Archive for category replication

The 1000 Cell Spreadsheet Kills an Innovation

Posted by Tim in business models, innovation strategy, replication on 30 December 2009

A few years ago I had a consulting job where my task was to help a company figure out how to sell the waterless composting toilet that they invented. They had already had other consultants working on the problem, but they weren’t happy with what they got from them. The only constraint that I had was that the plans had to be for selling the toilets in China and India (by the way, I was thinking about adding a picture to this post, but I think we all should be thankful that I didn’t…).

You may well ask why China and India? I certainly did. The company was already active in the US, Europe and Australia with other products, so why not go there first? Their answer was the 1000 cell spreadsheet. One of their early joint venture partners had developed an incredibly detailed market analysis spreadsheet. It considered about 25 possible markets for the toilets. The spreadsheet started by taking the population of these 25 countries, and then running this number through a huge long string of calculations (% of people buying a toilet each year, and a number of similar things). At the end of all of that, the two markets with the biggest potential were China and India. Why? Because every calculation that they made was either a multiplication or division of the original numbers, with no variation between countries. So the end results (value of the toilet market) were directly proportional to the original set of numbers (population).

This is clearly poor analysis, for a number of reasons. The key one is that trying to sell high technology toilets in countries where huge numbers of people lack access to any toilets at all is a bad idea. If someone hasn’t spent US$20 on a pit toilet, why would they spend 10 times that on a composting toilet? The answer ended up being: there’s no reason for people to do that. The company that I did the report for is now out of the toilet business, and their great invention never got into the market. This is a situation where thinking about value networks and business models would have helped.

I wish I had realised it at the time, but I hadn’t started thinking about this in great detail yet – the company had no business model for their new idea. Their value proposition was basically ‘it’s a cool toilet’. Consequently, the markets that they targeted were simply the biggest ones they could find in terms of number of people. The fact that there was no value network to plug into in these markets was never considered, despite my best efforts.

If their value proposition had been ‘this is a waterless toilet’, then they would have looked for markets with limited water supplies, like Australia, South Africa and the Southwest United States. All of these countries already had value networks for water-conserving toilets that my firm could have tapped into.

If their value proposition had been ‘composting toilets are green’, then could could have looked for markets that valued this highly, like many European countries, or California.

All of these would have been better options than China and India.

I think there some important lessons here.

First, don’t mistake numbers in a spreadsheet for reality. When you’re modeling markets for your new idea, you need to question all of the assumptions that underpin your model. The assumptions in this firm’s 1000 cell spreadsheet were deeply flawed – consequently, so were the decisions based on that spreadsheet.
Second, don’t automatically go for the markets with the most people. The number of firms that have gone broke following the ‘If we can just get everyone in China (or India) to just buy one of our products’ strategy is very high. You’re much better off finding a smaller market that values your idea.
Finally, you have to be clear about the business model for your idea. You need to know the hook for your new idea, who will value that, and what value network you’re in, among other things. Lack of clarity on just a couple of these can sink your entire effort before it really gets off the ground.

I just took another look at the report that I put together for them. I’m still happy with the analysis that I did, but I was answering the wrong question. I tried to lead them to better questions, but at the time I was too inexperienced to get that to work. The lesson of the 1000 Cell Spreadsheet is that while analysis is an important part of building your business model, a barrage of numbers is a poor subsitute for clear thinking. We have to get our business models right if we want our innovations to spread.

3 Comments

You Don’t Need Any More New Ideas!

Posted by Tim in innovation strategy, replication, selection, variety on 30 December 2009

Scott Berkun let out the secret of innovation today in an outstanding blog post. It’s a secret that Rowan Gibson tried to let out of the bag recently, and so did Braden Kelley on Blogging Innovation. I’ve tried to tell you about it too, using both analogies and statistics. The secret idea of innovation is this:

You don’t need any more new ideas.

Here is Berkun on the what we really need:

If there’s any secret to be derived from Steve Jobs, Jeff Bezos, or any of the dozens of people who often have the name innovator next to their names, is the diversity of talents they had to posses, or acquire, to overcome the wide range of challenges in converting their ideas into successful businesses.

That’s it. The problem is executing your ideas. Here’s an example – yesterday I talked about mousetraps – here are some interesting stats.

The patent for the flip-trap mousetrap design was filed in 1899. That’s a better mousetrap, right? We’re still using that design over 110 years later, so it’s probably pretty good. And yet, since 1899, the US Patent Office has granted over 4400 mousetrap patents. They receive more than 400 new mousetrap patents every year. So there’s no shortage of ideas. But fewer than 20 mousetrap designs have led to products that have actually made money. The problem in innovation is executing your new idea, and getting it to spread.

There is so much effort put into improving innovation by generating more ideas. This isn’t necessarily wasted effort, but it’s not the smartest use of resources. My MBA students evaluated innovation within their firms:

This approach is flawed, and my MBA students demonstrated why. They came from a wide range of organisations – huge multinationals, small start-ups, government departments, and educational institutions. Despite these different backgrounds, their findings were remarkably consistent – only 3 of the 60 organisations that they work in are ideas-poor. The other 57 (that’s 95%!) have problems with either selecting or diffusing ideas.

Here’s more from Berkun:

The closest thing to a real secret is this: In my years studying and teaching all things innovation, there’s one fact that’s the hardest for people to swallow and it goes as follows – To invent or create is to take a bet against the unknown. No matter what you do, you are still betting you can do well in the face of many things that are out of your control. Don’t like that? Don’t want uncertainty? Then do something else. Comfort with risk and uncertainty is the real secret. Or at least acceptance of the fact you can work your ass off for uncertain rewards.

Where does this leave us? Here are some conclusions:

If you’re going to get some help to improve innovation at your firm, don’t focus on generating ideas. Get help on selecting ideas, or on getting them to spread. Those are the hard parts.
Innovation is a bet – you’re betting that your new idea will work better, that it will meet needs, that it will fit into the value network. All of these things have to happen for your innovation to work. Like Berkun says, this is a leap into uncertainty.
Most of the innovation problems that organisations face are problems with innovation diffusion – the challenge is to get your new ideas to spread.

The new idea that I’d like you to accept is that you don’t need any more new ideas. Instead of generating more ideas, let’s develop some plans for getting better at executing our ideas. That seems like a good idea heading into the new year, doesn’t it?

11 Comments

When You Don’t Want Ideas to Spread

Posted by Tim in business models, innovation strategy, replication on 27 December 2009

One of the themes that I talk about a lot here is the importance of getting our ideas to spread. It is a central part of innovation – if our new ideas are not adopted, then we’re in trouble. But what about when our business model is based on ideas not spreading? The idea seems counterintuitive, but there are actually a lot of businesses that depend on our ignorance for their profits.

I had this idea while reading Socialnomics by Erik Qualmen. He talks about a new social-based search model and talks about why Google might not adopt this innovation:

Coincidentally, the search engine incumbent (e.g. Google) would be the least likely to introduce such a model. By endorsing such a shift, Google could see itself as a potential loser because it currently genreates a fair amount of money from the inefficiency of the old model.

I don’t agree with this quote – I think that Google is very well aware of the vulnerability they have with the current method of search and that is the big reason behind their constant introduction of new ideas. But Qualment raises an interesting point – Google does actually make money right now because our current methods of getting ideas to spread are not very efficient. We need search engine assistance.

However, there are more obvious examples of business models built around preventing ideas from spreading. Some of them are:

Media: The issue of file sharing is an obvious example. But there are plenty of others. What about the crazy regional code system on DVDs? Or the many streaming video sites that are only available in the US (Hulu, Netflix, and the one that kills me – Yahoo’s free NHL stream)? All of these models are built on the idea that the people generating the content will control how and when the idea spreads.
Car Dealerships/Real Estate Agents: Buying a new car used to be a horrible experience. You never knew what a fair price for a new car was, and as a consequence, people often ended up paying more than they should have. This is much less of a problem now as pricing information is much more widely available, as is data about which dealers have the cars of the make, model & features that you want in stock. There are similar issues in real estate – much of the purpose of real estate agents is based on the ignorance of home buyers. When we’re buying a house, we usually don’t know all the houses available in the area we’re interested in, what the going rates are, or how quickly houses are getting sold in the area. Well, it used to be the case that we didn’t know this – now we know a lot more.
Firms with poor customer service: It used to be much easier for firms to get away with bad service. If you weren’t happy with them, who could you tell? Ten of your friends? Ever since Jeff Jarvis & Dell Hell, it’s a lot easier now to tell everyone.

There are actually a lot of business models built on the idea that the firm has access to information that you can’t get. These worked for a long time, but they are increasingly difficult to maintain now. The key point is that if you are working in an industry whose profits are based on information asymmetries, it is unlikely that these will last for too much longer. That is the threat. The opportunity is that these industries provide fertile ground for business model innovation.

In particular, you can use strategies that aggregate, filter and connect to reconfigure the way these industries operate. We’re seeing that in real estate, as firms are starting up that collect all of the active house listings in one place (aggregate & connect), that provide sales histories for particular neighborhoods (filter), and that enable private listings to show up on google maps (aggregate, filter & connect).

We still have to get our good ideas to spread, but it’s getting increasingly difficult to prevent ideas from spreading. If your business model is based on preventing people from having information, you better get moving on coming up with a new business model. If you don’t, someone else will.

8 Comments

A Few Innovation Ideas

Posted by Tim in innovation strategy, replication on 26 December 2009

How do we get ideas to spread? It’s a critical question, and one of the ways that we distinguish between invention and innovation. For me, an invention is a clever new idea, an innovation is a clever new idea that is packaged up in way that enables it to spread. There’s a big difference between the two. I ran across several interesting articles today that shed some light on how to get our ideas to spread.

The first is a profile of Duncan Watts from Fast Company. The article looks at the question of whether or not you need to spread ideas primarily through targeting superconnectors, an idea put forward by Malcolm Gladwell in The Tipping Point. Watts suggests that peoples’ tendency to be receptive to new ideas has a greater impact on whether the idea spreads than who starts spreading the idea does. This result is very similar to the findings of our colleague Andrew Stephen.

This has some interesting implications. One is that ‘Influentials’ actually aren’t any more effective at sparking trends than normal people. This leads directly to the second point, which is that ideas spread most effectively when the time is right. Taken together, this makes it very hard to predict which ideas will spread, and it also makes it difficult to develop a strategy to make your ideas go viral.

I think that the best response to this is actually to approach innovation alogorithmically. What this basically means is that the way to innovate is to generate a lot of ideas, figure out ways to try them out cheaply and quickly, and then scale-up the ones that seem most promising. The FC article describes an advertising strategy devised by Watts that functions very similarly to this, and I think that it is the way to go.

The second article that caught my eye was Atul Gawande’s New Yorker piece on health care reform in the US. In assessing the bill that the Senate passed last night, Gawande applauds the way that this experimentation mechanism is built into the bill. It does not specify precisely how the new health care system will function, rather, it provides a platform for encouraging experiments and a path for getting the best new ideas to spread. Gawande describes how this approach worked during reform of the US agricultural system at the start of the 20th century, including the efforts of Seaman Knapp in Terrell, Texas:

Knapp knew that the local farmers were not going to trust some outsider who told them to adopt a “better” way of doing their jobs. So he asked Terrell’s leaders to find just one farmer who would be willing to try some “scientific” methods and see what happened. The group chose Walter C. Porter, and he volunteered seventy acres of land where he had grown only cotton or corn for twenty-eight years, applied no fertilizer, and almost completely depleted the humus layer. Knapp gave him a list of simple innovations to follow—things like deeper plowing and better soil preparation, the use of only the best seed, the liberal application of fertilizer, and more thorough cultivation to remove weeds and aerate the soil around the plants. The local leaders stopped by periodically to confirm that he was able to do what he had been asked to.

In a very poor year for cotton, Porter’s profits jumped substantially. This led him to use the new ideas over his entire farm. Many other local farmers followed suit, and the federal government gave Knapp more funds to expand the program. This was happening all over the country, and it transformed the agricultural industry – first in the US, then around the world.

So, again, try things, figure out what works, then scale up. That’s an innovation algorithm that works.

Note: If you’re reading this and you feel bad about thinking about work in the middle of the holiday season, remember that we’re just doing what Chris Brogan is telling us to do:

12 Comments

Tradition is Not a Business Model

Posted by Tim in evolving economic entities, innovation strategy, replication, time on 23 December 2009

I’m currently reading The Nature of Technology by W. Brian Arthur. It’s a fantastic book. This morning I ran across this quote discussing the spread of innovations:

There is another reason the old pricniple persists beyond its time, an economic one. Even if a novel principle is developed and does perform better than the old, adopting it may mean changing surrounding structures and organizations. This is expensive and for that reason may not happen. In 1955 the economist Marvin Frankel wondered why cotton mills in Lancashire did not adopt the more modern efficient machinery of their American counterparts. He found that the new machinery in the English setting would indeed be more economical. But it was heavy, and to install it the Victorian brick structures that housed the old machinery would have to be torn down. The ‘outer’ assemblies or elaborations thus locked in the inner machinery, and the Lancashire mills did not change.

Oh, and, the Lancashire mills went out of business in time as a consequence.

This story made me think of two things. The first is that when you come up with a new idea, it does not spread automatically. Like I said yesterday, you have to fight to get the innovation embedded into the economy. There are always surrounding structures and organisations that will have to change to accomodate an innovation. You can’t just build a better mousetrap – you have to get people to want a better mousetrap too.

The second point is that if you are well established in an industry, and you are surrounded by structures and organisations that lock you in to one way of doing things – you have to be aware of what is going on in the marginal niches. You always have to be asking if we were starting today, would we do this? The worst excuse for doing something is ‘that’s the way we’ve always done it’. When I hear that phrase, my heart always sinks.

Our current structure always constrains what we can do. Even if we are not innovating ourselves, we have to take this into account, and think about what competitors without those same constraints might be able to do. This is hard to do, but responsible managers have to do it.

As the authors of the journalism Internet Manifesto put it – Tradition is Not a Business Model!

(picture from www.historic-uk.com)

1 Comment

How to Assess Your Innovation Capability

Posted by Tim in innovation strategy, replication, selection, variety on 18 December 2009

How do you know how good you are at innovation? One of the tools that we have found very useful for assessing innovation within organisations is the Innovation Value Chain. The tool was developed by Morten Hansen and Julien Birkinshaw and published in an article called The Innovation Value Chain in Harvard Business Review in 2007.

I’ve talked about this before – There are two key points with this model. The first is that there are three stages in the process of innovation: idea generation, selecting & developing ideas, and diffusing ideas. The key part, however, is that all three parts of that process have to be working well in order to innovate.

The three step aspect of the innovation process is important. Measuring idea generation, selection and diffusion helps organisations get around the problem of simply equating innovation with ideation. Organisations that do this often find that they have plenty of ideas, but they’re still not being very innovative. This is because innovation actually doesn’t occur until you execute new ideas. To do that, you have to be good at having ideas, but more importantly you also have to be good at selecting ideas and getting them to spread.

This leads to the second key point of the Innovation Value Chain – your innovation process is only as good as your weakest link. This is not simply a linear model of how things happen, it is a description of a complex system. For example, if you are bad at selecting ideas, people will become less willing to give you their new ideas. This means that there are feedback loops between the three parts of the process. If you are going to improve your innovation, the whole system has to get better. You can use the IVC to identify your weak point and take steps to improve it. Then you can move on to whichever step is your weakest point now. If you keep doing this, you will build excellent innovation capability within your organisation.

Here is a Special Deal!

Our research has shown that while organisations usually first try to improve their idea generation, 95% of the time, this is not their weakest area. I’m curious to see how broadly this is true – so I would like you to please

Fill out this Survey!

In exchange for your time, I’ll give you some feedback on your results. If you’d like some information about what your organisation’s innovation strengths and weaknesses are relative to others who have taken the survey, just leave an email address when you take the survey. If you would like several people from your organisation to take the survey, I can compile the results – just have everyone indicate the name of the organisation when they fill out the survey. All results are, of course, confidential. To get the most meaningful results, please tell everyone you know that might be interested about this survey. Thanks for your help!

Innovation is about more than just coming up with new ideas. If we’re going to be innovative, we have to be able to execute new ideas. The Innovation Value Chain is one tool that can help us get better at this.

1 Comment

Using Networks to Spread Ideas

Posted by Tim in networks, replication on 18 December 2009

Yesterday I talked about some of the benefits and challenges of distributed innovation within organisations. One of the biggest challenges you face when you make everyone responsible for innovation is this – how do you get new ideas to spread throughout the broader group? This is part of what John and I are studying in our major research project at the moment. We have a three year grant to look at innovation networks within project-based firms. As we’re getting further into the research, it is becoming clear that this issue of idea diffusion is one of the biggest problems that these firms face.

Earlier this week, we did a pilot study for a student’s part of the project. Their question concerns how people search within their networks for information that they need. Because we haven’t made a good video talking about this yet, here is Venessa Miemis explaining some of the issues:

(there’s more good stuff from her here)

So the network facilitates innovation, as well as the diffusion of information – but how? That is what we’re trying to figure out because the ‘how?’ part has generally been treated as a black box. To get at this, we will map networks within four groups of people in one firm that share a knowledge area, but who are spread across a number of different locations. This week, we tested the survey on a small group in the firm, and we learned some interesting things even from this.

NetworkKnowledgeProvision

This is one of the networks that we mapped. It shows the links based on responses to the question ‘who provides me with significant knowledge?’ In this case, we defined significant knowledge as that which was essential for solving a work-related problem. There are a couple of interesting things that we learn from this.

The first is that it is a relatively sparse network. This surprised the group – the manager thought that we wouldn’t learn much from this team because they worked very closely together and they are highly cohesive. Still, even within a highly cohesive team, knowledge is not evenly distributed.

The second issue concerns the diamond formed by the four people in the middle of this network. This group of four was at the core of all of the different networks that we mapped. The surprising thing here is that this structure actually reflects the formal hierarchy of the group pretty closely. Organisational network analysis often shows that the informal networks are quite different from the formal structures of the firm. But that doesn’t appear to be the case here. We’ve actually found this in other parts of our research in other firms as well. So we’re starting to think that in distributed innovation networks, hierarchy is actually more important than we expect it to be. This is still very speculative, but it’s potentially interesting.

The bottom line is that when our innovation efforts are distributed, it is critical to understand the structure of our knowledge-sharing networks.

11 Comments

James Boyle’s Important Ideas on IP & Innovation

Posted by Tim in business models, innovation strategy, replication on 14 December 2009

Intellectual Property rights encourage innovation, right? Right? Well, not necessarily. Actually, people that study this empirically consistently find that the evidence suggests that they don’t. Here’s a fantastic talk by James Boyle discussing his book The Public Domain, which addresses this exact issue:

(Thanks to Gerd Leonhard for the tip on this talk)

Boyle’s book is outstanding (there’s a link to a free download of the book here). His central thesis is that we tend to underestimate the benefits of openness while overestimating the potential downsides. Consequently, we have too much legislation supporting IP rights. However, he effectively documents all of the research on this topic, which consistently shows that IP rights in general do not encourage innovation, they suppress it.

The original intent of patents was to encourage the use of technology so that new ideas could be built on these technologies as soon as possible. At the national level, this has several implications:

Class, repeat after me:

(1). Green jobs are NOT a zero sum game where nations are competing for a fixed number of them.

(2). If China or Germany or anyone develops “innovative energy technology”, that is NOT bad for us. It is in fact *awesome* for us, as we can then adopt it and use it.

People, ideas are public goods. That is the whole basis of new growth theory. If China is now doing cutting edge R&D, that is an unmitigated blessing for everyone on the planet.

Will Wilkinson picks up on that post:

Worrying that other countries are pulling ahead is like worrying that the other oarsman in your boat will beat you to the destination if you’re lazy. You’re in the same boat! The smart thing is to goad everyone else into going as fast and hard as they can.

So if we don’t have IP rights to protect our ideas, how do we make money off of them? We win by building the best business model.

The value of an innovation is not in the idea that it is built on, but on the execution of that idea.

(thanks to Venessa Miemis for the link to the Wilkinson post)

No Comments

Seeing what’s coming

Posted by Tim in innovation strategy, replication, time on 2 December 2009

When Alexander Graham Bell developed the telephone, he offered to sell the patent to Western Union. He knew that getting the idea to spread was the hardest part, and he figured that a big firm that was already in the communications industry would be better equipped to get the idea out there. This was part of the minutes from the Western Union board meeting that decided the telephone had no future:

Bell’s instrument uses nothing but the voice, which cannot be captured in concrete form…. We leave it to you to judge whether any sensible man would transact his affairs by such a means of communications. In conclusion the committee feels that it must advise against any investment whatever in Bell’s scheme.

I hear similar things all the time from firms. We often get so caught up with what is good about what we’re currently doing that we fail to see any value at all in alternatives. Consequently, we miss out. Or, worse, we are made irrelevant by someone else.

What threats (or opportunities!) are you missing right now for the same reason?

7 Comments

the hardest part of innovation

Posted by Tim in innovation strategy, replication, selection on 28 November 2009

I was thinking about my talk from yesterday, and one bit that I just spontaneously threw in is probably worth expanding on. I spent a lot of this week marking assignments from my MBA students (who were an exceptionally good bunch this year). For the major assignment this year, I had them analyse their own firm or organisation using the Innovation Value Chain model developed by Morten Hansen and Julian Birkinshaw.

There are two key points with this model. The first is that there are three stages in the process of innovation: idea generation, selecting & developing ideas, and diffusing ideas. The key part, however, is that all three parts of that process have to be working well in order to innovate.

Both John and I have talked about the dangers of over-focusing on idea generation at the expense of execution, so we find this to be an extremely useful model. In particular, we have frequently observed organisations decide that they have to improve their innovation, and then sinking all of their resources and effort into idea generation.

This approach is flawed, and my MBA students demonstrated why. They came from a wide range of organisations – huge multinationals, small start-ups, government departments, and educational institutions. Despite these different backgrounds, their findings were remarkably consistent – only 3 of the 60 organisations that they work in are ideas-poor. The other 57 (that’s 95%!) have problems with either selecting or diffusing ideas.

So when firms focus on improving their idea generation, it is a mistake for two reasons. The first is that this is almost certainly not where their problem lies. They’re probably worse at execution. The second is that it does not take into account the entire innovation system. I just saw this quote from Russell Ackoff (via Venessa Miemis):

Improving the performance of the parts of a system taken separately will necessarily improve the performance of the whole.
False. In fact, it can destroy an organization, as is apparent in an example I have used ad nauseum: Installing a Rolls Royce engine in a Hyundai can make it inoperable. This explains why benchmarking has almost always failed. Denial of this principle of performance improvement led me to a series of organizational designs intended to facilitatethe management of interactions: the circular organization, the internal market economy, and the multidimensional organization.

Innovation within an organisation is a system. It is much more than simply idea generation. And if you focus only on improving your ideation, there’s a pretty good chance that your overall innovation performance will actually get worse. The hardest part of innovation is idea execution, and we simply must get better at it.

16 Comments