* Extended comments are highlighted in blue.
Recently the co-op that I work with, Tillamook County Creamery Association, decided to change the way it pays producers for milk-quality premiums.
Historically, it had in place a fairly aggressive program that pays significant premiums for higher-quality milk. Frankly, it is one of the best examples I know of "getting what you are willing to pay for” in that it has resulted in one of the most consistent, high-quality milk supplies that I'm aware of in the country.
Prior to the change, we monitored each incoming load of milk for somatic cell count and did weekly standard plate and preliminary incubation counts for each farm. These counts were then averaged and the averages compared to a framework in which lower numbers pay higher premiums.
But about two years ago, we began using a technology called BactoScan for assessing bacteria counts. BactoScan uses the presence of bacterial RNA and DNA to assess bacterial levels rather than the traditional count of colonies that grow.
While there is a strong correlation between the two strategies, they are different enough that the numbers they generate don't always agree. In addition, BactoScan allows us to do a bacteria count on every load of milk as it comes to the plant rather than doing weekly counts. That, in turn, allows us to pay premiums on a daily basis.
Managing milk quality
is a process, not an event. Part of the process involves generating data. However, this data is not particularly useful unless you have made sure it is organized and presented in a timely manner. You also must know how to interpret the data in a way that contributes to problem resolution.
The process should end by defining the root cause of a problem, a corrective action and a preventive action. Simply put, there needs to be a system in place that efficiently translates data into information into action. A program that pays premiums for high-quality milk must be paired with an aggressive support system to help pinpoint and resolve problems when they occur.
So when you switch to a different strategy for accumulating data, as we did with BactoScan, you need to go through a process of reinterpreting what this data is actually telling you. If you don't, you run a high risk of wasting time solving problems that don't exist or not focusing fast enough on problems that need attention.
We have learned
a number of things from our new system. We have shifted our attention more toward watching bacteria counts, and we spend less time looking at SCC. This may be partly due to the fact that the average SCC across all the milk we receive runs in the mid-100,000s/ml.
We see a high correlation between the daily bacteria count information and SCC counts.
In many cases, we will see bacterial activity before we see significant shifts in SCC. In fact, I could argue for the bacterial status of the milk being more significant to postprocessing product quality than SCC.
We have found that while weekly bacteria counts gave us useful information, they by no means captured all the dynamics. We frequently find situations where the weekly data missed significant events that the daily screening catches. Having more timely data has also contributed to faster resolution of problems.
The chart above shows the types of organisms that we found across several thousand elevated raw bacteria counts gathered in a period of two to three years. Most, if not all, of the Strep counts, as well as the coliform counts, were, in our opinion, cow-sourced rather than equipment-related.
Controlling farm milk quality