts to stray too far, you can always point to the meeting purpose and look very stern. You won’t be popular, but you’ll be respected for running a good meeting.

Know the rationale behind your methods

As you present the results of your competitive analysis, you may run into troublesome meeting participants who question your methodology. They may raise questions about your selection of competitors or criteria, or your technique for capturing data.

There is only one surefire way to address this risk: Don’t invite these people to the meeting. But, if that’s unavoidable (and it usually is), the second best way to address the problem is preparation. Methodological questions are easy to anticipate, and if you think about your rationale before going into the meeting, you can usually quash these hecklers. (OK, maybe they have legitimate concerns, but you don’t have to like it.)

Say you’re building this pet web site, for example, and your analysis looked at a handful of sites. That might not stop one of your participants from saying, “I do all my online shopping at JeffersPet.com. Why isn’t that in your competitive analysis?” If you’ve done your homework, you can respond with “Given the time constraints on the competitive analysis, we had to keep the number of sites down to four. We included DrsFosterSmith.com among our reviewed sites to represent the non-retail-store competitors. If, after our presentation, you think there are some aspects represented in JeffersPet.com that we missed, let’s talk about it offline.” Ah, the offline discussion, the Internet consultant’s secret weapon.

Open your mind to varying interpretations

It’s one thing for meeting participants to question the number of competitors or your method for scoring, quite another for them to poke holes in your analysis. Questioning your conclusions is a scarier risk than questioning the methodology, but in reality you may be less married to your conclusions.

In other words, once you’ve done the research for your competitive analysis, a serious concern in your methodology might mean throwing out the entire analysis and starting again. Questioning a conclusion, on the other hand, means revisiting the data for a different interpretation. If you’re convinced of your conclusions, be prepared to defend them vehemently. If you’re open to discussion about what the data mean, try spreading out the data and soliciting alternative interpretations. This could lead to worthwhile discussion.

Unfortunately, the only way to mitigate this risk may be to revisit the raw data. If you can rationalize your conclusion, you may not need extensive discussion (which would derail the meeting), but if you try to simply quash the criticism of your conclusions, your clients may start to question your integrity as a consultant. A vehement defense of ideas is one thing, but outright defensiveness is another. The bottom line is that you need to go into these meetings with the mental and emotional preparation to both unpack your conclusions and to go back to the raw data and observations if necessary.

Competitive Analyses in Context

The competitive analysis is a strategy document: It does not describe the user experience itself, but it’s a stepping-stone to getting there. It’s not an output of the design process but rather an essential input to it. However you employ your competitive analysis—whether for arriving at an overall strategy or for zeroing in on a specific design problem—you’ll need to anticipate how it will fit into your process.

Using a competitive analysis with other documents

With competitive research, your aim is to establish a context for discussions about specific design problems or overall strategic decisions, such as which features to include. Therefore, most of your competitive analyses will have to cooperate with other deliverables, which more directly document those decisions for your site.

Competitive analyses and user-needs documentation

User-needs documentation—at least in the context of this book—is any documentation that contributes to or comes from research into your target audiences. A competitive analysis can work with both aspects by helping to define an agenda for user research or put the results into context.

If you do your competitive analysis before talking to any users, you can use results of the analysis to identify issues you’d like to clarify through research. For example, in looking at different pet-related web sites, you might notice that the competitors use two different strategies for home-page navigation. Namely, some home pages just use pet categories to get users into the catalog, while other sites offer additional product categories (such as sale items, or “new this week”). Your user research can attempt to shed light on these different strategies from the user’s point of view. Your competitive analysis, therefore, can make reference to upcoming user-needs documents (such as personas) that will elaborate on these particular issues.

On the other hand, you might save your competitive analysis for after you’ve done some user research or usability testing. In this case, the user research can set the agenda for the research into competitors. For example, in your pet web site project, suppose you start the process for upgrading the site by soliciting feedback from some of its users in a usability test. That test might reveal certain aspects of the site that are most important, functions or features of the design that users respond to either very positively or very negatively. Imagine, for example, that the current version of your site offers extensive pet care advice that helps sell products from the catalog, and that users respond to this content. In this case, your competitive analysis might look at how other sites make use of advice content. This competitive analysis would use the usability test results as a rationale for examining particular criteria.

Competitive analyses and other strategy documents

Competitive analyses are strategic documents: They help define a design direction without defining the design itself. Other strategic documents described in this book are concept models—a tool for describing complex ideas that serve as the foundation for design—and content models—a tool for keeping track of all the content on the site.

Since each of these strategic documents aids the design process in a very different way, it’s highly unlikely that they’d reference each other. Still, you employ these tools as part of the competitive analysis, using them to describe or highlight different aspects of the competitors’ web sites. Put simply, you can do a concept model or a content model for each competitor as an additional means of comparison.

Competitive analyses and design documents

Even though competitive research can help drive design decisions, it’s difficult to draw specific references between your analysis and a set of wireframes, for example. By the time you’ve gotten that far along into the design process, the information from the competitive analysis has already been absorbed into the design team’s approach. As a result, the relationship between a design decision and a specific observation in the competitive analysis may not be explicit. When competitive research addresses very specific design concerns, it should acknowledge the source of the design. For example, the design team behind the pet web site may be trying to decide where to locate the search box, a very specific design decision. A survey of competitors might reveal that they all put their search box at the top of the page, and the majority on the left-hand side. When putting a wireframe together for the new site, you could point to the competitive analysis by way of rationale. At this stage of the design process, however, such references are sometimes unnecessary.

Acknowledging the competition

A competitive survey, no matter how rich the information gleaned from it, will always play second fiddle in the design process. After all, just because your competition does something doesn’t mean you should. The competition is a good place to get ideas and to establish a baseline, a cost of entry. But the value of that information in making design decisions is limited at best. Innovation moves fast online. Regardless of how the design process evolves, information about how other sites address the problems you face will always be valuable, because it keeps you abreast of the latest trends in technological change. With innovation comes a change in landscape. At the most basic level, your audience has to make a choice and it’s our job as designers to make that choice easy, even when we don’t understand all the factors that go into the decision.

When the commercial Internet emerged, retail stores faced a new kind of competition. Suddenly, competitors were lurking around every corner. This is still the case. What drives your understanding of the competition, therefore, should be a keen understanding of your audience. On the Internet, the competition is more than simply every other site trying to do the same thing you’re doing. The competition is nearly any site that can attract and hold a user’s attention. By knowing how your audience spends its time and how it makes decisions, you can anticipate how other sites or technologies are getting its attention, thereby expanding your survey of the competition.

---

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.

Related Topics: Business, Critique, Planning, Web Analytics