What would you choose? Why?

Coming late to market translates into lower profits over a product’s lifetime. High-tech products that come to market on time but over budget show a much higher return than products that arrive late within budget, according to a McKinsey model 1

This McKinsey study, probably due to Reinertsen, also indicates that a six month late entry into the market place reduces the life time profits by nearly 33%1. The actual numbers may vary by industry and product line.

Still not sure?

For your key product, pull out your cash flow analysis spreadsheet; instead of just looking at net present value, internal rate of return, and return on investment, go through sensitivity analysis and compute the cost of 1-6 month delays in entry into the market. Generate several scenarios using expected time to enter the market for a key competitor. Make a summary and display the results through a chart. This chart will provide you with the ammunition to get additional resources to speed up the project even in bad times.

Being first to market, ensures higher free publicity, higher name recognition, higher market share and, yes, higher profit margins. Even Wall Street and stocks rewards you with higher valuations.

How to reduce Time To Market on a Single Project?

You may ask, “How do we reduce time to market? We are dealing with so many unknowns in the chemical product/process development R&D stage?”

I propose two solutions, one from Information Theory, another from understanding Work-Flow.

Information Theory:

A hard question that needs to be asked, but is hardly asked, is “What is the purpose of a single experiment?” How do we achieve that purpose?

We all agree that the purpose of an experiment is to generate new information. So a question that we need to be asking is “how do we run the experiment, so the information is maximized?” This is indeed a hard question. And since we do not know how to measure information, we seldom ask this hard question.

Paul Shanon, thinking in a higher dimension, related Information Content to Success. If probability of Success is p, then Information Content I is given by

I = p* ln(p); and I is maximized when p =½

So to Maximize Information Content, plan and run experiments such that probability of Success is = ½.

This is a very unusual result. Most scientists are very conservative, they want to succeed every time, so they take little risk and make small changes. Shanon says, run experiments so that probability of success is ½; in other words, half the experiments should succeed, half result in failure. Simply translated, “Be bold, but not brash, in running your experiments.”

Improving Work-Flow

The second solution comes from bringing new knowledge, new insights to the table. As Einstein would say you need to think in a higher dimension… What is a higher dimension? It is a different dimension. For example, in manufacturing we talk about batch sizes. One principle in the context of just-in-time and modern Kaizen is “Reduce the batch size to one, if possible.” Most scientists do not think of batch size or batch time in the context of their R&D experiments. But if they think through, there might be opportunities to reduce total experimental time.

Total experimental time = Time to run experiments + sample(s) prep time + test time

There are at least two approaches to handle this situation.

For convenience and labor efficiency, quite often, a scientist chooses to run the above experiments, sample preps and tests in three batches, serially.

Run all experiments first,

Prepare all test samples

Test samples, get and report findings/measurements.

If an experiment takes one day, and sample preparation and testing take another day, total experimental time would be 16 + 16 =32 days.

If however, these steps are done in parallel, total time would be 16+1 or 17 days, nearly 50% reduction.

How to reduce Time To Market across Multiple Projects?

Same concept can be extended easily across many projects.

Scenario 1: Three scientists are working on three separate projects, each expected to take three-man-years for completion. They complete these three projects in three years. Average project completion time = (3 +3 +3)/3 = 3 years.

Scenario 2: All three scientists work on Project 1; Project 2 and Project 3 are put on hold!

So Project 1, which required three-man-years to complete, is finished in Year1.

Similarly, Project 2 is finished in Year 2, and Project 3 in Year 3.

Average project completion time = (1 +2 + 3)/3 = 2 years. A reduction of 33% in time.

And since % reduction in time translates to earlier market entry, our profits go up!

Also notice, our key Customer 1 for Product 1/Project 1 is THRILLED, he got it in one year, instead of three. Customer 2 is VERY HAPPY, he got it in two years! Customer 3 is SATISFIED, he got it when it was promised, in three years.

By thinking in terms of serial vs. parallel process, and taking advantage of the opportunity to work in parallel, we can:

If an experiment takes one day, and sample preparation and testing take another day, total experimental time would be 16 + 16 =32 days.

If however, these steps are done in parallel, total time would be 16+1 or 17 days, nearly 50% reduction.

How to reduce Time To Market across Multiple Projects?

Same concept can be extended easily across many projects.

Scenario 1: Three scientists are working on three separate projects, each expected to take three-man-years for completion. They complete these three projects in three years. Average project completion time = (3 +3 +3)/3 = 3 years.

Scenario 2: All three scientists work on Project 1; Project 2 and Project 3 are put on hold!

So Project 1, which required three-man-years to complete, is finished in Year1.

Similarly, Project 2 is finished in Year 2, and Project 3 in Year 3.

Average project completion time = (1 +2 + 3)/3 = 2 years. A reduction of 33% in time.

And since % reduction in time translates to earlier market entry, our profits go up!

Also notice, our key Customer 1 for Product 1/Project 1 is THRILLED, he got it in one year, instead of three. Customer 2 is VERY HAPPY, he got it in two years! Customer 3 is SATISFIED, he got it when it was promised, in three years.

By thinking in terms of serial vs. parallel process, and taking advantage of the opportunity to work in parallel, we can:

And we have still not accounted for benefits of three people working together, sharing ideas, challenging one another and knowledge transfer from one to two others.

Use of microarrays for experimentation and Chinese innovation of creating towns around a singular manufacturing focus are two extreme examples of learning from work-flows.

So crank up your Product Development engines...

And let the fun begin!

In the next issue I will share another reason to be first to market...

1Ref: National Center for Manufacturing Sciences, Focus, Oct 1991