Solving The Innovation Equation

Innovation is analogous to a complex mathematical equation. Identifying
the constants in the equation is the key to solving it.

The day after the IBM PCjr was announced in 1984 the headlines in the Wall
Street Journal read, “the PCjr is a flop.” The critics were right—it was a
flop, but how did they know? What metrics were they using to evaluate the
PCjr? As a member of the PCjr development team I wished we could have
gotten our hands on those metrics well in advance of
announcement—preferably before we even started developing the product. If
that were possible we could have simply built the product to address the
stated metrics and achieved success.

In the years that followed I worked on the IBM PC product planning team,
trying to figure out a way to build successful products around
customer-defined metrics. As part of that effort, four questions had to be
answered:

1. What criteria/metrics do customers use to evaluate a product?

2. Is it possible to capture those criteria/metrics in advance of a product
being developed?

3. How can we be certain that the evaluation criteria will not change from
the time a product is conceived until the time it is announced?

4. Will using the criteria/metrics enable us to effectively predict if a
product will win in the marketplace?

To answer these questions, I found it helpful to think about innovation as
analogous to solving a complex mathematical equation—an equation that
requires a company to look at the universe of possible solutions and figure
out which solution will best address a number of unmet customer needs. To
solve this equation, as with the analogous math problem, there must be
constants in the equation—the inputs can’t all be variables.

When we consider the two key inputs into the innovation process (solutions
and needs), it is safe to say that “solutions” are always the variable in
the innovation equation—they are, by design, ever changing over time.
Consequently, to solve the innovation equation “needs” must be defined as
constants in the equation. After all, if “needs” are also variables (ever
changing) then innovation will always be a random, iterative process.

This is the innovation challenge that most companies face today:
somehow companies must figure out a way to define need statements
that are stable over time—as constants in the innovation equation.
Fortunately, this problem can be solved when a company focuses on
the customer’s “job-to-be-done.”

A job-to-be-done is the underlying process a customer is trying to execute
when using a product or service. “Repairing a hernia”, “passing on life
lessons to children”, “managing cash flow”, and “staying informed on a
topic of interest” are all examples of jobs that people are trying to get
done.

Because we always define a “job” as a process, a job is stable over time,
thus creating a stable target around which customer needs can be defined
and gathered. If the goal of a product or service is to help a customer get
a job done better, then we can define “needs” as the metrics customers use
to measure success when getting a job done. We call these metrics “desired
outcomes” and they form the foundation of our
innovation process: Outcome-Driven Innovation.

When “needs” are stated as “desired outcomes” the innovation process is no
longer random and unpredictable. Here’s why:

Desired outcomes are the metrics customers use to evaluate a product.

They are the metrics that customers use to decide which products get the
core job, consumption chain jobs and related jobs done best. Emotional jobs
and financial metrics are also part of the equation. Customers typically
have 100 or more metrics to consider when completing a product evaluation.

Desired outcomes can be captured before a product is conceptualized and
developed.

Since desired outcomes are tied to the job-to-be-done and not the product,
these metrics can be captured well in advance of product development. They
can even be captured in markets where products do not yet exist.

Desired outcomes do not change from the time a product is
conceptualized until the time it is announced.

Since the job-to-be-done is stable over time, desired outcomes statements,
which define success along the way, can be constructed in a way that makes
them stable over time as well. Desired outcome statements are devoid of
solutions, making them constants in the equation.

Desired outcomes can be used to effectively predict is a product will
win in the marketplace.

With the metrics in hand, a company can evaluate an idea it has for a new
product against the set of metrics to see if and how much better it will
get the job done. This analysis also makes it possible to discover the
weaknesses associated with a new product concept, thus providing a path for
enhancing ideation and improving the solution before it is even developed.

Consider how this thinking was applied by Cordis Corporation as they tried
to win back market share in the angioplasty balloon market. They had to get
the job of “restoring blood flow through an artery” done better than
competing solutions. The interventional cardiologist’s most underserved
desired outcomes are shown in Figure 1.

Figure 1: Desired outcomes on the job-to-be-done

With these metrics in place, Cordis had the ability to test different
solutions against the unmet desired outcomes and quantify which solutions
had the potential to get the job done best. Confident that they
conceptualized a winning solution, they approved the new line of
angioplasty balloon products for development. They introduced the new line
of products 18 months later, 19 products in total—all of which became
number 1 or 2 in the market, and increased their market share from 1% to
over 20%.

With “needs” defined as constants in the innovation equation, innovation
becomes a predictable process.