On the Importance of Time in Innovation


Innovation is executing new ideas to create value.

If you think about that, you can see that time is important – you can’t do all three things instantaneously.

Here are two important ideas:

  1. Within an organisation, innovation is the process of idea management.
  2. New ideas diffuse along an s-curve.

If we put these two ideas together, we can see how time works in innovation.

Here are the parts of the idea management process:


If we order the steps as:

  1. Idea generation
  2. Idea selection
  3. Idea execution
  4. Sustaining ideas
  5. Diffusion

Them we can map them onto the S-Curve like this:



Here is what is going on:

  1. Idea generation: there is no idea to diffuse if we don’t have them in the first place.  However, the best processes have constant feedback built in to them, so that ideas improve as we go along.  This is an essential part of the lean startup approach (which we can use in established organisations as well).
  2. Idea selection: there are two ways to approach this.  First, we can select ideas before we start working on them, so they happen at the start of the process as shown. Second, we can use a more experimental approach, where we choose ideas as we test them out.  This means that there is idea selection taking place during the invention phase too.
  3. Idea execution: we have to make the idea real – there is no innovation without this.
  4. Sustaining: this part is tricky for larger organisations.  Sustaining involves keeping people involved and excited as we go through the periods marked Y and X in the diagram.  The value for X is always larger than we expect, and it is easy for people to lose enthusiasm for an idea as we go through the start of the diffusion process.  This can kill innovation.
  5. Idea diffusion: like selection, there are two approaches to this.  First, we can wait to work on getting our new idea to spread until after all of the development work is done.  This makes diffusion much harder.  It is better to use the second approach, which is to be thinking about idea diffusion all the way through the process.

There are several important points to come out of thinking about innovation in this way:

The best innovation processes have all five parts interacting.  This follows from the discussion of both idea selection and diffusion.  It is more effective to think of this as a continuously interactive process, rather than a linear one.  To that end, instead of thinking about the process as I’ve put it in the first diagram, you can just as easily picture it this way:



Here is how Eric Ries puts it in a must-read post from Tren Griffin today:

All of our process diagrams [in major corporations] are linear, boxed diagrams that go one way. But entrepreneurship is fundamentally iterative. So our diagrams need to be in circles. We have to be willing to be wrong and to fail.

The second point is that the timing of our new innovations is important.  Timothy Lee wrote a great piece on Vox on the timing of newspaper innovations called Newspapers weren’t too late to online news – they were way too early.  He makes two important points.  The first is that it takes a long time for us to figure out what new technologies are for (the value of X in the diagram above is much longer than we expect!):

30 years is a typical period of time between the first experiments with a new technology and mainstream commercial success.

The first mouse was invented in 1965, but it took until the mid-1990s for mice to be a standard computer feature. The first packet-switched network was invented in 1969, but the internet didn’t become mainstream until the late 1990s. Multitouch interfaces were first developed in the early 1980s, but didn’t become a mainstream technology until the iPhone in 2007.

That suggests we shouldn’t underestimate the disruptive potential of technologies, like self-driving cars, personalized DNA testing, and Bitcoin, that seem exotic and impractical today.

The second point is that experiments in the flat part of the S-Curve typically fail.  Lee outlines many of the ideas that newspapers tested in the 1980s and 1990s designed to bring news online – they all failed.  The result:

Indeed, part of the reason that newspapers were slow to adapt to the web is that by the time the web got big in the 1990s, newspapers had been experimenting with online services for two decades. And based on those experiments, they concluded that online services weren’t a serious threat to their business.

This reflects two common innovation problems.  The first is that we often expect new ideas to be adopted instantly.  This, of course, is never true.  That is why it is so important to understand the S-curve.

The second problem is that we often expect new technologies to simply fit into existing business models.  This is what happened to newspapers.  However, one of the reasons that the S-curve starts out flat is that new technologies require new business models.  Business model innovation is the learning tool that helps us discover the value in new technologies.


This means that business model innovation needs to be part of our innovation management process.  The right idea at the wrong time is still wrong.  Business model innovation is one of the tools that can address this problem.

Innovation takes time. When we don’t understand the importance of time to innovation, we substantially decrease our chances of success.  We end up looking for quick wins, with a guaranteed return on investment, rather than patiently building the systems that will enable us to succeed over the long term.

To innovate successfully, we must understand the role of time.

Student and teacher of innovation - University of Queensland Business School - links to academic papers, twitter, and so on can be found here.

Please note: I reserve the right to delete comments that are offensive or off-topic.

8 thoughts on “On the Importance of Time in Innovation

  1. Hi Tim – an excellent post covering some fundamental concepts. The other element of time is time to market. For incremental innovations when the job to be done is relatively clear, the competitive context is understood and the target customer is known; time to market should be as soon as possible. Post-launch iteration is important but is often forgotten because resources are rapidly moved to the next item in the pre-launch queue.

    For more radical innovations, the temptation is also to go as fast as possible, the danger is that the JTBD is not understood; there are no competitors; and the target customer is mysterious. Even then I would still move fast and iterate as much as possible and, as Paul Sloane wrote recently, remain in Perpetual Beta.


    • I agree with all your points Kevin. I think it was Sam Altman in his Stanford startup class who said something like the big advantage in radical innovation is to iterate as quickly as possible…

Comments are closed.