Scott Cook, founder of Intuit, talks about innovation at CHI 2006

Scott Cook opened CHI 2006 with a plenary on fostering innovation. Live-blogging notes have been posted on the CHI Blog. Here are my notes, which are not a word-by-word report on the talk but more of an annotated reconstruction (quote at your own risk).

Creating Game-changing Invention

The brief –  Innovation happens at the junction between business and customer needs, not from executive ideas or lonely geniuses within the company. Indeed, innovation bottlenecks are often at the top. Creating a culture of innovation is about nurturing customer observation, incubating new ideas, celebrating failure, and staying out of the way. Read Thomas Kuhn (The Structure of Scientific Revolutions) and Peter Drucker (Innovation and Entrepreneurship).

Intuit: Changes lives so profoundly people can’t imagine going back to the old way.

Examples of innovation

Cook gave some examples of innovative business ideas and showed how they needed both deep understanding of an unmet customer need and really good innovative solutions.

Twenty-five years ago, Guy Laliberté had the idea to create a new form of circus entertainment and founded Le Cirque du Soleil. There was no business sense in the idea, according to the market wisdom of the time, but Le Cirque du Soleil turned out to be one of the largest cultural export from Canada. “They were able to do this because they invented a new game.”

In the mid 20s, 3M produced sandpaper. At the time, Richard Drew was working at 3M as a product tester on a new type of sandpaper. In the same period, two-tone paint was becoming popular. Drew was visiting a car shop when he heard complaints about how hard it was to get clean lines between the two colors. The tape available at the time was too strong and tended to ruin the first layer of paint. Drew thought there was a business opportunity for a “wimper tape.” He presented the idea to his boss, who didn’t see any value in it. “We make sandpaper, not tape!” he declared. Drew worked on this new idea without support from his company and invented the masking tape. Few years later, he invented cellulose tape, also known as scotch tape. (Read more about Richard Drew.)

The bottleneck is always at the top of the bottle.
– Peter Drucker

Five models of innovation

1. The lone genius
2. The boss is a genius
3. Copy competitors’ inventions
4. Cluster the geniuses in a lab
5. Make your people the geniuses

Scott Cook prefers model #5: “This is what really works.” None of the other models are sustainable or effective in creating a culture of innovation in the long term. Models 1-4 tend to create pockets of innovation that are limited and isolated from the rest of the company.

The source of invention is unlikely to be the big executive: innovation comes from where the business connects to customers. It scales, because it allows the company to create many “contact” groups between the company and the customers.

Five principles of corporate innovation

1. Inventions comes from mindset changes
2. Mindset changes come from seeing things differently
3. Savor surprises… as learning
4. Focus managers on a customer metric
5. Nurture and protect innovation teams

1. Inventions come from mindset changes

In the 40s and 50s, innovation in cargo transportation was all about ship speed. Companies were racing to build faster and faster ships. In 1937, a trucker from North Carolina, Malcom McLean, was observing people unloading his truck on a dock in New Jersey and realized that the largest delay in shipment was in loading and unloading goods from and to trucks. He realized that the truck trailer itself could be loaded on the ship, with great savings in loading time. This was the seed idea for the invention of the container, which McLean started producing in 1956. Containers allowed a 35-fold reduction in cargo loading cost.

The invention of the container originated from a mindset change: McLean didn’t focus on the speed of the ship, as everybody did at the time, but on the speed of loading and unloading ships.

Thomas Kuhn, in The Structure of Scientific Revolutions, talks about paradigm changes in science. This is how innovation happens: there are these young scientists, usually under 30, who introduce a new and revolutionary scientific paradigm. Nobody believes them at the beginning, but the new paradigm works and eventually changes the way we see the world.

In the industry, this means creating products that change lives so deeply that it makes it impossible to do things the old way.

Cook talks about the initial failure and amazing success of Quickbooks. Quickbooks was launched in 1992. It had half the features of competitors’ products and was twice as expensive. Intuit ran an horrible ad to promote it. The product was buggy. Not small bugs: the customer would enter business data for months and one day all the data would suddenly disappear. Supporting calls were going through the roof.

And then, suddenly, Quickbooks became the best selling product. Out of nowhere. How did that happen?

A survey showed that many people used Quickbooks for their business even though the product was designed for home use. After ignoring this piece of data for a while, Intuit started to investigate why people where using Quickbook to keep track of business expenses. The study revealed that small businesses very often don’t have an accountant and owners hate to keep track of expenses in “in” and “out” columns. Quickbook worked for them, and they made the product a success.

2. Mindset changes comes from seeing things differently

Empathy is not walking in another’s shoes… First, you must remove your own shoes.
– Indian proverb

The problem is not how to get new, innovative thoughts into your mind, but how to get old ones out. Every mind is a room packed with archaic furniture.
– Dee Hock, founder of Visa

Typical development process

1. Define requirements
2. Design solution
3. Involve UX
4. Find customer problems

Intuit Process

1. Involve UX
2. Find customer problems
3. Define requirements
4. Design solutions

You can look at Intuit marketplace position and feel very satisfied. Intuit market share is very good as compared to the competitors. If you consider people using other methods (manually, with spreadsheet, etc.) for keeping track of their finances, however, the penetration of Intuit products is not that impressive anymore.

There are 5 types of taxpayer types. Only two are targeted by TurboTax (one of those is the “Self-directed”). One of the groups that is not a good match for Turbotax is the “Worry warts.” These people want to be in control of their tax return but are afraid to make mistakes and don’t trust themselves. They are the ones that use H&R Block: they are looking for a product that comes with an “expert included in the box”.

Intuit created another version of TurboTax called TurboTax Personal Pro. The team started with a concept product (a storyboard). First, the customer fills out the tax form using a combination of web interface, avatar, and chat with a tax expert. If customers have doubts they can leave the field empty or chat with a tax advisor. Then, the client calls the tax preparer and goes through the tax return to complete information and fix errors. This product is still a prototype but it seems to be working well for these types of taxpayers.

3. Savor surprises… as learning

Victoria Secrets hired a consultant firm to conduct a study on their customers. The consultant went to the houses of Victors Secrets customers and looked in their dresser’s drawers. They noticed that many of these women did not wear Victoria Secrets lingerie. They asked why and discovered that clients did not consider Victoria Secrets lingerie comfortable to wear. After this finding, Victoria Secrets introduced a new bra that was designed for comfort: it was seamless and had broad soft straps. Victoria Secrets’ market share in that sector went from 25% to 45%.

“Follow people during the workday, listen carefully when they call, go to Town Hall meetings, create decks in the halls with customer information.”

Peter Drucker and the seven sources of all innovation

1. New technology and scientific findings
2. Changes in public perception
3. Demographic changes
4. Industry market and structures
5. Process needs
6. Incongruities
7. The unexpected

The most important source of innovation is the unexpected, the least important is new technology.

The insecticide Raid was created to kill insects by poisoning the nest. The bug that has been sprayed with Raid goes back and bring the poison to the nest, killing most of the bugs living there. But when Raid observed customers using the product, they realized that Raid users sprayed much more product that they had imagined, drowned the bug in the insecticide and killing it on the spot. This behavior made Raid quite ineffective because it allowed to kill only one insect at the time. Raid reformulated the product to have a much lower dose of insecticide so not to kill the bug immediately even in high doses.

When studying Quickbooks use for businesses, the Intuit UX team came back with three findings:

1. Accounting was the problem
2. POS is custom installed for $4000+
3. Intuit sells the software

The team concluded that to make Quickbooks more appealing to small businesses, Intuit had to offer both software and hardware (credit card swipe, recept printer, and bar code scanner).

Scott Cook said: “Hardware? No, way! Were are a software company” (Remember, the bottleneck is always at the top of the bottle.) But the team did not give up. They looked at all the objections, one by one, and found answers to all of them. Quickbooks shipped a version with software and hardware that became very successful.

Creating an environment for innovation means that there will be a high failure rate. But the few successes keep the business strong and competitive.

At Intuit, we celebrate failure. Literally: Intuit has a greatest failure awards. Because every failure teaches something important that can be the seed for the next great idea. So, it’s important to celebrate the learning that comes from failure.

4. Focus managers on a customer metric

Everybody is big in financial metrics, but lousy in customer metrics: no customer metrics are identified, or too many customer metrics are monitored, or they lousily defined. To nourish innovation, customer metrics need to be as strong, well defined, and decision-driving as financial metrics.

An important customer metric at Intuit is how well products do when clients are asked the question: “Would you recommend this product to a friend?” (on a scale from 1 to 10). People who rates a product 1-6 are detractors, 7-8 are passive, and 9-10 are promoters. The percentage of net promoters can predict almost perfectly the 5-years revenue growth of a company. (Scott Cooks had several graphs showing a very high correlation between percentage of net promoters and 5-year revenue growth in various industries.)

The success of a business is when your customers are happy and tell their friends about it.
– Andy Taylor

When Intuit looked at tiny businesses, they realized that Quickbooks didn’t work for them and created a new product targeted to very small businesses, Simple Start.

5. Nurture and protect innovation teams

When people work on innovation, they take more risks and they need empowerment to be successful. One way to nurture and protect teams is allowing them to blur of the boundaries between disciplines. “Great teams don’t have barriers, they do each other work and collaborate.”

Keep management from interfering with teams. Senior management tends to ask too many questions too early. In each phase of design, there is only one or two relevant questions, and this is all should be asked.

Cook ends talking about Intuit Medical Expense Manager. The idea for this product came to an employee whose child had cerebral palsy. In the first year, his medical bills topped a million dollars. Although the health insurance covered most of the expenses, dealing with medical bills and claims was a nightmare. It was clear that there was a need for a software that could help people going through an emotional and difficult time dealing with such a complicated administrative chore.

In the Q&A session, Scott Cook talks about simplicity, the importance of studying customer behaviors rather than listening to their wishful thinking on their behavior. How to change the culture from the bottom-up? The important people to convince are those who work on the project.

There is a problem when large sets of quantitative data are used to reaffirm current thinking within the company and to kill innovation. This happens especially with surveys, because the way questions are asked in a survey reflects the current thinking of the company. We should never ask customers what they want, we should use customers to understand what the problem is. Customers are not going to provide the solutions; solutions need to be built by a skilled team that has a deep understanding of the problem.

At the end of the talk everybody applauds and I wonder: how does it feel to have your boss saying all the right things?

[Update: Check out a good and more literal report of Scott Cook’s talk by Sébastien Paquet, together with other CHI 2006 session blog-scripts]

Tags: CHI2006, computer-human interaction, conference, Montreal, innovation, Intuit, Scott Cook