Web site undesign and redesign
Topics
Web site redesign and undesign
Improving engagement and conversion through web site redesign
How often should redesign occur - TIMITI?
References and Further Reading
Introduction
When was the design of your organisations web site last changed? If it was launched or relaunched more than a couple of years ago then there are likely to be significant opportunities to improve its effectiveness. Of course a web sites effectiveness depends on what you are trying to achieve through it. So we start by looking at how setting practical objectives can be linked through to web marketing evaluation. We then look at how different design philosophies can give poor results and how the concepts of redesign and undersign have arisen as a result. In the last section, practical guidelines to improve the marketing effectiveness of a web site are given.
Site design - Objectives and evaluation
Large organisations may have a range of web sites such as the corporate site, partner extranets and microsites for managing response to campaigns. But, here we concentrate on the main customer-facing web site. We look at practical objectives which can be used to inform and assess changes to design rather than strategic objectives.
Primary traffic volume metrics
Regardless of the type of market and product offering, there is a common factor in setting objectives, and that is traffic. Web site measurement tools give us three alternative measures for traffic.
1. Page impressions (views). One page impression is one individual viewing one page.
2. Visitor sessions (visits). Usually defined as a series of page requests by a visitor without 30 consecutive minutes of inactivity.
3. Unique visitors (uniques). The number of individual visitors to the site within the reporting time period such as month or quarter.
Note that the term 'hit' is no longer used by web professionals since it is not a reliable measure for the number of people viewing a page (it is recorded for each graphic or text file requested from a web server, so if a page contains 10 graphics, plus HTML file, this is recorded as 11 hits).
Reach
Which of these primary metrics is most useful for setting marketing objectives? Well, although they are all commonly used for reporting, if you pick one, it has to be unique visitors. This enables us to measure how many people we are reaching via our web site. It shows opportunities to communicate with our customers and potential customers and convert them to other outcomes on the web site. So specific objectives should be set for volumes of new visitors and customers in different segments according to the potential audience in a given country. Methods for assessing visitor reach in a market include:
ISP-based monitoring from Hitwise reports on the relative popularity of over 150,000 UK web sites in 150 industry categories (www.hitwise.co.uk)
Home and work based panels from Nielsen Netratings (www.nielsennetratings.com) can give an idea of your site reach compared to competitors for different demographics)
Using the Overture keyword suggestion tool (www.overture.com) gives an indication of how many are searching on terms relevant to your product offering.
Engagement metrics
A site may generate large numbers of visitors, but for the site to be effective, it needs to engage the visitor, to encourage them beyond the home page or the first page they arrive at. All site managers should assess their engagement rate. This can be calculated by dividing the number of visits where visitors only view one page by all site visits and then expressing as a percentage. If your engagement rate is over 80%, this suggests you are encouraging the vast majority of interested visitors into the site. Some visitors, may not be within your target market and are likely to leave immediately, whatever steps you take. However, if engagement is only 60%, there is clearly room for improvement and possible redesign steps to take are described later in this article.
Conversion metrics
For the visitors we have engaged, converting them to outcomes is a key objective. Typical conversion objectives include conversion to:
Lead generation through an e-mail acquisition offer such as a whitepaper download
A particular type of content, e.g. news stories
Product related content, such as items in a catalogue indicating propensity to buy
An online purchase conversion between the different steps leading to purchase is assessed so steps can be taken to reduce the attrition at each step
A Where to buy interaction directing visitors to a store or distributor.
A customer service request
For every site, it will be possible to devise several conversion measures which show the success of the site in achieving real marketing outcomes. When first calculated, conversion rates can be disappointing, but they are a starting point is required so we can use changes to the site to improve them.
Achieving conversion to capture an e-mail address from a first time visitor is particularly important since many site visitors arrive never to return. Once they have opted-in to receiving e-mail communications, then messages can be pushed out to their inbox.
Web site design philosophies
Your web site may not be fulfilling its potential if its design was influenced not by clearly defined objectives, but the design philosophies of its creators. Consider these alternative approaches:
Design-led - Here, the emphasis is on aesethetics rather than content. An elegant graphical design is to the fore, and the importance of information for users, or usability is marginalised.
IT-led at the other extreme attention may have been focused on the information architecture there is certainly a lot of well-structured content, but it may be difficult to see the wood from the trees.
Information led these sites are stripped of most graphics and focus on information the increasing popularity of weblogs such as Marketing Fix LINK: www.marketingfix.com suggests the popularity of this approach. This approach is an example of undesign which is described in the next section.
Usability led here the focus is on the user experience making navigation, interaction and access to information straightforward. Usability gurus such as Jakob Nielsen and Vincent Flanders have stressed the importance of this approach for a long time. Customer-centric design, has to be right, but it is not the only approach sites still have to contribute to a business and this requires customers to be led in the right direction. Which brings us to:
Marketing led this approach depends on the nature of product plus the perspective of the marketer do they see the site as a branding tool, direct response tool or a customer relationship management tool. I would argue that a marketing led approach, based on engagement and conversion of visitors is crucial.
Obviously, no single approach is best, rather it is a question of getting the right balance. To determine which approach works best, then we need defined aims, and a measurement system in place to assess whether we are achieving these aims. This E-marketing Insights Column explained the options for different tools: LINK: www.wnim.com/issue4/pages/Emarketing.htm.
Web site redesign and undesign
Web site redesign is simply a redesign of an existing web site to improve its effectiveness. It is commonly needed since it is impossible to get any web site right first time, or second time or any time. Business objectives change, branding changes, technology changes, the style in vogue changes and so audience needs change.
As Kelly Goto and Emily Cotler say in their book Web Redesign Workflow That Works:
There are many good reasons to redesign: You inherited an old, mostly brochure-ish site when you bought your companyYou want to update your image with Flash or other mediaYou need e-commerce capabilities now that your services have expandedYou need a content management systemYour site is a navigational mess, and your site users arent getting where you want them to go
What then is undesign? According to Justin Kirby, quoted in a recent article in Graphics International:
The web has finally found its voice. Design has become a way of bringing content out, not an excuse for lack of content He sees undesign as a backlash against overdesigned visual language on the web: designers from print, video games, contemporary music or street fashion whom he accuses of not understanding the craft of information design.
So undesign implies simplification of design with a greater focus on content and enabling access to content through usability. For examples of undesign see the portfolio of Jason Kottke (www.kottke.org/portfolio/portfolio.html).This article also suggests the importance of weblogs or blogs in this trend with sites such as www.blogger.com now helping tens of thousands of Internet users to preserve their stream of consciousness by recording them on the web. More sophisticated content managements systems such as Plone (www.plone.org) enable a range of content types such as events, news and articles to be stored, updated and accessed through a clean undesign.
So, those are the principles, now for some practical steps to improve web marketing.
Improving engagement and conversion through web site redesign
To encourage engagement and conversion, we should think about the barriers that stop engagement and conversion and then use redesign to remove these barriers. Lets look at three common barriers.
Barrier 1 visitor cant see the point:
The first barrier is when the visitor first arrives on a site. They make a quick assessment of whether to stay or go based on the design of the site and the key messages. A good design establishes credibility. A key part of this is the online value proposition. Research reported by Jakob Nielsen shows that new visitors to a site, particularly of a company they are unfamiliar with, will seek out a tagline or graphic that explains what the site can offer (WIFM or Whats in It For Me).
Indeed, for UK provider egg, their current tagline is egg: whats in it for me? Users decide quickly (often within a few seconds whether to stay or leave a site and the tagline is a large part of this).
We can suggest that we need to think about a sites Online Value Proposition or OVP It should:
Reinforces core brand proposition and credibility, but also shows:
o What can a visitor get from your site that
They cant get from you offline
They cant get from competitors
Some of the best taglines have been developed by the dot-coms, for example:
Compare. Buy. Save. Kelkoo (www.kelkoo.com)
Earths biggest selection. Amazon (www.amazon.com)
Search the largest inventory of cars and trucks on the Internet. More than 1.5 million listings, updated daily (www.autotrader.com)
Of course, graphics, features and benefits lists and the content itself all can help communicate the OVP and establish credibility.
More specific call-to-actions can also help engage the visitor. View this example from software company Microstrategy (LINK:www.microstrategy.com). The top third of the screen is used for branding and to highlight offers that are continuously present to help acquire leads through e-mail addresses. So think about how use the space on the homepage for different messages.
In his book Homepage usability, 50 web sites deconstructed, Jakob Nielsen evaluates sites for different messages. For the Citigroup Corporate page (www.citigroup.com) the proportions at the time of publication are in the table below. Since then, the company has redesigned the site and the new figures are shown as rough estimates. It can be seen that the design has been revised to increase the space for marketing messages and reduce unused space.
|
Screen element |
Original |
Current (est) |
|
Welcome and site identity |
8% |
>10% |
|
Self-promotional |
0% |
>15% |
|
Advertising/sponsorship (third party) |
0% |
0% |
|
Filler (background graphics) |
16% |
<5 |
|
Content of interest |
6% |
>15% |
|
Navigation |
4% |
<10% |
|
Unused |
48% |
<30% |
|
Operating and browser system controls |
19% |
19% |
Barrier 2 visitor cant do what they want
Visitors cant go where they want when there are practical problems with the site. These include problems with download speed and availability. The 8 second rule from Zona Research suggests that if your site takes more than 8 seconds to download, you lose half of your new visitors.
Poor usability is also a problem particular when the needs of different types of audience havent been assessed. As Jakob Nielsen says in his Novice vs. Expert Users alert box LINK: http://www.useit.com/alertbox/20000206.html):
Web users are notoriously fickle: they take one look at a home page and leave after a few seconds if they can't figure it out. The abundance of choice and the ease of going elsewhere puts a huge premium on making it extremely easy to enter a site.
Visitor types
Many sites focus site design on the visitor or the customer, but without really getting to grips with the vast diversity of visitor types and visitor needs. This goes beyond different segments. To start with we can identify different types of familiarity and corresponding actions:
1 Familiarity with the Net In many countries, there are still substantial new users to the Internet, so we need to keep it simple for them and reassure them. At the same time, there are now many experienced users on the Internet and we need to provide shortcuts for them. One thing research shows is that they love a search facility, since theyre so familiar with using it on search engines.
2 Familiarity with organisation for customers who dont know your organisation, content is needed to explain who the company is, what it does and how it does it, and demonstrate credibility through About Us options and customer testimonials.
3 Familiarity with organisations products even existing customers may not know the full range of your product offering. So ensure the online value proposition, navigation and page layout shows the range of products and allows for cross-selling opportunities.
4 Familiarity with your site Site maps, Search and Help options are not nice to have options for an e-commerce site since you may lose potential customers if they cannot be helped when they are lost. It is also worth thinking about offering specific content for First time visitors.
Visitor scenarios and customer journeys
For each of the audiences identified for a site, it makes sense to evaluate visitor scenarios or customer journeys. As Patricia Seybold says we need to get inside the lives of our customers, to really understand the way they think; what motivates them, what worries them as they traverse a web site with a goal in mind. It is clear which sites have done this type of evaluation. Design and navigation supports different types of visitor at different stages of the buying decision. For example, the Nationwide bank (LINK:www.nationwide.com) greet the visitor with:
How can we help you today They then have options for: Get a quote, Access your account, Find an Insurance Agent, Claim
Providing such a task-based navigation scheme can help visitors on their journeys whether they are a new visitor or an existing customer.
Navigation schemes
A range of navigation options should be considered in a redesign since different visitors think in different ways and have different needs on arrival. The most widely used navigation approaches are:
1. Product or service-based different menu options take the users to different company offerings.
2. By audience segment Dell for example offers menu options for Home users and different types of customers.
3. By familiarity new or existing visitor or customer.
4. By offer the site design has a range of calls-to-action to appeal to the customer.
4. Task based (see above).
Using all of these is not the right approach for all sites, but more than one is usually best. See http://www.harcourt-international.com for an example which uses the first three. The British Gas site (www.house.co.uk) also uses a range of navigation approaches.
Barrier 3 visitor feels insecure
Many Internet users feel insecure on an unfamiliar site, particularly if they are asked to provide details such as e-mail, contact or credit card details. In the UK, the Consumers Association conducts an annual survey (http://www.which.net/surveys/key.html) of Internet users and non-users. Essential reading for all those who want to understand e-customers. In 2001, the survey showed the extent of this problem 44% of Internet users still think it is unsafe to buy things on the Internet with a credit card. This has only decreased slightly over the past two years from 51%. However, the research also shows that guarantees and independent certification can reassure 73% of online shoppers say they feel more confident in the knowledge that the site included some form of independent guarantee.
So clear privacy statements and explanations about security can help reassure visitors. For example, when collecting e-mail addresses, we are most concerned about our data being passed on to third parties. So a form which reassures the visitor we will not share your information with any third parties is much more likely to be completed than one where this statement is buried inside a linked privacy statement.
How often should redesign occur - TIMITI?
Major redesigns involve considerable investment, so they are unlikely to occur more than once a year or every other year. However, this does not mean that the design should remain static in between site re-launches. This is a missed opportunity. In his book on web metrics, Jim Sterne has described what he calls the TIMITI approach to improving web sites. TIMITI stands for Try It-Measure It-Tweak It! In other words, key features of the design such as the calls-to-action that are likely to improve marketing outcomes should be changed, the results measured, and further changes made to improve engagement and conversion rates. So the answer to how often should redesign occur is continuously!
References and Further Reading
Chaffey, D., Mayer, R., Johnston, K. and Ellis-Chadwick, F. (2008) Internet Marketing: Strategy, implementation and practice. Financial Times Prentice Hall, Harlow, UK. 4th edition.
Farrelly, L. (1998) Design and Undesign. Thames and Hudson.
Goto, K. and Cotler, E. (2001) Web Redesign: Workflow That Works. New Riders. Publishing, USA.
Graphics International (2003) Low-fi allstars. Graphics International 102, p33-34.
Krug, S. (2000) Dont Make me think a common sense approach to usability. New Riders Publishing, USA.
Nielsen, J. (2000) Designing Web Usability. New Riders Publishing, USA.
Seybold, P. (2001) Get inside the lives of your customers. Harvard Business Review. May 2001, pp80-89.
Sterne, J. (2002) Web Metrics: Proven Methods for Measuring Web Site success. John Wiley & Sons, Inc, New York, NY.
Links
Jakob Nielsens Useit (www.useit.com) Including Alertbox column on usability
Vincent Flanders Web Sites That Suck (www.websitesthatsuck.com). Learn good web design by looking at bad web design.
ClickZ site design and writing online channels (www.clickz.com)
