graykuglen - Tumblr blog

graykuglen · 8 years ago

Text

Data Driven Design

Both data analysis and design research share a common goal and that is to answer questions about why. It might be better to say ‘data analysis’ seeks to come up with a hypothesis based on insights found in data, and ‘design research’ looks to gather evidential insights by direct observation, but I’m still chasing the same question of why?. I’m learning more about data analysis to inform design and it’s really exciting. Now what if we lead with data analysis to inform design?

Using some simple investigation and some statistical analysis tools (”R” + “R Studio”, Excel) I explored the topic of home buying. The information I found helped inform some concepts, which I created in Sketch and then prototyped simple micro interactions using Principle, a prototyping tool. I’ll keep it brief and go over the thought process using a data analysis framework (below).

DEFINE

There’s a growing disparity between people who can afford more than they need in a home, and those that cannot afford much of anything at all. An unfortunate outcome are skyrocketing prices in metro areas with a limited supply of homes. Stated more plainly, people with more money drive up the prices for others. Nobody likes to pay too much though, could more helpful data about the market/sales help home buyers (across the income spread) not overpay?

I’ve lived in two of these metro areas - San Francisco and Portland. An interesting side note is that San Francisco’s high prices are actually driving up Portland’s from displaced Californian’s moving to Portland.

DISSASSEMBLE

How do home buyers make purchase decisions?

Is the buyer at a disadvantage (asymmetry of information) without the right information to make better informed decisions on price/value?

What information (about the value of a home) may make buyers more informed about pricing, if any? Could this new information slow pricing increases?

QUESTION 1 - How do home buyers make decisions to purchase

In a market that has so many buyers seeking the same type of home there’s gonna be competition. It’s the market economics of supply and demand, if demand is high and supply low, prices go up. More information for the buyers may not make much of a difference here. However, I spoke with a few very successful real estate agents about how their clients make purchase decisions, here’s their response (paraphrased):

“Buyers are very rational when they begin shopping for a home (mostly for first time buyers) making financial decisions around affordability and needs paramount. But in the end the deciding factor is emotion and desire.”

Let’s assume this expert view is valid. It is very probable that first time home buyers enter the buying process logically and rationally and see what they can afford. We may focus first on influencing the people who are driving up prices. AND, we'd want our focus to be in the early stages before emotion steps in.

QUESTION 2 - Is the buyer at a disadvantage without enough information to make more informed decisions?

Price/value information available for new buyers, while “still rational and pre-emotive”, may be limited. Consider the following:

MLS data shows all the facts about a property, But there is little pricing or market data in it.

Online websites offer the same data in a better presentation, but may be stale and not timely. Regional MLS's see to that.

Websites do offer price $sq/ft which is great. Some even offer estimates of value, but with no insight as to how they arrived at the price.

Buyers are beholden to their agents for “comps” or comparables. These are 1-1 comparisons of similar properties which sold recently.

Some online websites do offer comparables, but the properties can be questionable which may create doubt and require more analysis.

Getting immediate pricing of what is selling/sold is guarded. Other free options take effort and added analysis.

You may pay for sales data using services like Dataquick but this is not ideal

Both selling and buying agents get a commission on the property sale, the higher the price the more they make - just saying.

The market data THAT is available is aggregate and shows averages of what properties go for in the Zip code or neighborhood, seldom at the property level that is accurate.

So, maybe there could be an opportunity to show more value-based information.

QUESTION 3 - What information (about the value of a home) may make buyers more informed about pricing, if any?

I evaluated the information available online to buyers. There definitely is some good information to help but it’s scatterd. Key areas I find most interesting and worth exploration include:

Realtime sales data of comparable properties (may or may not be attainable)

Exploring and expanding on a metric that already exists e.g. price / sq ft

There are different qualities of homes, why not grade the properties with desirability or a property grade A, B, C... (interesting, but is this viable?)

Better yet, why not make this about ‘my needs’ for a property which could be looked at differently

EVALUATE

I scraped data online for 3BR, 2+BA homes in San Francisco (Redfin.com), active or sold within the last 90 days - there were 347 records. Using a data analysis tool called “R” and “RStudio” I began exploring the data.

Scraped active listings for 3BR, 2+BA

Scraped comps (past 90 days) for 3BR, 2+BA

Queried the two together to get a better understanding of the price per square foot for 3BR’s using summary data, median, mean, histograms etc..

Insight

Sales prices in San Francisco are higher than the list price by an estimated 12-13%. The insight here “what if this information could be more readily available to buyers”? Could it influence offers?

Look at the histogram (above) for Active listing (For Sale) 3BR/2+BA single family homes, and you’ll see the largest number of homes in the median, or middle.

The pricing for sold homes (above) is considerably higher by 12-13%. That could be for several reasons and something to investigate.

More questions arise

San Francisco is one of the hottest markets in the US. How do “cool-markets” compare to “hot-markets”?

Can I get a list-prices to compare with my sales prices for 3BR/2+BA? This may lead to a correlation analysis and possibly my first regression of 3BR/2+BA’s (sadly I cannot get this data)

Because it’s a hot-market, it may be wise to use control groups (in two different markets for hot and cool) to test effectiveness of any price/value design concepts. In other words, test with concept in hot market, and w/out concept in hot market, and do the same for a cool market.

Design insights to explore

Show buyers a histogram? If we show a live histogram (shown above), could it help buyers understand pricing of properties better?

Use price per Sq Ft to show a distribution of where ‘this’ home lays on a curve of 3BR/2+BA homes.

Define an acceptable pricing range based on actual sales data. If a property is over price range for 3BR/2+BA, indicate it somehow.

Don’t waste buyers time - show only properties that are in the rational realm of their minds and avoid properties outside their decision criteria. This may also avoid the “I love it and need to have it” scenario.

Is it possible to tailor listings to the buyer’s decision criteria? For example, what is an acceptable match to my simple stated needs --> 60% of what I want, 70% of what I want? Might there be a way to show this? Maybe we could re-look at saved searches....

DECIDE

My original question was “is there an imbalance in the available information to help home buyer’s make more informed decisions that don’t drive up prices for everyone?”

Two concepts explored

The first concept is to help buyers by better matching them with what they set out to buy originally. It would reflect the decision criteria they started with, e.g. 3BR/2+BA, 2000sq ft+ etc.. This could be a simple metric to show at what % does it meet my stated needs. Buying a home is very time consuming and stressful, this is intended get buyers to physically visit only the best matched properties.

The second concept explores the value of a home compared to actual sales data. It would have two parts - a histogram highlighting where this property falls on the distribution curve, and the second part would show variance away from the median by some measure (yet to be defined), If its over 10% (or whatever) higher than median, indicate its more expensive than properties of this type.

Queried Data

Bedrooms = 3

Bathrooms = 2+

Sq Ft = 2000

Location = Zip codes 1, 2, 3, 4

The above prototype shows someone browsing a list of found properties. In the right hand column of the interface are the two concepts and a histogram to give home buyers more information about value:

the top icon is showing an estimation on a match (94 = 94% match to my wants) with my stated needs for a home

The bottom icon (below it) indicates “is it over priced or not?”. That calculation is based on an acceptable variance (maybe SD) amongst properties sold over the last 90 days that meet my stated needs for a property

(function(i,s,o,g,r,a,m){i['GoogleAnalyticsObject']=r;i[r]=i[r]||function(){ (i[r].q=i[r].q||[]).push(arguments)},i[r].l=1*new Date();a=s.createElement(o), m=s.getElementsByTagName(o)[0];a.async=1;a.src=g;m.parentNode.insertBefore(a,m) })(window,document,'script','https://www.google-analytics.com/analytics.js','ga'); ga('create', 'UA-93946844-1', 'auto'); ga('send', 'pageview');

0 notes

graykuglen · 9 years ago

Text

Design of a product to measure customer experiences

vimeo

Over the years I’ve utilized many different design and research methods to evaluate and design software products. I designed and built a product using one of these methods to measure a design, but made it 60-70% more time efficient (wisepix.com). The method is great at measuring the design of a product which also mimics how we humans make decisions.

Our brains are generally wired to make decisions based on our past experiences - or our own “rules of thumb.” There are three general ways we make decisions:

with intuition, quickly by gut and in the moment with very little data

using heuristics, a few options to base your decision on

and optimized where you have all the data and everything is specified

Going on gut is less effective since its less informed, and optimization is hard since we don’t always have all the data. The product I designed and built uses heuristics with a few options or “rules of thumb”.

Using heuristics is pretty flexible and can be used for many types of situations. For example it can be used for the obvious design choices like websites, web applications, or marketing landing pages. You could also apply it to specific activities like sign-on (authentication) or within a specific domain such as financial services, healthcare, or customer service. If there are good rules of thumb to leverage of what we know works, we can measure an experience. I built the product to start to measure the omni-channel experience for in-person experiences and digital.

WHY DO THIS?

Software development and deployment is costly. Any issues you may catch early are 1/30th of the cost compared with after a release.

Establishing goals up front (rules of thumb) can be a great benefit to the team and educational for those who are not designers

It can be done very cost effectively by experts, or those that know the domain of the product and the “rules of thumb” to measure by.

There is no substitute for putting your product in front of users (primary formative or summative user research), however utilizing experts (in addition) will give you actionable recommendations on best ways to address an insight. Users won’t do that.

A view of the Chrome extension in a Chrome Browser - view more about the product here http://wisepix.com

PRODUCT VALUE

Collaboration where multiple people do the evaluation (2+)

Flexibility, allowing users to create their own rules or guidelines to be measured which makes it useful for many applications

Generates an executive summary of the findings for you, no more formatting

Simple, push-of-a-button to export findings for sharing

Can be used to measure mobile and desktop products, as well as, in person experiences from photos.

The product will be available the first Quarter of 2017 in the Google Web Store, where anyone can install and setup a free account. Currently it is available to do evaluations, but privately ([email protected]). To register to use the product in advance goto wisepix.com.

0 notes

graykuglen · 9 years ago

Text

Law Enforcement: Using technology to rebuild relationships with communities

Law enforcement organizations need to rebuild their relationships with the communities they serve. Technology has come a long way and can greatly aide them in this effort. Let me explain.

While working with a client recently I was tasked with exploring the ‘citizen experience’ to report non-violent crimes to law enforcement - file a report. The client was looking to apply their technology to create a report, rather than the current wide usage of HTML forms. In the end I prototyped an experience to gather reports (bike theft) using a CHATBOT and defined some personality traits to be infused into the interactions.

[play with the bot I built below]

MY PROCESS

I recruited and interviewed 5 people who had reported a non-violent crime within the last 14 days

Audited and evaluated 5 different precinct/districts in the states of California and Oregon

Spoke with an expert familiar with law enforcement and the process to manage incidences

Reversed engineered the process for one precinct and one crime to look for opportunities for improvement - can this be done.

DETRACTING RESEARCH INSIGHTS

The interviews were over the phone with open ended questions probing for more information about their experiences. In summary my findings are simply that there’s a need to improve this process for citizens, it’s not perceived as helpful, and may even damage perceptions of efficacy. Here’s some top-level findings...

Most users were dissatisfied with their experience to file a report - citing communication issues, instructions, and lack of “what’s next” information. [DISSAPPOINTMENT]

Most also were told that someone would follow-up with them and that never happened [CREDIBILITY]

Enough of the participants (I interviewed 5) shared that they would not file a report if this happened to them again, [TRUST] unless compelled to for insurance or compliance of some sort.

This to me is a missed opportunity, one where the Police could build a better relationship with the community without a tremendous effort. A 2015 Gallup Poll shows the lowest confidence the the Police in 20 years, with 18% having little or now confidence (highest ever).

THE CASE FOR BOTS

This is not an exhaustive investigation, but provided more than enough evidence (from my 2-3 days of investigation) to conclude there is a need and a bigger opportunity. Current online forms just don’t measure up. Technology is not always the right solution and may create problems. This is not the case here and it’s an ideal example of where technology can do good. Why?

Bots are simpler to fill out and relieve burden from the end user - they no longer need to know as much about the types of crimes

They can be more efficient for creating a crime report (than filling out a form), which means less time on task to complete the form

Technology can provide a human-side to the law enforcement that reflects traits the public hopes to encounter, albeit artificial, its much better than a form.

Bots may be deployed in more areas including mobile which would impact outcomes

Lastly, it may even influence policy and baseline requirements forcing departments to re-think and analyze how they interact with communities i.e. what needs to be said, how it’s said, what works etc...

Above is the bot experience I created using Motion.ai to prototype the workflow of creating a report - below is that actual bot you can play with.

When automation hits, and it’s gonna hit hard, BOTS do a better job at (and more affordably) gathering information, taking reports, or answering questions. One additional aspect to consider here - police officers are not necessarily the best at interpersonal communication with the community. Just Imagine now each officer could have a representation of their digital self or, even their own personalities which could be configured to build a better relationship.

PERSONALITY

The above script was derived from the required process flow a person would walk through but now I overlaid personality into the response to guide and support the person filing the report.

Staying on track here...we’re talking about using technology to eliminate the process of taking police reports. Sure, if a bike’s stolen it’s gone and the Police are not staffed to chase down misdemeanor crimes. Technology will not directly change the outcome of your bike - it’s gone. However, any added negative experience from an online form, or police station personnel can only lessen the community perception. Even the small negative experiences add up to deteriorate people’s mental models of the Police. Participants came away from my interviews questioning a crime report’s value. That’s a problem and one that alters the perception.

DESIGN THINKING w/BOTS

Providing the ability to file a report online is fantastic (current technology) but it does NOT have to be perceived as a futile effort. At first I was skeptical about Bots - machines that offer un-manned communication, however, automating the mundane and recurring tasks is what machines are good at. Take a look a the example prototype I created (below).

0 notes

graykuglen · 9 years ago

Text

1st generation collaboration software design for large display walls

We were hired by Jupiter to develop an entirely new product which would end up altering their market - large display wall imaging and software. We provided the design strategy, and experience and visual design of the new product named Canvas™. This new product category would help them win awards and drive sales.

We worked with Jupiter to create the first mobile application to collaborate with situation rooms. Jupiter makes the best hardware and software in the world to run large video walls and was recently acquired by Infocus. These large video walls exist in situation or control rooms worldwide to display live information graphics or video for monitoring and managing information. Some example Jupiter clients included: the centralized Bejing Traffic Bureau (in China); or at CAISO (California Independent System Operator) where moment-to-moment energy consumption is monitored for the entire state of California. These rooms are where the information is and where status and response occurs.

“Congratulations to you guys and thank you. I don’t think it was just the company that changed this week. I think we may have changed the trajectory of the industry.” – Brady Bruce, Vice President, Marketing & Strategic Alliances

The demand to effortlessly share information and collaborate in real-time is increasing. Rather than have situation rooms be a closed environment (common at the time of the product development) Jupiter was looking at an entirely new open solution to redefine how different teams may interact from anywhere and not just the situation room. This new product would offer teams in different locations the ability to share any information in real-time and also to write on it, label it, draw and erase it – “Now you don’t need to be in the room to share and participate”, says, Brady Bruce, Vice President of Marketing & Strategic Alliances

People can now gather in different control rooms or from their desktop or even via mobile, to see the same things on or off the wall(s) from anywhere they're connected with the Canvas™ software product.

“I had 14 press meetings and all of them said the user experience was better than anything else they had seen in the industry, including display wall solutions, switches, and any other related product.” – Brady Bruce, Vice President, Marketing & Strategic Alliances

At the end of the project and at release, Jupiter was able to demonstrate the product at InfoComm, which is ‘the show’ of trade shows for the commercial AV industry. At the conference Jupiter summarily impressed their customers and competition with a demonstration of Canvas.

“By noon on Thursday, our booth had been visited by the senior execs of all of our largest competitors asking for a demo. I’ve never seen anything like it” – Brady Bruce, Vice President, Marketing & Strategic Alliances

Canvas™ was the talk of the conference showing how to write/draw/edit live on any wall or screen using their new collaborative multipoint visualization software. They also received a Best of InfoComm Award for Canvas™. At the time Jupiter has defined a new product category for they’re industry and were aggressively competing with much larger international companies. As mentioned they were acquired by Infocus after the product hit the market.

0 notes

graykuglen · 9 years ago

Text

Collaborative site redesign using design sprints

I worked with the fine people of Tintri to help in a redesign effort for their website and to improve their content effectiveness and sales funnel. They were looking for a full site redesign by an upcoming conference to help them communicate the value of their product. I proposed Design Sprints with a cross functional team - internal and external. We delivered the planning, strategy and concept design including concept testing, and they took on internally the detailed visual design and delivery/development. We worked as a team all along the way.

Their products optimize storage for IT and arguably is the best product on the market. It does such a great job and is still so new to the market that nobody believes what it does even after multiple independent studies. We helped with that too.

We introduced them to our approach and utilized design sprints. We worked with the team to define who their customers are (through workshops), created journey maps for each key customer type, and began a rapid approach (sprints) to redesign their site and put it in front of customers for testing and experts with heuristics.

How do you do redesign a site with hundreds of pages and years of content in a Design Sprint? Well, you don’t and can’t actually. But you can chunk up the work into multiple Design Sprints working toward the same goal. That’s what we did. For a design sprint and to be effective, the planning is critical. So some may be able to do a Design Sprint in a week (read Jake Knapp’s great book called SPRINT), but this was not one of those projects. Always plan ahead and do your homework which is critical.

FINDING: If you are planning to perform a usability test as part of a Sprint, clearly identify who your customers are in advance of the Design Sprint. It takes a little back and forth discussion amongst the team to elaborate on who they are and also who they are not. This is important so that you can begin recruiting. If you do plan on recruiting be mindful of the quality of your participants. If you have very specific needs, it may be better to enlist the help of a recruitment firm and make sure you allot at least two weeks before testing – see what I mean about planning! With this project we were dealing with specialized and technical users of a product, not someone you can get on a list board or through friends and family.

Overall Site Architecture

One of the many activities and part of any redesign is cleaning house. We had multiple concepts for the information architecture which were informed by our stakeholder interviews, site traffic and end goals. Though this is not the prettiest of diagrams (above) it communicated to the team the evolving new structure, and was baked into the evolving prototype making it shareable for the team.

We did modify with the overall site structure and content utilizing the journey maps we had created for the targeted users. Our focus was on the desired interactions and pathways for these users, but also taking into consideration the user’s needs for content and support at the different points along the sales cycle. Keep in mind that this was not a consumer product you’d place in a cart to purchase, this is an enterprise software product costing many thousands of dollars with a longer selling cycle. Each sale could take months depending on the procurement process for each customer.

Design Sprints

Once we felt we had good design-data we began prototyping our experience for sharing with the team and evaluating the website in a responsive manner. This is very important step since we are really beginning our explorations into what it would be ‘like’ and how it would behave for customers. I believe Prototyping is a fundamental step to getting the experience right. First it helps the designer work through the details, and second it becomes a communication tool for all members of the team.

(Above) Output from a Design Sprint, these are the sketches from the team highlighting individual preferences.

Prototyping & Heuristic Review

With the prototype (30-40 screens) and overall interactions underway we tested the design with our targeted users. We also had several UX pros evaluate the product using a design review with heuristics (wisepix.com) who were not involved with the design efforts from the team. We did this to ensure we had an outside perspective on the overall design we had chosen. Both the user research and the design review’s revealed some key findings that we had overlooked.

FINDING: If possible use a prototyping tool that more than one person may contribute to the design and parse out the work - we used Axure. Having said that, use a tool that efficiently allows multiple people to have a versioning system to check-in and out different areas of work. We had three interaction designers working collaboratively to divvy up the work.

We went the extra mile in the redesign and built out as many screens as possible to demonstrate the overall interactions and flow of the site. We would NOT doing the development work and clear communication of how the new site would work with it’s interactions was critical to eliminate any confusion later which could be weeks later.

FINDING: Plan for multiple iterations to improve and critique the design and pages. Keep a realistic hit list of changes and why they were suggested and agreed to. The further away you get from the design decisions the more hazy your recollections will be. We were not perfect, but we could document key decisions and why they were made so not to waste time.

http://www.tintri.com

#design sprint #site redesign #user experience strategy

0 notes

graykuglen · 9 years ago

Text

Transforming the US Postal Service

Well known enterprises today are under heavy pressure. New technology and business models are challenging their existence and hastening their demise. What once worked to produce money may now be a liability. An example is the US Postal Service which had a net operating loss in 2015 of 5.1 Billion. Reducing costs and changing laws will only take it so far, so what’s next for the Mail-person?

The United States Postal Service has to change. But, not only that, it has to TRANSFORM itself. Not an easy task for a large organization. Their bread and butter - first class mail - has been steadily declining. The world is going digital, so what differentiates them and what can be delivered digitally?

A WORLD IN FLUX

The USPS must change or cease to exist (less physical mail means less revenue) with postal mail decreasing at about 7-10% each year. The future growth is obviously digital, but how? Above we have examples of two changes occurring simultaneously - one depicts the shrinking volume of 1st class mail (in gold above), the second shows the explosive growth of online advertising. The post office delivers mail to citizens, can they also deliver digital advertising to citizens?

DIFFERENTIATION

What's so unique to the USPS is that they can deliver to every household and business in the United States and do it efficiently. Think about that, every business or residence in the U.S. each has a unique address but why not have a digital address to represent the physical as well? This would create a new channel for delivery and management of mail and messaging for the future. Every house could also have a unique digital address, maybe even reflecting in some way the physical.

CURRENT PHYSICAL DELIVERY

FUTURE DIGITAL & PHYSICAL DELIVERY

There would still need to be delivery of physical mail in a 2-day time period and the workforce to support it. However, the workforce should be reduced to meet obligations and market. They would need to be downsized of course, but with new opportunities in digital there could be a retraining, something our country sorely needs to meet the needs of our changing economy.

MARKETPLACE

Advertising is about targeting the right people with the right offers to demonstrate value and sell products or services. There is a considerable amount of physical mail that goes unwanted resulting in a waste of resources. What if the USPO could leverage their delivery knowledge bur reverse the flow. Customers now go to the USPO (online) to find the deals or even opt-in to the right digital delivery of advertisements offered locally - waste could be reduced with many benefits across the board. The big point here is that the post office would be an intermediary between merchants and customers who may opt-in or opt-out of the right offer. Each customer has a digital version of their home address with the post office. They just simply access the mail and offers online and not wasted in the mailbox.

AN EXAMPLE

I need a step-ladder for yard work. I could go online to the Google and do a search which would result in a list of retailers (hopefully) to purchase a ladder or wade-through the malaise of pages and links and offers etc. What if I just entered I was looking for a ladder and the offers and promotions came to me from my digital address through the USPS? I could easily connect with the deals and offers within my geographic region.

Below are some examples of building blocks for evaluating a business model. I’ve only added a few and have not dissected the current USPS model to show in comparison, but enough to communicate more of the idea.

VALUE PROPOSITION

The USPS creates a marketplace where I can go to find the local offers when I need and not stuffed in my mailbox and trash bin. Junk mail is not in my mailbox, but now sortable online - it’s there if I want/need it. Sure, I can do this with Google, but that’s honestly work with so much noise and dead-end paths when I’m trying to find what I need.

For businesses looking to reach customers they are offered a less expensive option to reach customers. They save costs on printing materials and are provided better tools to find and target new customers.

CUSTOMER SEGMENTS

The USPS now begins to strengthen the relationship with all U.S. consumers and businesses. The different segments determine their needs and opt-in to share. People may even be offered anonymity and share more demographic information to help businesses attract the right and desirable customers. There is no need for inferring from online behaviors when someone simply raises their hand and says “I want that”.

CHANNELS

The channels now expand for the USPS to include home delivery AND online. The USPS offers an online instance of your address to extend delivery to every household via post AND digitally. This saves resources in printing and opens up new revenue streams. Charge vendors less for commercial mail, offer better targeting as customer segments opt-in on what then need or want.

REVENUE STREAMS

The Post Office is currently constrained and desperately trying to find new revenue streams. They are a trusted and reliable service - build on that. The new options for a digital channel to deliver mail, promotions etc. are vast. They could create new and more affordable commercial rates for businesses to find the right consumers. Options to create revenue from consumers could be through a host of services around digital mail delivery or for new features to help manage and store digital offers or personal correspondence that is never printed.

CONCLUSION

With the Postal Accountability & Enhancement Act in place the Post Offices is handcuffed and unable to make the right business decisions to stay afloat. Postal worker pensions are a serious liability and they need to dramatically reorganize their workforce. They can only reduce costs so much before running out of runway. Well, actually, they ran out of runway already and in 2015 they operated with $5.1 Billion loss.

What I’m proposing here are big directional concepts that would require research and planning and cost millions to implement. But definitely better than offering new tools for junk-mailers to target you with unwanted advertisements which go wasted and have minimal conversions, which is what they are doing now. This is indeed a very BIG IDEA and possibly not original, Hey, Kodak knew of digital photography and had a part in it’s creation years (maybe decades) in advance of their demise. The world is changing more rapidly than the Post office so they need to alter course. The cost of such an endeavor would be high and take time, however NO change is inevitable defeat.

0 notes

graykuglen · 9 years ago

Text

Healthcare mobile app and platform design

I was hired by AppMedicine to create a mobile and responsive application to improve pre-operative procedure preparation – all the things that need to happen before and after a medical procedure. I led a small team to conduct and deliver the user research, strategy and experience design for this new product.

The product would later be called eProtocol. Before beginning we created a detailed list and understanding of the project goals. Sometimes it’s easy to jump right into solutions with the client or team, but this can be very troublesome and add confusion. One of the goals was to make this a multiple user system (two products) one for healthcare providers to monitor patients, and another for patients to engage with providers. We aligned the value/benefits between the two these two user types. Multi-user systems can get very complex fast.

We recruited and interviewed eight former patients who had recently undergone this type of procedure. Our first goal was to understand the current pre-operation process since we will be asking the end users of this new product/service to modify their behavior. The interviews were recorded and the notes and findings summarized into a Journey Maps for sharing with the client team.

When working with a Journey Map, we want to summarize the information we do know to be representative of the many people we’ve spoken to during the interviews. One Journeymap is sufficient for a unique group of users. In our case we also added a standard graphical model (at top) of origination - or where patients started into the funnel from – some came from primary care, some came from specialists etc… These entry points were very important (depicted it at top of the Journey Map) since these are the first touch-points of the patient experience we were modeling, but improving upon.

Application Architecture

After research we looked at other applications and frameworks across industries. We were also looking for parallel experiences which may be adapted to the needs of patients. In our case and for this procedure we were looking at an older audience and that had big implications on the approach to design. From our research there were enough participants (and from 3rd party research) that did not have the latest smart phones and were not super savvy users – the application had to be simple with as few screens as possible.

Once the team was in agreement of some directions we filled in the blanks and began discussing content requirements, screen navigation (Android and iOS), interactions for different features, and messaging for patients. We were to design two applications, one native app for iOS and future version for Android, then the larger Provider application view to manage the patients.

Prototyping and design

We began prototyping our experience to share with the team. This is very important step - we are beginning our explorations into what it would be like for patients and providers. Prototyping is a fundamental tool we believe to getting the experience right. First it helps the designer work through the details and learn early and often, and second, it becomes a communication tool for the team.

Detail design

With the prototype and overall interactions established we began on the details for the platform – in this case we started with iOS. By the way, we did have some challenges on the project including new management and technical disagreements which altered the design further along. This forced us to modify some of the application – it happens.

We used vector graphics for icons and the user interface elements we included in our UI-Kit. From our experience Sketch was built from production perspective so it is fantastic at producing what you need for mobile development, in other areas not so much.

A UIKit is a combination of many of the graphical elements. It’s a great resource for getting a one-stop view of the different patterns and designs. We consolidated the iconography and translated screens for various applications into one Sketch file. After the assets, we worked with development to implement and modify. In the end the client was not successful and closed doors about a year later.

The product offered the ability to save costs for healthcare providers through introducing efficiencies. This reduced cancellations that many doctors and practices were scheduled 4-6 weeks out - time and money that would be lost. It was also for the patients, research showed that the better a patient is informed and prepared, the better the outcomes of the procedure.

#moblie #interaction design #user experience #strategy

0 notes

graykuglen · 9 years ago

Text

Design of a 1st Generation Internet of Things (IoT) product

Interaction and Architecture Design

I lead and designed the first generation product to connect mobile wireless cameras with sensors for Netgear, Inc. They selected my company to assist them in the redesign of Vuezone’s mobile application to view-manage-wireless cameras and motion sensors. After the the redesign the Vuezone was rebranded and is now known as Arlo.

Design and prototyping

Netgear re-architected the platform and was adding new functionality to make way for smarter in-home features like motion sensors, infrared vision and video recording. These camera sensors would be the first of many new capabilities for Netgear as they ventured into the Internet of Things (IoT). Now a camera motion sensor could possibly activate a light, or send a notification. Our role was to design the mobile phone and tablet user experience for iOS and Android, as well as to plan out the responsive web application and visual design for all.

There were multiple systems for registration, user identification and new systems being developed for promotion that needed to be understood. So….it’s really critical to get the data out of people’s heads and onto paper that is shared and spelled out. Logic flows were useful for the designers to learn, but critical for the team to comprehend.

I prototype for just about every project now. There are many things to consider before you decide to build a prototype for your project like who is the audience for the prototype, what is most important to prototype etc.. For this project we were mainly working with an internal team at Netgear and was not to be shared outside the group, which made things easier (less polish). For sure it’s added work to prototype, but we find it’s critical for the team and us and the key is knowing when to do it. I’ve made many mistakes prototyping too early which created additional work and time. Prototyping is critical for several reasons:

The act of prototyping is also one of deep learning. It forces designers to go through the process of modeling the experience including discovering problems and making mistakes early

Great for sharing and communication with a team. It’s critical to get everyone looking and understanding the same thing. There’s no way to talk around anything anymore, it’s right there for the team to discuss, debate or wrestle over.

It helps reduce the documentation since now the prototype is the document and better describes an animation, or a behavior, so much more so than a paragraph of text with a pretty picture.

Interaction and Architecture Design

There were many stages of detail we went through. We mostly worked in iterations of functionality and screens to represent the critical areas of the experience. We were designing for both iOS and Android and started with iOS and worked from there simplifying against the platform guidelines. As an example we chose to remove many gestures and go with the most widely understood and avoiding gestures that seemed gratuitous at the time.

I’ve made the mistake previously to create too much documentation - it just is a wasted effort. We tried to focus on the experience and details with a prototype to be created/delivered for offshore development. We translated the design into Android after iOS. Where necessary we did provide side-by-side comparisons to speak to iOS vs. Android.

The intention at the beginning of the project was to provide a fully responsive HTML4 product, in addition to the native iOS (before Android would come after). We designed for the small screen first on an iPhone, mindful of the screen sizes we’d be accommodating. We also provided layouts for the different resolutions (or media-query sizes) for key screens of the application.

Detail Definition & Visual Design

The visual design was a separate project. Netgear had just completed a rebranding effort corporate-wide and the design elements which would influence the project were being created. On the visual design we utilized what we could, but the rebranding effort was focused mainly on print and web with very little for the mobile interfaces or iconography. We worked closely with the Marketing team to review and approve our visual design language additions and also created visual style and interaction guidelines for the company.

To ease in the handoff efforts after design a UI library was created which included all the icons and icon states for the mobile and tablet apps. Symbols like motion sensor states and colors, settings and signal strength were shown. This was the graphical index for the team and a very powerful tool. We were fortunate to have an excellent team and this included visual design. We would have preferred building a live index page with interaction/hover states and links to all the icon assets for development, but there was not enough time.

0 notes

graykuglen · 9 years ago

Text

A model to gamify your product or service

For those that are not familiar with this concept to “gamify”, over the last 6-8 years businesses have been bringing the psychology of game theory to their business applications to drive usage through improved engagement. I augmented an existing model of game mechanics in an attempt to describe how any business could truly align their business goals with end user goals.

On a recent project I heard a client make similar requests repeatedly to “gamify” the experience of their B-B-C (business to business to consumer) platform for crowd sourcing. I realized shortly thereafter that they did not know what that meant (or could articulate it) and I was unable to clearly communicate how to go about building a process to do just that. Without a framework and plan I’m afraid we’d end up wasting valuable time. I dove in for a little while and this is what I came up with which we utilized.

FIRST

Leveraging the exceptional work by Social Scientist B.J. Fogg and his Fogg Behavior Model I modified it to better align with a business perspective. This is not a writeup of Game Theory. My intention is to communicate the model and process, not go into too much detail about game frameworks or mechanics - there are plenty of great books and more qualified people to speak to it than I. The Fogg Model has three parts to it that occur in order for a person to perform some behavior. Understand these, and you have a model to guide you on how and why people do things.

Trigger - what prompts the event or desired behavior

Motivation - why does someone perform the behavior

Ability - is the person able to complete it

If you can successfully preform at least 2 of the 3 above you’ll increase the likelihood to have the user perform the desired behavior. This is fantastic, but from a design perspective I needed to look at it differently.

SECOND

I chose to modify the labels to better align with a design process and so now it becomes = [Motivations + Actions + Feedback]. The Fogg Model still stands and is better at describing the behavioral phycology, but for a team of engineers and business people to contemplate and understand their goals of what they want to happen I chose these different labels. Then, I butterflied the model (flipped on horizontal axis) to reflect the business side, as well as the direct user (see illustration above). After all, we want business and primary users motivations to align.

USER (Left)

Motivation1 - what motivates the user to perform an ‘action’

Action1 - the action performed by the primary users

Feedback1 - how will we reinforce the action with feedback

BUSINESS (Right)

Motivation2 - what is the motivation of the business to have the user perform this action

Action2 - what does the application/process/platform need to do after the user performs this action

Feedback2 - what does the system/business do to record/reward/communicate this action was performed

If you’ve ever laid out an application with multiple user types and interrelated interactions it can get really complex quickly. For this client there were two customer types - the primary user of the product and the business customer.

HOW TO USE

So, to keep things simple I usually prioritize the most important aspects that are key to the success of a product first - for example what 3-4 tasks/activities we absolutely need to get right to succeed here. Then we needed to get the overall game mechanics agreed to - are these points we’re giving out, reputations we’re building, do we want people to share...etc?. A framework describes the broader aspect of what you are looking to build. For example is this a collaborative expert audience where reputation is more valued than points, or a competitive one where users go head-to-head to attain top ranking with awards and a leaderboard with ability to redeem points? Each would involve a different set of mechanics and approach.

When we identify the top activities that will have a high frequency and importance to all parties (top priority), we use the model to guide. Below is an example of one scenario of a crowd-sourced activity to take a photo and report a problem within a community so that it could delt with. In this example there are two direct primary users interacting (Mary & Ignacio) with the system. This is just an example, you can substitute the scenario.

0 notes