Page 1 of 1

Satisficing: A Behavior That Drives Corporate R&D

PostPosted: Fri Dec 11, 2015 10:17 am
by Dustin Levy
Satisficing is a term I never heard of as a graduate researcher, but have since learned that it drives many behaviors in industry. This term, a combination of “satisfy” and “suffice”, was introduced by Nobel laureate Herbert A. Simon in 1956 and is a decision making strategy that involves searching through available alternatives until an acceptability threshold is met. Business managers who seem to make decisions based on limited or incomplete information instead of taking the time to understand all available facts and data are often satisficing. These managers have an acceptability threshold in mind and they’re off and running once they have enough data to meet that threshold. Their decisions will usually turn out okay as long as the acceptability threshold was accurate and they were made aware of key risks before making the decision.

Researchers are trained to seek truth and perfection. They spend long hours in the lab and library and rarely settle for “good enough”. Business managers, on the other hand, subscribe to the theory that “perfect is the enemy of good” and as soon as a product or service is good enough to make money, they accept its imperfections and sell it. Apple is on its 6th generation iPhone and each version has had its issues, but Apple has made a lot of money along the way.

As a technical contributor in a R&D company there are times when it is appropriate to satisfice. As a rule of thumb, you satisfice more the further downstream you are in the R&D process, i.e., closer to the point of delivery to a customer. If you separate R&D into three main activities: research, development, and validation, then research is the most upstream and validation is the most downstream. Industrial researchers spend their time on core technology development, have less-strict deadlines, and more freedom to strive for perfection. They hand off these technologies to developers who implement them into products and services that can be sold to customers. Developers have some freedom to be creative, but are held more accountable for hitting milestones and deadlines by project managers who drive them to move onto their next assigned task when the current one is “good enough”. The validators are closest to the point of delivery to the customer and are responsible for verifying that the product or service meets the minimum requirements to be sold. This job is the most black and white and has the least room for creativity.

In the context of the definition of satisficing:

Researchers spend most of their time developing alternatives, Developers spend most of their time choosing and implementing one of the alternatives, and Validators spend most of their time verifying that the implementation meets the acceptability threshold

The tendency of corporations to satisfice has 3 implications for early-career scientists:

1. Choose the right path. If you value creativity and independence, quality control may not be the right role for you.

2. Interview the right way. If you’re interviewing for a general role in R&D, be sure to give examples of when you’ve operated in a creative open-ended mode, but also when you’ve been very task-focused when a deadline had to be met. You may miss out on an opportunity if you present yourself as too one-dimensional.

3. Do the job asked of you. If you’re in quality control, don’t hold off on a product delivery because you have an idea to make it better. Yes, you should share your ideas at the appropriate times, but your job is to authorize delivery when the product is good enough.

Re: Satisficing: A Behavior That Drives Corporate R&D

PostPosted: Sat Dec 12, 2015 11:30 am
by Ana
I partly disagree on your view Dustin of satisficing being a characteristic of industry that goes against researchers nature.

You say “Researchers are trained to seek truth and perfection. They spend long hours in the lab and library and rarely settle for “good enough”. Actually that is a bit of a stereotype, most academics I’ve known, which should not be under the same business pressure as industrial scientists are, see their job as the creation of publications and data for grants because that is how success is measured, by publications/impact factor and funding received. Many publications are based on (quoting from your post) "limited or incomplete information” which is why it is so hard to reproduce many publications, and why many seek minimum publishable units to inflate their CVs as they nurture a few good big papers. I think researchers are trained to both seek truth and perfection (which will make a big publication) and to seek those minimum publishable units/salami slicing publications that meet the publishable threshold and help you get more papers our, get your studying a PhD… etc.

I think that doing just enough to get what you need (paper/drug advanced) is a normal human behaviour when confronted with limited time. Not really liked to one specific field.

That said your main message is that different jobs have some different quality requirements, some requiring the use of standard operational protocols while others are a lot more creative, and that you should know yourself and the specific job you are applying to and make sure that aspect matches. And that's a very important message.

Thanks for your review posts, they are very interesting,


Re: Satisficing: A Behavior That Drives Corporate R&D

PostPosted: Sun Dec 13, 2015 11:47 am
by Lydia
I have not heard that term before - to me, it sounds more like a combination with sacrifice than with suffice.

Re: Satisficing: A Behavior That Drives Corporate R&D

PostPosted: Mon Dec 14, 2015 9:23 am
by BMK
Thanks Dustin for another informative post! One thing I thought I might comment on

Ana wrote:I think that doing just enough to get what you need (paper/drug advanced) is a normal human behaviour when confronted with limited time. Not really liked to one specific field.

I think the real issue with an Academic's ability to satisfice isn't so much that the concept goes against their nature (as Ana deftly explained); in my opinion, it's more that their perception of time itself is skewed: when it comes to publishing, they don't BELIEVE that there is limited time. I'm sure everyone who published during their PhD program can recall their professor/advisor going "let's also do X, in case the reviewers think Y", and this "let's also" can sometimes feel like it is repeated ad nauseam (never mind that there is ALWAYS something in a paper a reviewer can harp about if they really want to, every study has limitations). Without a hard deadline (like grants have) things have a possibility of never reaching the top of the to-do list, which in effect yields Dustin's outcome/synopsis, even if the root-cause is slightly different.

This issue happens with both MDs and PhDs in my experience; it's a very pervasive habit in Academia, otherwise the old adage "give them a hard deadline to reply to you by or your going to do/submit X" (almost like a threat) is so often handed out to people dealing with a stubborn professor/researcher. That said, dealing with it as a postdoc/junior scientist gives one valuable experience in "managing up" (and also becomes a great answer to certain behavioral questions as well, hint hint).

Lydia wrote:I have not heard that term before - to me, it sounds more like a combination with sacrifice than with suffice.

Contextually, I think the concept of this portmanteau works with either pair of words: you sacrifice the gain from doing additional work/experiments because the current product/results suffice (I actually initially thought of sacrifice and suffice as being the pair at first glace as well, until I realized there was no "t" in either word). Just my two cents.

Re: Satisficing: A Behavior That Drives Corporate R&D

PostPosted: Mon Dec 14, 2015 11:44 pm
by Dustin Levy
Thank you all for your very interesting comments. Much has been written about the need for companies to go-to-market with “good enough” incremental product innovations instead of waiting for perfection or near-perfection (see Agile/Scrum, Lean Startup/Minimum Viable Product literature). Ana’s observation of research groups working toward minimum publishable units is consistent with this line of thinking, but I wonder if research students, particularly who land in industry, are learning these behaviors. Is it possible that the MPU approach is being driven by PIs who have learned how to work the system, while many students maintain, in BMK’s observation, a skewed perception of time? That those students who learn how to play the game get grants and PI roles and those who don’t land in contributor roles while struggling to manage their time independently?

I still have an image of researchers being trained to seek truth and perfection, still held to a very high quality standard for their dissertations, while PIs pluck off MPUs along the way. These PIs were also trained to be a truth-seeking, but have adjusted their approach in response to their incentive system, much like a scientist/engineer turned business manager. If the PI is handing out arbitrary deadlines to achieve MPUs at a high rate and inflate CVs, then their students probably aren’t learning great research skills, nor learning how to self-manage. This is a disservice when it occurs.

Re: Satisficing: A Behavior That Drives Corporate R&D

PostPosted: Fri Jan 08, 2016 3:33 am
by D.X.
Hi, a bit delayed but I'll add one point. Again be careful with the bucket term of Corporate R&D, because it may not apply to Pharmaceutical sector.

Satisficing in Pharma R&D in Terms of a driving our products forward cannot accept "satificing". Our products are going into humans and with the benefits they have, they also carry risks (safety). On the technical side, that risk (safety), can be linked to Quality.

I have not experienced our Industrial Operations Teams (where our Technical Operations Teams can sit)demonstrate Satisficing. Delivery of our products demands that we are fully aligned and very much within specifications as mandated by regulations (ICH, GMP, GLP, CHMP, EMA, FDA). In fact, i've seen quite the opposite, the technical Teams can be very active in ensuring that Quality is best as possible - most times they have increased activities that are actually providing diminished Returns on what can be acheived Quality wise. In fact some times we get the opposite, they push and push until we're out of specification. The behavior then is not "good enough", the behavior is an evidence-based, in depth, Analysis to derive a conclution of "we can't improve, we're best we can be with current Technologies and procesess". There is a Balance to be achieved but its not satificing - our regulations can be quite absolute. If there is satificing, then it is within very very narrow specifications/regulations suffice to say its not satisficing, its complying and knowing our Limits.

In Pharma R&D we can't afford satificing - there is sufficent history (in recent times) to Show when that was done, there was impact to Patient lives. So in conclusion, I would not Support that thre is a perception that satisficng is a behavior that is well accepted in Pharma R&D where Rx/Products are in Focus.

Diagnostics and Device divisions may allow for a satisficing culture based on the products and Business landscape with associated risk-tolerances they may operate in.


Re: Satisficing: A Behavior That Drives Corporate R&D

PostPosted: Fri Jan 08, 2016 6:53 am
by Dustin Levy

You point regarding the Pharma industry is well taken. A major reason that this industry is so highly regulated is to avoid satisficing. In the absence of that regulation, Pharma companies would cut corners too often. There's a natural progression over history - innovation occurs within an industry, companies compete for profits within that industry, the public may be harmed in some way by that competition, government regulation ensues. Companies satisfice and exploit loopholes in those regulations, public is harmed, more regulation occurs. Before you know it, the industry is tightly regulated by a huge number of three-letter acronyms (ICH, GMP, GLP, CHMP, EMA, FDA).


Re: Satisficing: A Behavior That Drives Corporate R&D

PostPosted: Mon Jan 11, 2016 3:17 am
by D.X.
Hi Dustin,

I would agree that regulations curb satificing, but I would challenge that many regulations were not generated for that primary purpose. Most were put in place to adjust for Managing evolving ethics, scientific thinking, and societal expecations on changing behaviors.

If we look at ICH, that was born out of Need to protect human rights from harmful and non-concentual experimentation - the Major violators here were on European and US soil and was not the industry at that time really. The Evolution was driven by a Need to harmonize the conduct and collection of data.

If we look at Drug Safety/Pharmacoviligence, that was born out of tradjedy, as in the case of Thalidamide. We didn't know back then, scientifically, that there was a Need for saftey monitoring in the Long-term, it was not routinely done both in the academic, government or industrial Settings. Regulations were Put in place to prevent such Events from Happening again and have been evolving, not because of satificing but due to learnings, for example we learned more about how to capture more robust safety Information in Trials, not due to satisfising but due to increased knowledge on what we should be looking at. If I look at a clincal study conducted in the 1980's vs. today, safety wise (and efficacy wise too) there is a huge difference on safety data capture, again not due to satisficing but due to evolved scientific thinking where regulations have followed.

I can Keep going on, definately I agree the reguations prevent such behavior, but most where not born out of preventing satisficing. One I can think of that was born out of preventing satisficing was the EMA Guidelines on Safety Monitoring Plan/pre-clincal package before First in Human study.

Ist not to say there wasn't and isn't bad behavior but alot of those behaviors were not due to satisfying, it was just more..plain old bad behavior.

A key benefit of many regulations is that reduction of satisfcying, i agree but the Drivers for most had nothing to so with satisfcing. More reacting to evolving changes in ethical views, emerging scientific understanding, and changing societal expectation - with some bad behavior in there - and on the Topic of bad behavior, its not just industry, remember it takes 2 to Tango.