Macadamian Blog
Visual style - let’s make beautiful software
How many times have you heard "Design isn't how something looks, it's how it works"?
Rubbish.
I understand how we got here - for years, UX and Usability professionals have been fighting with executives and engineers about how hiring a graphic designer to put lipstick on the pig doesn't a good product make, and that you need get serious about usability. But now we've swung too far the other way. Usable is great. Useful is better. But, if you want to create really desirable, compelling software, that people are going to want to buy, then you'll have to start paying more attention to aesthetic and desirability.
Mobile: What can we learn from the web?
In my previous blog post on "Has the mobile created a gap in your product offering?" I started exploring a model on how to rethink the experience of my product line given the mobile wave hitting us now. In this post I want to look back on what my overall big lesson was from the web days. I want to see if I can use it to understand how I can move my mobile strategy forward.
The Next Big Innovation In Healthcare
As patient volumes increase, it can become difficult for providers to remember medical background material specific to each and every case. A Clinical Decision Support (CDS) system combines both background knowledge and case-specific information to help the provider make better decisions. A CDS can point a provider to reference material and information, perform certain actions, or supply alerts.
A Common Clinical Decision Support Configuration
What are the opportunities for innovation?
EMRs are benefiting from CDS system integration by keeping medical knowledge at the fingertips of healthcare providers. Opportunities to push the envelope include:
- Learning patterns of care. Some CDS systems do not require a pre-developed knowledge base, but rather apply machine learning algorithms on medical examples fed into the system. Others can apply machine learning to build knowledge about user behavior. This concept exists in academic publications but the challenge remains to make this a widespread commercial reality.
- Reducing alert fatigue. When users receive too many alerts, they may begin to ignore serious alerts.
- Improving usability. Organizations can innovate by increasing the intuitiveness and speed of access in CDS systems. To date, whenever a CDS has been integrated into an EMR, the combined system has tended to behave as two distinct systems with different flows. Companies that can address this problem and improve usability stand a good chance of becoming leaders in the space.
- Sharing CDS. Multiple healthcare facilities can share anonymous data about clinical decisions so that they can benefit from each other’s behavior and improve the quality of patient care.
- Improved matching algorithms. Strong algorithms that match case instances to a knowledge base will deliver more pertinent information to users.
How are CDS systems used in healthcare today?
Most of today’s CDS systems provide clinical reminders for patients, identify possible risks for adverse events and errors, analyze clinical performance, and encourage adherence to standards of care.
The most common use of CDS/EMR integration is to identify drug-to-drug interactions. Popular solutions for e-prescription and drug interaction detection include DrFirst, RxNT, Medi-Span and a host of others.
A CDS can also be deployed as a stand-alone package that provides a lookup of medical knowledge. The Isabel and Elsevier CDS tools currently offer this functionality.
What constraints does CDS technology face?
- Patient data model accuracy. Patient data must be entered consistently for a CDS to contain accurate information.
- CDS system accuracy. The CDS system needs to be intelligent enough to be able to correctly identify similar cases and present them to the user when requested.
- Alert fatigue. A CDS system must take care to only alert the user in critical situations. Otherwise, the user will suffer from alert fatigue and ignore all alerts.
- Usability. CDS systems are often developed independently from EMRs and then integrated. This process limits the usability of the CDS. Users may choose not to use the CDS or become frustrated when it takes them too long to get to the information they need.
When a CDS becomes a nuisance as opposed to an aid – either because of inaccuracies, poor usability or alert fatigue – the system loses value and users simply abandon it. Organizations that can closely integrate their CDS with an EMR and overcome the obstacles mentioned above will have the best chance to stand out from the crowd.
Successful Mobile Apps - Offer Real Value
Unless you’re recently back from a long stay on a deserted island, you’ve been bombarded with news reports, market forecasts and customer input about mobility.
But for every monumental success like Angry Birds, there are thousands of apps that fail. iTunes alone boasts 330,000 mobile apps, and this number is in the millions when you consider other app stores, direct downloads, and mobile web applications.
We’re constantly developing new mobile products for our customers, and we’ve distilled some of key elements that make the difference between a mobile product that becomes part of your customer’s daily lifestyle, and a product that is lost in the shuffle, 'collecting dust'.
Ignore the User Experience, Lose A Billion Dollars
From this recent case study - "Walmart based this incredibly expensive misadventure on what customers said, rather than what they did. And the customer experience is all about what customers do."
This is more or less right. The survey - and what Walmart did with it - are subject to three obvious problems I can see:
- You are not asking users why they are responding that way
- You letting a non designer design (if you take a suggestion verbatim) and just implement
- You are mixing up perceptions (what they say) vs. behaviours (how they will actually act); the point up above from the article.
The general point is that if you want to design anything (software, a store, a culture) and you want to do research to inform your designers (software, store, culture) then you need to be aware of both the benefits and limitations of the research.
You need:
- a clear purpose and question(s) (walmart didn't) correct methodology (survey wasn't it)
- experts to carry out and interpret (they didn't have it)
- talented, knowledgeable well informed designers who know what to do with the information (again, missing from Walmart)
This is what software companies absolutely need in their product management and design process. If Walmart had 90 seconds of our time (or any professional product manager or UX researcher's time), they could quite conceivably have saved a billion dollars.