Tuesday, December 16, 2014

Anyone heard of paying chemists by the reaction?

The different ways to pay chemists:

1. A yearly salary
2. An hourly wage
3. By project (i.e. deliver X grams by Y time for Z dollars)
4. Subsistence wages, supplemented with food and alcohol (kidding)

Has anyone heard of paying people by the reaction? If so, how did it work out for you? How much did you get paid?

Seems to me that this is a bad deal for any chemist. (How do you determine how you get paid? Do you have to work up the reaction, or just set it up? How many stir plates do you get?) Surely there is some kind of labor law about piecework that might apply to such a scheme.

I haven't heard of this before, so I thought I would ask. Readers, what say you? 


  1. Not heard of this before. But one thing springs instantly to mind, very large compound libraries, you know the ones with hundreds of thousands of compounds, if not millions. Even at 1 cent/reaction this is serious money.

  2. I worked with a CRO that made custom compounds and only charged us if they successfully delivered the compound at our target purity. It was a great deal for us. They didn't really vet the compounds we asked for, and many of them turned out to be difficult or impossible to make, so I always wondered how they could make money. After a year of working with them, I found out they paid their employees based on the number compounds made (i.e. they only got paid when they made the compound). We were much more careful with what we asked for after that...

  3. Depending on who chooses the reactions, this could work out very well for chemists: I'd be all over the Schotten-Baumann, and would steer clear of anything requiring degassing.

    This system (at least used to be at a company, now defunct due to idiot management, I worked for) has been, in effect, used by crappy little btechs that tracked how many "new" compounds each chemist registered and submitted for testing and included this as a criterion in performance reviews/bonus. The obvious outcome was people choosing less complex chemistry and registering every possible intermediate, which then got tested regardless of how senseless ("wow, that compound really inhibited cell proliferation, can you make more of this sulfonyl chloride?").

  4. At my previous employer (large Pharma) some groups were not "paid" by reaction, but were evaluated by the number of reactions that were run over the past year. So, at the year-end evaluation, they would simply put down the number of reactions that they ran (and compounds registered) for evaluation. This, obviously, is a bad idea - you get out what you put in (i.e., people would simply run libraries for the sake of libraries to get the reaction count up and register every compound, whether it was a final compound or intermediate). My group was evaluated by "impact" to the project (which is subjective, I know) but felt if done correctly was the way to go. Needless to say, my group was not the most "productive", but we were almost always the inventors on patents since more thought was put into the compounds that were made, for what it's worth.

    1. My company, although they won't admit it, still uses compounds submitted and reactions run as a criteria during the stack-ranking that happens this time of year....

  5. I worked at a CRO that did a project for one of the big pharmas. They paid per compound with the contract price going to the lowest bidder at a global auction. About half of the designs were fundamentally flawed, yet projects still got bid down to numbers that could only be profitable in chindia. We went after the easy stuff first, but it was clear some bidders were using the Jackson Pollock approach.

  6. From a purely management point of view... paying by reaction would be hard to do well. You get what you reward, and if you're rewarding number of reactions, you aren't rewarding quality unless you build some sort of quality measure in... by which time, you're probably back to something that looks a lot like paying per project.

  7. Hmm, so would a deprotection/cross-coupling count as one reaction or two? Do I get 1.5x credit for a Sandmeyer?

  8. I had a similar experience as Lyle Langley at a big pharma. Some other people would submit obviously toxic compounds for bio testing (alkyl iodides, etc). I never did as I felt it was wrong to waste time and resources doing so. Once my manager gave away submission credit for 80+ compounds I made to another person who did not bother to thank me for synthesizing them. That person had a very loose connection to the work in contributing to the core the team had been working on for several months. All the analog design and synthesis was my own. Purification was done by analytical - who never got submission credits so did not need them.

    I was laid off when they shut down our division. I was asked to look for another job internally - but did not bother. Happy to be out of that employer.

  9. I worked with a CRO and their main evaluation strategy was to count the number of reactions a chemist would perform. Do I have to mention that this is a really lousy and bad strategy. First, how can you compare 20 g reaction with a 5 mg scale reaction? And yet they count the same! The difficulties with the work-up cannot be compared.
    Some customers ask for impossible to make compounds (computer-generated) so CROs will waste time trying to make these. In the end they wind up losing lot of money, and seeking constantly cheaper and cheaper work-force.
    If you want to be treated well at work, don't study chemistry. Because of constant R&D savings, chemists can never work good enough. You will have PhD but will be treated worse than a garbageman.


looks like Blogger doesn't work with anonymous comments from Chrome browsers at the moment - works in Microsoft Edge, or from Chrome with a Blogger account - sorry! CJ 3/21/20