Sensory Myths: Flexibility & Functionality

Just the other day, I was chatting with a friend in the industry, sharing in our woes regarding how much time and effort it can take to manage and maintain a strong sensory program; specifically the time it takes to set up, run, break down and analyze sensory panel data. Don’t get me wrong, managing sensory may be the most fulfilling and exciting part of the brewing process (and yes, tasting is part of the process). It is where all of your hard work planning, brewing, cleaning, fermenting, finishing, tracking and packaging ext.) culminates in the flavor of your final product. It is where you can take pride in the product you’ve lovingly created, knowing that it is indeed in spec and true to the brewer’s original intent. It is where you bear the responsibility to defend your brands, where you must put on your sensory science armor (usually a lab coat), and protect your brewery’s reputation by guarding the flavor of the beers that have brought your brand success.

Seems safe to assume we all believe that formally tasting your beer through a sensory program is worthwhile, but we’ve all felt defeated by our lack of time, telling ourselves: “I simply do not have the time to devote to sensory”. Yes, managing a sensory program certainly takes time (just like anything truly worth doing), it requires a consistent effort from the panel leader, and in a dynamic brewing facility where sensory is but one of many hats the panel leader may wear, this can pose a unique challenge.

After getting off the phone with my friend, I started thinking about the myths within the sensory community that have posed as giant, fabricated barriers to entry. This has frustrated me from the beginning; the notion that running a sensory program requires an all or nothing mentality - you’re either “in or out”, “with or against”. I too, am discouraged when I read papers from sensory academics that state you must train a panel for at least 100 hours before utilizing them to gain useful data (who has that kind of time in the craft brewing industry!?). This has left many of us with our heads hung low, sacrificing the good for the mirage of the perfect.

Now, I’m by no means condoning mediocrity! You can run a rock solid sensory program where you’re capturing meaningful and actionable data without devoting hundreds of hours to training, in the perfect booth space (with positive ionized air pressure, of course) and a PhD applied statistician/brewer/sensory guru at the helm. There are ways you can start right now. To use a shamelessly appropriate cliché, you can “do what you can, with what you have”.

During the conversation with my friend he had to put me on speaker on 4 different occasions so he could finish cleaning glassware before moving on to the cellar to help dry hop, filter and package beer. We spent most of our hurried 20 minutes discussing ways to more efficiently set up, run and analyze panel data, so that’s where I want to focus today’s post.

Here’s how I like to break it down, you have two different sets of questions to face as a panel, high risk questions and low risk questions. Understanding these questions and what it takes to answer them is the first step to properly budgeting your time towards Sensory.

High Risk Questions

“Is the beer True to Target and free of off flavors?” This is the question you need to be asking before every critical point along a beer’s journey. The further down the line it goes, the more money you have invested (especially in packaging), and the higher the risk to the brewery.

Clearly, high-risk questions require a decent amount of attention, but there are some tricks to skim off time if the proper system has been put in place. First, these questions should be answered in your formalized panel setting. If “formalized panel setting” conjured up images of serious panelists being pushed samples from a secret door under red light, don’t stop reading! All that is required for a “formalized panel setting” is a devoted space and time where your team can evaluate samples.

  • Where: The basic requirements for a sensory space is that it remain free of odors (the Brewhouse is out!), is quiet (run panel before your taproom opens) and is accessible to your team (nothing offsite).

    If you have a devoted sensory space that is easily accessible you’ll save time looking for a spot to land when beers need to be tasted, you’ll save time lugging samples to distant places and getting your panelist to travel to get there.

  • When: We are all creatures of habit and once we get used to doing one thing every day, it is hard to break the cycle. Play on our human/robot nature and hold panel in the same place and at the same time during a regular frequency determined by the number of samples you will be evaluating in your brewery.

    If your panelists know where to show up and when to go, you will save time trying to wrangle your busy crew at variable times of the day/week.

  • How - Setting Up Panel: Once you come in to the brewery, assess what beers will be moving that day (or a few days later) and schedule them to be tasted. Bonus points if you can run two or more of the same beer on panel and blindly randomize them within the set. This is easy if you have another independent sample that requires tasting (same beer, different tank for example). If you don’t have another independent sample ready to taste, you could use a “control” sample that has already been tasted, and passed panel. This will prevent any bias from the panelists to automatically “pass” or “fail” a beer if they have a hidden agenda or are privy to production information that may key them in on a potential defect (crafty subconscious brains!).

    Scheduling beers for panel is as easy as clicking and dragging using the “Schedule Tasting” functionality in DraughtLab Pro. There is no requirement to enter superfluous information; all the panel leader needs to do is spend less than a few minutes selecting brands for tasting, and tagging other important information about the sample. You can even have the software randomize the samples for you!

  • How - Analyzing Data: Data analysis for high-risk samples must be performed immediately so that a decision can be made to move the beer forward, or stop. Anyone who has ever manually entered sensory data (bless us all) knows that it can take hours of time and is incredibly prone to error, making your data weak and potentially compromising to the reputation of the sensory program. Data analysis does not have to take a long time if the proper system has been set in place ahead of time (see Batch-to-Batch Variation in Brewing: Let P-Charts Do the Work).

    At the end of panel, you can quickly bring up the tasting results and visualize where the sample falls within the brand’s normal variability while reading comments exactly as entered by the panelist.

Low Risk Questions

These questions include: “how are in-process samples (maturation) beers developing” and “how are packaged beers aging (shelf life)?” They encompass any sample at any stage that do not require an expensive “package/don’t package” or “sell/don’t sell” decision. These decisions can be made in a less-formalized (and less time-consuming) setting, so there is no need to wrangle your team and set up a formalized panel. These questions do not require as much ‘hands on’ time from the Sensory analyst. They’re like the recipes that take 5 hours to prepare but only require 30 minutes of hands on time (I’m looking at you Cassoulet!).

  • Where: Just like having a simple, devoted meditation space where you can go to clear your mind, so should you have a simple and devoted tasting space where the evaluator can physically remove themselves from the process, giving them the space to devote their full attention to the task at hand: evaluating the beer. This doesn’t have to be anything fancy, simply a corner in the cellar or a special table by the loading docks will do just fine, but removing oneself from the production floor is an important step in conducting a mindful evaluation.

  • When: Whenever your tasters are available and focused on evaluating. You may be at a brewery that has three shifts, each run by one or two people at a time. If you want to get everyone to taste the same sample, you can set up a timeframe in which your tasters can evaluate the beer when it is convenient for them.

    This will save you time having to wrangle everyone to the same place at the same time. Setting up the tasting is as easy as scheduling it and setting a timeframe in which the sample must be evaluated. The samples will then appear in your panelist’s “Available Tastings” box in DraughtLab Pro and your panel can evaluate within the allotted timeframe.

  • How - Setting Up Panel: In this “low risk” case there is no devoted panel time, so as the panel leader all you need to do is alert your tasters which samples need to be evaluated, how they need to be evaluated and by what time.

    Using DraughtLab Pro, you do not need to physically find every person and explain what needs to be tasted. All the panel leader needs to do is spend a few minutes scheduling beers to be tasted; and all the panelists need to do is select the beers they are tasting to begin their evaluation.

  • How - Analyzing Data: Once the allotted time has passed, the panel leader follows the same robust analysis procedure described above and in Batch-to-Batch Variation in Brewing: Let P-Charts Do the Work.

    The panel leader should be able to quickly check a brand’s results against the normal variability, while reading comments exactly as entered by the panelist.

Our breweries thrive on our brands, and our brands stand on their flavor, and we honor our beers by lending them our time and devoted palates. Running a sustainable sensory program is all about biting off exactly what you can chew, while skirting the bogged down, overwhelmed and defeated mentality created by the image of the perfect program. Use your time wisely, choose the most critical tasting points to evaluate, and use a system that allows for flexibility without skimping on functionality.

Cheers, to growing your culture of quality one panel at a time!