Lipstick on a Pig Session Notes from Web Visions 2010

Lipstick on a Pig
Can UX Differentiate a Software Company?
Gene Smith
@gsmith
  • Story of transformation from user experience design to product design
  • Retrospective of what worked and didn’t work
  • Wanted to transform from a services company to a licensed product
  • Great user experience team, great development lead, great development team
  • Idea: we could bring web 2.0 design to Microsoft Sharepoint
  • You could go to one spot, see all your sharepoint sites, docs, etc.
    • Sat on top of sharepoint
    • Everything designed to be very simple to use, straightforward, very easy
  • Mint.com
    • Founder had used Microsoft Money and Quicken, found them to be very tiring to use. A lot of work to enter all transactions.
    • Mint is a thin user experience layer on top of Yodalee. 
    • Had:
      • Engaging experience
      • Clever solution to common problem
      • Amazing Timing (right at recession, people needed tools to manage money)
      • Way to make money
  • Flickr
    • Relative to competitors, was a step above everyone else
    • Interesting social features was novel
    • Flickr became popular just as digital cameras were booming
    • Before flickr, people were sending an email with 25 photos attached.
    • Pro account had unlimited storage
    • Had:
      • Engaging experience
      • Clever solution
      • Timing was right
      • Way to make money (pro accounts)
  • Slideshare (bit of counter example)
    • People tweet about how hard uploading, annotating, and sharing presentations is with Slideshare
    • It meets most of the criteria (engaging experience, clever solution, etc), but the experience isn’t quite there from an engaging experience
    • However, Slideshare reaches over 28 million monthly people. It’s a top site, ranked  in the top 400 sites in the US.
  • Getting Starting...
    • “What if we built BaseCamp on Sharepoint?”
    • This turns out as a terrible way to start an idea
      • Will: I don’t believe this.
    • Sharepoint is a billion dollar product today.
      • Sharepoint ecosystem is a $13B system.
      • Market that we were entering was massive.
    • And we know that companies out there were having problems using Sharepoint. You could have hundreds or thousands of sites in a network. Companies wouldn’t know what was out there. People would spend 60% of their time giving people permissions. Another company was spending hundreds of thousands on training for Sharepoint.
      • You know something is wrong when a person who uses Facebook for two hours a night needs a two day training session to use Sharepoint.
  • Product: kiiro
    • Make a product that focuses on getting your work done, instead of endless tuning of Sharepoint.
    • Make a central dashboard, so you could see everything relevant across multiple sharepoints.
  • UI Design stripped everything away... very clean.
  • First idea was to have a two way sync between Microsoft Project and Sharepoint.
    • This wasn’t so good. 
    • Project health metrics and so forth made for a very good demo, but doesn’t match how people actually do project management, which is much more messy.
  • Then decide to focus on social stuff... an activity timeline of what you and your coworkers are doing.
  • Took twice as long to build on top of Sharepoint as opposed to a regular web application like Django. Regular web apps have easier databases, security, etc.
  • We wanted to make this thing a black box. Organizations have trouble with people continually customizing sharepoint. We took most of that way. 
  • We wanted to have a very purpose focused interfaced.
  • Market Response
    • Not particularly good. 
  • Products like Mint, Flickr --> the buyer is the user.
    • Even if they don’t pay money, they make a decision to invest their time
    • The person who is evaluating the product is different than the person who will use the product
  • Enterprise market
    • The buyer is not a user.
    • What benefits does the buyer get?
      • They want hard benefits like increased revenue, decreased cost
      • And soft benefits like increased effectiveness and efficiency
    • e.g. we can save 10 minutes a day x 2000 employees x $57/personhour = $xM savings
      • Not really true, because if people have 10 extra minutes a day, they will just go on Facebook...
  • Problems...
    • It was hard to make the business case for the product
    • The Buyer, the Business Owner, IT, and the User all different...
    • The marketplace is complex: independent software vendors, system integrators, buyers
  • Finally did a market analysis of which companies would be a good match for their product. About 24,000 companies were a good match. 
    • But the next problem was diffusion: the companies were spread across industries and markets.
  • Next problem... system integrators.
    • Companies with complex sharepoint needs go to a system integrator, not directly to 
  • 1:3:5 Rule
    • For every dollar an independent software vendor makes...
    • Microsoft makes 3 dollars in licensing fees...
    • system integrators want 5 dollars to integrate the software
    • And system integrators don’t encourage companies to go directly to an independent software vendor: “oh, you don’t want that software, we could build that for you...”
  • 2:1 Rule: Marketing vs design and development
    • You need to spend twice as much on marketing vs. design and development
    • And which means that about 10% is left for UX design
  • Conversion...
    • If you have a $5K product, and $500,000 in costs, then you need to sell it to 100 companies.
    • If you have a 1% conversion rate, you need to pitch it to 10,000 companies to sell it to 100 companies.
  • Because the product was strongly differentiated... it allowed them to more easily get companies into the funnel.
  • But because the product limited customization, many companies, once they learned about the limitations, wouldn’t use it.
  • And one of the biggest potential customers would be system integrators (who would resell it)...
    • Except that because it was a black box and limited customization, the system integrators wouldn’t really use it. (their specialty is customization...)
  • Kiiro Had:
    • Engaging Experience
    • X Did not have Clever Solution
    • X Bad Timing: in the trough between SP 2007 and SP 2010
    • Way to Make Money
    • Market Opportunity (good)
    • X Market Fit was bad...
  • On the scale of fidelity and functionality:
    • I would spend as much time with low fidelity and low functionality as long as possible researching and learning.
    • When I was ready, I would go for maximum fidelity: Mock-Ups (and then add functionality)
    • As opposed to going for maximum functionality: Working Product (and then improve design)
  • Going beyond... New Version of kiiro
    • Highly customizable
    • Better Sharepoint Integration
    • Targeted to System Integrators
  • And created a new tool... Parachute
    • Backs up Basecamp Files
    • Complete HTML Export
    • Low Cost
  • Releasing something out there in the world is thrilling. User interaction design is fun, and it pays the bills, but releasing product is exciting.
Post a Comment