Empower 3 - are custom fields commonly used in industry?

Empower2018

There is a lot of information on custom fields for Empower 3 plus training courses offered by Waters but from my own experience, their use varies a lot from one company to the next. My first job was QC lab and they were scared of it, they didn't understand it plus nobody had the skillset. Another job they were very widely used but were often too complicated or they crashed. 
Im asking Waters staff here or people with a lot of industry experience- are custom fields common in industry and is their a lot of knowledge from clients on how they operate etc? From my own experience, they take at least 2-3 years of constant upskilling and practice to get the hang of them, much more than any other element within Empower 3. 

shaunwat

I've been in the R&D pharma / medical device industry for 20 years. Approximately six of those 20 years has seen me serve as an independant Empower consultant for others. At many of the companies for which I have worked as a scientist has also seen me placed into an Empower administrator role where core responsibilities were end user calculations and reporting among others.

Your assessment is pretty much correct. Custom field use varies wildly from company to company. Many that do not use them simply don't know how. Those that do use them most often could be doing a lot more with them or the custom fields are very poorly coded.

As a result of the above, many have chosen the route of validated excel sheets or manual calculation. When I was a consultant I always told clients: if you can perform your calculation in Excel, then I can do the same in Empower.

So, in my experience custom fields are generally underutilized and there is next to no knowledge from clients on how custom fields truly work.

Overall, I find this disappointing because custom fields are so easy to learn. Everything in your routine testing environment needs to be turnkey anymore and if it is not, then you are losing money. Your bench chemists need to know how to setup and operate Empower. Once their analysis has completed and their data has been processed there should be no additional ancillary work that they need to perform (such as disso calculations, content uniformity, further impurity processing, etc, etc).

Empower2018

Thanks very much for your reply Shaunwat. Its interesting to hear that you had the same experience over different companies. I know with the last company I was with, I noticed they constantly used real numbers in their Bool/Enum fields such as GTE(Amount, 0.157) or a Result Bool like EQ(Average(Custom_Field),0.150) when using CConst fields is much more practical. The two biggest barriers I have come up against in labs when trying to implement change using CFs were 1. An unwillingness to learn new ways of doing things from long term staffers who were so used to keying in figures into Excel for years and years that they couldn't comprehend changing the system , even if it were proven more efficient. So I heard a lot of statements like "That's a bit complicated isn't it?" or "We are doing it this way for years with no problem, why change?". Not to mention some people are simply not interested in learning new things. 

2. The lack of time to train others and gain a deeper knowledge- most labs are just too busy to sit down and change systems so it takes a lot of pushing and dedication to plan even small changes to the established system. 

Im not sure I would agree with you about CFs being "so easy to learn". I found them quite tricky when I was trying different types. Yes, the basic ones like Area/CCalRef1[Area] are very easy to understand but when I got into intersample summaries and Enum functions, it took a lot of study, trial and error before it would click and im still learning. Particularly hard for me to grasp was (and in some ways is) the differences when you omit syntax in intersample summaries for example no Label equals current sample, no channel equals current channel, but no injection only means the second to last result plus other events like using Boolean functions in real formulas, it took me quite a long time to get where I am now. Shame there is no book available on exactly how they are coded but I don't imagine they would ever release such a thing!

shaunwat

I see what you mean...

I found that I was able to make my way through the learning curve that is custom fields fairly quickly because at the end of the day it is a mathematical construct and may be manipulated as such. There is true beauty in how Waters designed / developed custom fields.

It might have also helped that I came out of undergrad with a dual major (chemistry and math).

MJS

I think that there's a significant under-utilization of many of the advanced features that Empower has to offer.  In addition to the custom fields, electronic signatures utilization is very weak in my area, based on our service engineers, and nearly unheard of with regard to implementation with the SQT testing as I apparently am the only customer my service engineer says uses ES for the SQT work as everyone else just prints the re-qual reports (note I use it for both software and system QT even).  I think that for these advanced features, while the current system may work for people, the main driver is where to spend your time.  If you want to use custom fields and implement advanced features to their full extent, you need to invest time up front to develop the formula and possibly validate it where it can be hard to justify.  Dealing with data in Excel later can be justified as a necessary evil or the "cost of doing business" to some extent where it can be hard to quantify the benefit of implementing the automation in Empower.