Take the 2-minute tour ×
User Experience Stack Exchange is a question and answer site for user experience researchers and experts. It's 100% free, no registration required.

I redesigning a GUI of a programm, that is in use by customers for several years. Now the icons should be replaced with more simplified versions (like Metro style) and also the design regarding the shape of components and its color should look more puristic. So the GUI will change reagrding icons and slightly regarding other styles like color and shape.

Now I am thinking, wether its a good idea to offer some sort of Classic Mode showing the application in the old style in addition to the new style e.g. for customers who are used to the old design and don't want to get used to the new one.

Ignoring the extra effort to implement the design switch, i don't like this idea, because it breaks the rule to have an consistent GUI for every user. On the other hand i like the idea because it takes care of the less flexible users who don't want to change to a new GUI.

So my question: Is it a good idea, to offer a Classic Mode in addition to a new design in an application, that is already in use by customers for many years?

share|improve this question

3 Answers 3

up vote 4 down vote accepted

Good question.

I would say there are two approaches to handling this situation

  • Do incremental updates to small sections to see the user feedback and adoption : This will allow users to quickly adapt to the changes slowly while also retaining the existing user flow. You can also use tool tips and information bubbles to inform users of the change so that they dont get overwhelmed. Google while moving it to a new look had a video which walked users through the change

enter image description here

  • Make the complete design change but allow users to learn about what has changed as shown in the above example. Also give them a chance to temporarily move back to the old design with a limitation that that design will eventually be phased out. Gmail did this effectively with an option to go back to the old compose option as shown below.

enter image description here

TED talks does this with a link back to the old site so that users can opt for the old site as shown below

enter image description here

share|improve this answer
    
Thanks for the overview, i think the google aproach in letting the user know about whats new is a good way in general –  alex Feb 7 at 10:49
1  
I think gradual change is a good solution that let's you avoid double maintenance and still holds users by the hand while changing to a new visual look. –  kontur Feb 7 at 12:45

If you are going to have both old and new visual design as an option, you are following theme mentality. In some software, user have different options for changing the icons, colors and so on while keeping the main interaction flows same.

However, if you are changing also the main interactions and places of these buttons, you are redesigning the overall solution including visual design. In case of following this path, you should focus on the learn-ability related enhancements like FTE, Tours.

For a higher rate of learn-ability, the option of forcing user is a general strategy which has also its own set of risks. The acceptance rate of user is a crucial point.If you did not have any user test for validation, you should be very careful about what you are doing.

It might harm the business and get many negative feedback while thinking that you improved a working product. If you are the designer, you will be the first blamed one :).

It is a good idea to have both, it is the best design if you have one.

share|improve this answer
    
You are right with beeing careful and i also like your conclusion, that one design is best. –  alex Feb 7 at 10:50

If you are just doing a graphics/layout refresh, I would recommend against adding a theme switch (see reasons #2 and #4 below). Otherwise, there might be reason for it.

Several areas to consider:

  1. Usability. Will it improve usability of the site to have two themes? You need to add UI to change the theme (more clutter). Users will be faced with an additional decision in the way of getting work done. OTOH your old clientele will be able to keep using the system without re-learning.

    However, keep in mind that old users are a minority in a well-running company. "Death will take care of that," as Steve Jobs said on why he didn't add a touch-typing tutorial for the older users. Old users die, or retire, new users come in daily. So you'll be maintaining the Classic UI for a minority.

    Is that really worth it? Is it worth the added confusion to new users? And assuming your redesign is really better, don't you want your users to benefit from the better UI, instead of lazily sticking to the old, defective UI which they only prefer because they've gotten used to its idiosyncrasies?

  2. PR, brand image and outward appearance. Often, users learn about products from their friends. They see what they use, then look for something that looks and works the same. If your most enduring and thus happiest users, your multipliers, are using the old UI, your advertising will not work on their friends, because it'd obviously have the new look and not be recognizable.

    Also, if a user gets the announcement of a redesign and your claims that it is better now, but then sees they can switch back, they are bound to wonder why you provide a switch? Is the new UI still buggy? Are they not actually convinced it is better? Giving them the option to switch back encourages them to be complacent. To stick with the tried-and-true. So unless your new UI is missing important features, or is otherwise deficient, why would you provide a switch back? All you're doing is offer the user a way to switch back to a worse situation. You're providing a "Shoot me in the foot" checkbox.

    But be honest with yourself here; if your UI really is a radical departure that drops features or otherwise is likely to impede workflows a large number of customers has, you may want to let them switch the theme. Or alternately, you may want to fix that and just ship a little later, once your UI is actually better. Or, you could selectively allow theme switching (e.g. Twitter did a big redesign a few years ago, but a lot of the account settings still had the old look until they got around to properly redesigning that part).

    Also, consider the band-aid metaphor: Would it be more beneficial to tear off the band-aid right now, endure the yelling that users do when something familiar suddenly changes, and keep marching forward? Or do you want to slowly tear it off bit by bit, risking a constant group of niggling users telling people "use the old UI, it's better" or complaining that you should restore the old UI? Give them time to organize and start a petition to get the old UI back? It depends on your situation, on how radical the change is, on what kind of users you have, which of these approaches is right. But if nobody complains about the redesign, you probably haven't changed anything.

  3. Legal considerations. Features you advertise have to be available to users. If the old theme doesn't have that new feature, but still provides an old one being retired, you could run afoul of false advertising laws if it is not possible to use the two features together, for example. It's rare for this to happen, but something to keep in mind when there is a functional redesign.

  4. Effort. Apart from actually implementing your code so you can switch the theme and everything updates, switching themes also affects other areas of the site. E.g. what to do with FAQs and other documentation? Offer old or new docs depending on the current theme? You're also doubling the QA effort involved, because both themes need to be tested.

share|improve this answer
    
Thank you for the detailed discusion, i think i like the idea of having multiple modes less with each post –  alex Feb 7 at 12:02

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.