Are You Turning Data into Decisions?

Wikipedia defines data analytics as “the discovery, interpretation, and communication of meaningful patterns in data.” But what does this mean for agriculture? How can we break this down and turn it into meaningful insights in agriculture?

If we look at data analytics in four general categories–diagnostic, descriptive, predictive, and prescriptive–we can infer insights based on each category. As we progress through the categories of data analytics, we can move beyond diagnoses after the fact and start to be proactive and preventative.

  1. Diagnostic analytics deals with past and current information and can help applicators understand “how did a specific product perform in years similar to this one?
  2. Descriptive analytics involves mostly current information and may help a retailer, for example, in a case where corn in the area is in reproductive Stage 5, and s/he wants to know which products will sell best this week based on current information;
  3. Predictive analytics uses past, current, and future (forecast) information; based on the forecast, you might wonder what diseases are more likely to occur in Northern Italy next week. If you knew of a current pest occurrence, you could use predictive analytics to anticipate the progression of that pest emergence;
  4. Prescriptive analytics takes into account all available information and can help to determine course of action; for example, what chemical do I apply today and tomorrow, and in what quantity?

However, even before starting the discussion of data analytics and how it can help guide agricultural decisions, it is important to acknowledge that not all data sets are created equal. With so much data available, you must consider the data you have and the data you may want to acquire for your applications.

  1. Is your data relevant to your application? For instance, you may have a lot of soil moisture data; but if you’re concerned about a specific disease that is based on leaf wetness, that data may or may not be relevant to your application.
  2. Is historical data relevant or valuable? You may have data available for right now, but what you may really need to do is compare that information to the information from the past several years.
  3. Does your data take into account your location? You may have information for a location that is 100 kilometers from your current location, and for some, that information may not be valuable. However, if we’re talking about weather, that information may in fact be very valuable to you.

Watch Part I of “Analytics and Agriculture: Turning data into decisions” on-demand to learn more about how data, techniques, and expertise can help you make better agriculture decisions.  

Share this post!