What makes a good innovation analyst?

What is an innovation analyst?

An innovation analyst is somebody who does tech discovery, tech intelligence and scouting, market analysis, competitor analysis, or investment analysis for a living. But this does not necessarily mean that they do this full-time. In fact, most people in tech and business probably do innovation analysis somewhat “on the side”. Their focus is on developing new products, and on exploring tech and business opportunities more generally. But they know that somewhere out there, there are people, companies, and ideas that could help them innovate stronger. So they need some way of discovering and tracking these things. And this is what we call ‘innovation analysis’. And this is why you, even if you don’t do innovation analysis full-time, are probably an innovation analyst as well.

What do innovation analysts do?

Here are some examples of what innovation analysts do:

Discover solutions to technological challenges

This includes finding companies, people, or methods that can help you do or build something faster and better. For example, if you work in manufacturing, you might require new types of welding technologies to build your products.

Discover game-changing or disruptive technologies as early as possible

There might be a technology out there that could help you build something much faster, better, or cheaper than if you did it on your own. For example, if you do machine learning, new software-based approaches that use commomodity hardware could help you train your models much cheaper than specialized and expensive “AI chips”.

Identify new markets for your products and technologies

Here, the goal could be either to identify markets for a new technology, or new markets for an existing technology or product that you already have. For new technologies, it is often better to identify and address niche markets first.

Estimate the maturity level of a technology

A technology could be at a purely conceptual stage, or readily available off-the-shelf, or anything in between. In another article, we describe how you can use data for doing back-of-the-envelope tech maturity assessments. We used this method to estimate the maturity of various green hydrogen technologies, for example.

Identify possible future scenarios for technologies

What-if scenarios for what might happen in a given technology field. In another article, discovering strategies in additive manufacturing, we show you an example of this.

Identify and track competitors, including non-obvious ones

Tracking known competitors is often relatively straightforward. But identifying and tracking new, non-obvious competitors is a lot harder (but perhaps even more important). We have an example of this in another article, on the surprisingly interesting business of shock absorbers.

What are the traits and mindset of a good innovation analyst?

At Mergeflow, we interact with many people who do innovation analysis. Over the last few years, we have noticed certain traits, mindsets, or approaches that appear to work better than others. Below, I describe what I think are the most important three.

Know the book, but don’t always do things by the book

Very often, good innovation analysts ask “why not?”, rather than “why?”. This does not imply “breaking the rules”. Instead, it is about “expanding the rules”. In order to illustrate this point in a different context, Charles Elachi (former head of the Jet Propulsion Laboratory) used this photograph in a 2008 TED talk:

Guess who is the JPL employee in this picture? This person would also make a good innovation analyst.
Guess who is the JPL employee in this picture? This person would also make a good innovation analyst.

Only one of the people in the photograph was a JPL employee. I’m sure you can guess who. This person would also make a good innovation analyst. Notice where the other people are looking vs. where he is looking?

Wherever everybody else is looking, look somewhere else and do something different.

Charles Elachi, in his 2008 TED talk.

What does this mean in more concrete terms? For example, if you did tech discovery “by the book”, you’d probably look for solutions within your own industry. Say you’re interested in collaborative robots for manufacturing. If you went by the book, you would look for solutions in the manufacturing industry. But, and this might be more interesting, you could instead explore medical technologies for surgery, for example.

Mind and hands

“Mind and hands” means that you combine theory and practical action to solve problems. It is also the motto of my alma mater, MIT. Throughout the history of science and technology, many innovators have used this approach very successfully. For example, in another article, we wrote about how this approach was one of the innovation practices that made Galileo Galilei so successful.

Galileo Galilei, a "mind and hands" master. His strong network and 'science and business' mindset helped him excel at tech discovery and innovation.
Galileo Galilei, a “mind and hands” master. His strong network and ‘science and business’ mindset helped him excel at tech discovery and innovation.

In practice, I do not mean to suggest that you should do everything yourself, of course. But you should interact very closely with the people in your organization who build things, if building things is not what you do. You could also make an inventory list for the product or technology you want to build. For some more thoughts on this, see our article, “5 ways to escape innovation theater”.

Beware of certainties and predictions

Good innovation analysts know that tech discovery is about probabilities, and not about certainties. Even if this might be hard to swallow.

This is about intelligence gathering and analysis, not about mathematical axioms.

Adapted from a Tweet by Christopher Ahlberg.

And aiming for a first useful approximation tends to be much more useful and effective in tech discovery than trying to be comprehensive. We discuss the concept of “first useful approximation” in more depth in another article, “The most effective way to beat analysis paralysis”.

Similarly, good innovation analysts cringe when they hear ten- or twenty-year predictions about how technologies will develop. Yes, there are plenty of pundits who publish things like this. But once you really start digging on how verifiable and falsifiable these predictions are, you come back rather empty-handed. We talk about this in a bit more detail in our article, “Making technology and innovation predictions”. But you don’t have to take our word for it. There is a very good book by Philip Tetlock and Dan Gardner, Superforecasting. The book came out of a contest, The Good Judgment Project, that was run by IARPA. And in a 2015 talk, Steven Popper from the RAND Corporation discusses the topic of “plausible futures”. You can watch the talk on YouTube.

There are many ten- or twenty-year predictions about how technologies will develop.  But in most cases, you could also do cartomancy instead.
There are many ten- or twenty-year predictions about how technologies will develop. But in most cases, you could also do cartomancy instead.

Now, let’s move on and look at some methods that good innovation analysts use.

The methods toolbox of a good innovation analyst

Similar to traits and mindset above, we have seen several methods that good innovation analysts tend to use. Below, I describe three that I think are particularly important.

We have made a free ebook with practical tips for improving your tech discovery skills.

We have made a free ebook with practical tips for improving your tech discovery skills. You can get the ebook by clicking here.

Back-of-the-envelope estimating

When you do back-of-the-envelope estimating, you try to get approximate answers to complex questions. For example, let’s say somebody asks you, “How many piano tuners are there in Chicago?”. You could now put in huge efforts and try to get a precise answer. Or you could try to estimate. The famous physicist Enrico Fermi was very good at this. The piano tuners question is attributed to him, for example.

The physicist Enrico Fermi, a master estimator. Use his methods to improve your tech discovery skills.
The physicist Enrico Fermi, a master estimator. Use his methods to improve your tech discovery skills.

The idea behind “fermiing” is that you can relatively quickly do order-of-magnitude estimates. And in many cases, this really gets you quite far already. Plus, because fermiing is so quick, it is much easier to iterate and refine your hypotheses. You simply could not do this if every step took you days or weeks to complete. See also the “first useful approximation” method that I mentioned above.

The goal of tech discovery is not just to accumulate knowledge, but to stimulate a thought process.

Of course, when you do tech discovery, the number of piano tuners in Chicago is probably not on top of your list. But you can use the same method. For example, you can use it to do market size estimates. If you want to know how to do this, check out our article, “How to estimate the plausibility of a market estimate”.

Associating

In their HBS article, “Five discovery skills that distinguish great innovators”, Jeff Dyer, Hal Gergersen, and Clayton M. Christensen argue that “associating” is a very important skill for innovators to have (the other four are questioning, observing, networking, and experimenting).

Associating (…) is about discovering new directions by making connections across seemingly unrelated questions, problems, or ideas. Innovative breakthroughs often happen at the intersection of diverse disciplines and fields.

FROM FIVE DISCOVERY SKILLS THAT DISTINGUISH GREAT INNOVATORS. ARTICLE BY JEFF DYER, HAL GERGERSEN, AND CLAYTON M. CHRISTENSEN.

At first, associating might sound like a skill that you either have or don’t have. And while associating probably comes more natural to some people, there are resources you can use to learn it. Wikipedia, for example. For more details on this, please see our article, “How to use Wikipedia to boost your discovery skills”.

Communication between R&D and business

R&D-to-business communication can be similar to communication between East and West during the Cold War. I’m just slightly exaggerating.

Glienicke Bridge near Berlin was a central element in the difficult communication between East and West during the Cold War. Communicating between R&D and business sometimes seems almost as difficult.
Glienicke Bridge near Berlin was a central element in the difficult communication between East and West during the Cold War. Communicating between R&D and business sometimes seems almost as difficult.

Good innovation analysts are good R&D-to-business communicators. This also makes them better at tech discovery.

Being a good R&D-to-business communicator helps you get a 360° view on your topics.

“360° view” means that you look at a topic from various angles. From business, markets, venturing, R&D, and intellectual property. A 360° view helps you make better decisions. For example, if you only looked at R&D, and your topic shows a lot of momentum there, you wouldn’t know if your topic has a market as well. But knowing this is important, particularly if you want to develop a marketable product.

Good business-to-R&D communication means that you speak the languages of both sides. After all, business and R&D people often call the same thing by different names. For example, an R&D person might say “deep reinforcement learning”, and a business person might say “artificial intelligence”.

We talk about this in more detail in two other articles. One from the perspective of a scientist, “How to talk like a business person (if you are a scientist)”, and another from a business perspective, “How to talk like a scientist (if you are a business person)”.

Some further readings (books) on innovation analysis

To my knowledge, there aren’t any books on innovation analysis or tech discovery. There are on competitive analysis, but I’d argue that this is only one sub-aspect of innovation analysis. So I thought about adjacent subject areas. One such adjacent subject area is intelligence analysis as it is practiced in government agencies. Not to mention that these organizations have innovation or technology intelligence capabilities as well, not “just” geopolitical.

Just like an innovation analyst, a (government) intelligence analyst has to address questions that cannot be addressed directly. By “directly”, I mean something like, “OK, Google|Alexa|Siri, what is the population size of New York City?”. Instead, they need to find much more creative, indirect ways to address their questions. They probably use fermiing (see above) a lot. And just like innovation analysts, government intelligence analysts have to keep track of many fast-paced and disparate sources of information. Also, they deal in probabilities rather than certainties. And they need to get their message across to decision makers. This last point is very similar to the R&D-business communication gap I mentioned above. RAND economist Steven Popper, whom I mentioned above, talks about this in this talk. He refers to the difference as “quantitative culture” vs. “narrative culture”.

Anyways, here are some of my personal favorite readings, along with some comments:

Fingar, T (2011). Reducing Uncertainty.

Here, I really like the emphasis on the purpose of intelligence. How to have an impact on decisions, and how this may be achieved. And also how things can go wrong, even if intentions are good. All of this from a world class and hands-on expert, Thomas Fingar.

Heuer, RJ & Pherson, RH (2014). Structured Analytic Techniques for Intelligence Analysis.

This is a real workbook, in the sense that you can use it for and during work. It describes a range of techniques for different circumstances, as well as those techniques’ strengths and weaknesses. The book’s spiral binding keeps it open at any page you want. This makes it convenient to use as a reference on your desk.

Heuer and Pherson also talk about analysis paralysis.

Aldous, D (2009). A Tradecraft primer: Structured Analytic Techniques for Improving Intelligence Analysis.

This is a paper that is based on an earlier version of the Structured Analytic Techniques book above. The paper is by David Aldous, and you can download it here as a PDF.

Other (= non-book) innovation analysis readings

For example, I often read publications coming out of the RAND Corporation. In particular, I recommend keeping track of their Emerging Technologies topics, some of which we have integrated into “emerging technologies” semantic models in Mergeflow. In addition, their Science, Technology, and Innovation Policy publications are great.

Benedict Evans’ newsletter is great, too. I recommend subscribing to it.

And here is an interesting future scenario by Bob Gourley of what the intelligence workstation of the future may look like (you have to get an account to access the full content).

None of these materials provide cookbook instructions. But they provide very interesting examples and new perspectives. And making the transfer from these perspectives to one’s own questions, discovering new angles, and questioning your own assumptions are all part of why this is an essentially human and fun activity, right?

Leave a Reply