How to Effectively Manage the Fuzzy Back End of Innovation

You have doubtless heard of the fuzzy front end of innovation. It is another name for idea generation. But Jeffrey Baumgartner believes that the back end of innovation, where implementation is supposed to take place, is just as fuzzy. Many companies lack clear, efficient processes for implementing the ideas they generate.

In the world of business there seems to be a compelling need to label a simple concept with a bit of silly jargon in order to confuse people. Hence, the term “fuzzy front end of innovation” was invented to imply that there is something fuzzy and unclear about having ideas. However, I would argue that the back-end of innovation is just as fuzzy and twice as necessary as the front end.

Most companies I speak to today have implemented some kind of fuzzy front end to their innovation process. Usually it is an idea management software, some form of suggestion scheme or regular activities for generating ideas. Interestingly, nearly all of these companies also have a back-log of unacted-upon ideas. Why is this? There are two reasons.

First, business innovation tends to focus way too much on idea generation and this results in a large number of ideas that need to stored somewhere until someone can work out what to do with them.

Second, and the subject of this article, there is a lack of processes for testing and implementing ideas generated by the fuzzy front end. This is evidenced by CEOs who bemoan the fact that they don’t have a lack of ideas. They have a lack of innovation.

The lack of a fuzzy back end of innovation suggests poor planning. Think about it: Why create a process for generating ideas when you do not have a process for doing something useful with those ideas? It’s rather like building a factory to produce juice, but not creating any mechanism for bottling or packaging the juice for distribution. The result: more and more juice being stored in hopes that the odd liter of it might find a container now and again.

Where’s your fuzzy back end?

If your company is capturing ideas from employees, then you need to have processes for testing and implementing the most interesting ideas. Most likely, you need multiple processes. The steps necessary to implement a process efficiency idea is presumably different than the steps needed to turn a product improvement into reality. If being very innovative is important to your firm, you will also need processes for implementing the zany ideas that have the potential to become breakthrough innovations. In fact, you may even need to define a process for identifying these ideas.

Most suggestion schemes and idea management systems behoove managers to choose the most popular ideas, rather than the most creative which are often among the least popular.

In addition, you probably also need an improved process for testing and implementing ideas devised and developed outside of specific initiatives by rank and file employees, particularly those who are on the lower half of the corporate ladder. Anecdotal evidence suggests that most companies follow a similar, two step, low-risk approach:

  1. Employee suggests her idea to a middle manager.
  2. Middle manager rejects the idea.

Although this approach minimizes risk and additional workload in the short term, it is clearly less than productive in terms of innovation.

With all of these ideas coming in, you need processes for testing and implementing them. The form these processes take depends upon the way your company works and the corporate culture. Nevertheless, there are some key elements that should be in your testing and implementation process. Let’s look at them.

Submission and approval

In your company, when someone has a great idea, what can she do with it? To whom should she submit it? In what format should it be submitted? If this is not obvious to your employees, it needs to be made obvious. In addition, you should have specific requirements for submitting ideas for approval (as opposed to submitting them to a suggestion scheme or ideas software). I suggest you require ideas be developed to a certain degree, perhaps with a short business plan or by filling out a form that covers the criteria by which you would judge a project.

In order to help people who have developed ideas that extend beyond their area of expertise, you should provide partnering information so employees can collaborate with others who have the needed expertise. For example, if a marketing person has a product improvement idea, enabling her to find and collaborate with someone in the research and development division makes sense.

Once the idea is submitted, it needs to be approved or rejected, of course. Who will take responsibility for this task? Most likely it will be different people for different kinds of ideas and possibly different levels of budget. A $5,000 idea is likely to be easier to approve than a $5 million idea.

Perhaps more important than approval is rejection. How are ideas rejected? How is the rejection communicated to the idea submitter? Ideally the rejection should provide information about why the idea was rejected as well as whether the rejection is final, as might be the case for an idea that is not in alignment with corporate strategy, or whether it is provisional, as might be the case for an idea that would exceed budget possibilities but which might be viable if it could be developed at significantly lower cost.


One reason ideas languish in the fuzzy front end of innovation is that they are much cheaper to dream up than they are to evaluate, test and implement. An approved idea will almost certainly need money in order to be developed and implemented. What budget is available? Who authorizes it? How is it tied to the approval process? These are important questions and the answers should be transparent to employees. If your people know the money available for developing ideas, they can plan around those budgets.

In the case of relatively expensive ideas, it is likely you will want to provide several tranches of money at different approval steps. The first tranch, or tranches, would be made available for testing an idea. A sophisticated idea such as a new product incorporating cutting edge technology will probably need to go through several testing processes. Assuming the testing is positive, a further tranch of money will be needed for implementing the idea.


When ideas are highly creative or lead to complex projects, it is important to set milestones in the development of the idea into reality. At each milestone, there should be an evaluation of the idea together with automatic actions depending on results. If the project does not meet a milestone, it should be re-evaluated and possibly rejected so that resources can be reallocated to new ideas.

Do you have a process for this? If so, what happens when a team kills a project that does not meet a milestone? They should be rewarded! By killing an unpromising project early on, they will save your company a lot of money. Sadly, in companies where failure is not tolerated, this does not happen. When projects fail, even early on, those responsible often face consequences. In such companies, no one wants to take a chance with a highly creative idea that has real innovation potential. The risk of failure and reprimand is simply too high. What is the situation in your company?


How are ideas tested? Many product and process ideas can easily be prototyped, which is a great way to get a feel for a new idea as well as get feedback on the it. Ideas that may be difficult to comprehend when explained in words can be surprisingly elegant when put into physical form! Service ideas are often harder to prototype, but not impossible. I find role-play a good method for the initial testing of the idea. When people see and feel an idea being performed, they can more readily identify its potential strengths and weaknesses.

Trials of ideas, of course, are also good for testing ideas while minimizing risk. However, the danger here is that if you spend too much time testing an idea, you increase the likelihood that a competitor will learn of the idea and take it to market before you. So, sometimes you will need to take a big risk with a big idea. If so, how will you do that? Who will make the decision? How will you mitigate risk?

The truth is, you probably know better than I do the best way to test ideas in your company. The question is, do your employees also know this? They should if you want to encourage them to implement ideas!


The final step, of course, is implementation. How does that happen? Who makes the final decision? How is this made? How will budget be made available? How will success or failure be measured. This information needs to be clarified and made available to people who wish not simply to have ideas, but to make them happen.

A process and communication

Ultimately, you and your top management need to decide on implementation processes. Moreover, you know your business intimately (I hope!), so you can do it far better than an outsider. However, if innovation is more important to you than ideas, you need to design and implement processes that facilitate testing and implementing ideas. Moreover, you need to be transparent about the process and communicate it across the company so that when people have ideas, they know how to develop, submit, test and implement their ideas. A streamlined process will make innovation more efficient, more measurable and, most importantly, more doable!

In short, be sure that when it comes to innovation, you have a fuzzy back end for your fuzzy front end!

By Jeffrey Baumgartner

About the author

Jeffrey Baumgartner is the author of the book, The Way of the Innovation Master; the author/editor of Report 103, a popular newsletter on creativity and innovation in business. He is currently developing and running workshops around the world on Anticonventional Thinking, a new approach to achieving goals through creativity.