Monday, August 30, 2010

Its all in the data : How user interface design is driven by information.

As designers it's only natural to re-visit or re-use some design implementations that have worked best for us. At some point in our design careers we have a little cheat-sheet in our pockets that we use to allow for quick solutions. This creates individuality and gives character to our designs. There also comes a point where we append our cheat sheet with principles and methodology that have proven successful by other designers or solutions.

But are you considering the fact that seldom do 2 solutions function identically? If you take a close look at your project history, you may have worked on similar products or you may have worked in sibling products and the oddball case when your asked to re-create based on an existing product. Now whatever may be the case, the end solution will always be different.


The key underlying fact is the data offering. Based to the type and degree of data your, your designs will change. That's a fact. Lets put an example to this.


We are/have designed for 3 products, all solutions in the realm of data reporting. To take a more generic case, data is based on financial information.

  1. Product A : Low degree of data, has most of the public information and a little historical data.
  2. Product B : Low degree of data, has most of the public information and a little historical data but a more reputable brand name.
  3. Product C : Medium degree of data but has a huge database of proprietary information that is not available anywhere.

The Objective
All of the products sell a report that a subscriber will pay to obtain. The cost most certainly will vary, simply because of the quality of data.


Lets look at a hypothetical design for the three reports


Report A : Since most of the information is publicly available, its more about data consolidation and presentation that will be the key selling point for this report. The objective would be to provide the user with a one stop shop instead of doing the whole search and information gathering himself.

Some of the key design considerations
  • Present a simple report with formatted data
  • Use graphical representation for the information.
  • Highlight the span or spectrum of sources that has been used to obtain the data.
Report B : Now this company provided the same data but it all tagged with a popular brand name that you can trust. Right from the get-go your under the assumption that people will trust your data and information and the view that you provided which is an advantage.

Some of the key design considerations
  • Present a branded report with formatted data
  • Provide a different perspective at looking at the information : Graphs, charts, maps, etc.
  • Since the report if coming from a brand name that people trust, information source wouldn't be a big factor. Its a design decision.
Now you can see that although the both reports proved similar financial information, you will have the 2 reports laid out differently.


Report C : The main selling point its the core data that is proprietary to this company. This information is not available anywhere and emphasizing that information will drive the design.

Some of the key design considerations
  • The core data need to be highlighted in the report layout.
  • Probably section the report as proprietary and public views?
  • Use of graphics for the core data may be more useful than for the public data, but that's purely a design call!
  • The public data can be formatted or presented as is, another design call you would make.


Conclusion:
Ok I'm sure at this point you've got a lot of question on some the pointers above, its only an example and a hypothetical design. The point I'm trying to make is how the data will affect your design process and some of the design decisions you will make. To some of us it may be a given during our design stage, some may not realize it but for some who don't it good to understand the data your working with and then work at designing a solution tailored to that data.


A good example would be to quick case study on a product search results page. Different solutions have different meta data information they provide or let you search on and you will notice how the results vary.

0 comments: