Erin Hightower, an agronomist for John Deere dealer RDO Equipment who is based in Kennewick, Wash., often uses a sports analogy to illustrate the first step to successful data management.

“The best teams start by having honest conversations with themselves about where they need to improve during the off-season,” she says.

Data management starts with identifying the right farm management software for your operation. There’s no one-size-fits-all, but it’s advantageous to pick software that has growth potential, Hightower says. She likens the process to her own experience with financial software.

“I spend a little extra money for a really good personal financial software,” she says. “The reason I paid a little extra is because I’m choosing a software that continues to grow, one that’s always coming out with new features. It’s the same thing with farm management software. If it doesn’t grow with the times and the technology, then there will be an issue because we need a farm management software that will keep up.”

Hightower identifies a few basic prerequisites for a successful data management plan. First, make sure the data you’re compiling is in a format that allows you to easily obtain analytics.

“Data is only valuable if it’s usable,” Hightower says. “I worked with a farmer who refused to have his data in one place because of his mistrust of cloud-based data. He used John Deere, FarmWorks and Precision Planting software. His harvest data was in one place, and his planting data was in another. He couldn’t tell if something was working or not. It's hard to have those deep, analytical conversations if you have all of your data in a million different places or binders.” 

It’s also crucial to have an objective for your data, Hightower says. Determine what you are trying to measure. Examples include variable rates for fertilizer and seeding or timing and field location when it comes to planting. Don’t forget to back everything up at the end of the season in case of a catastrophe.

“I recommend farmers line all their tractors up in a nice row, put a stick drive in each of them, save the data, label it with the year, and store it in a drawer,” she says. “Then you’ll have a copy of everything in case something accidentally gets deleted.”

Measuring Data Correctly

Effective data integration means managing data 365 days a year and avoiding selection bias along the way. Selection bias can lead to overly optimistic beliefs because multiple failures are overlooked, as Hightower illustrates with an example from World War II.

“They were measuring the damage on airplanes that made it back, so that they could build a stronger plane,” she says. “But they weren’t measuring the planes that didn’t make it back. Crucial data from fatally damaged planes was being ignored. We’re not fixing the problem if we’re only measuring the things that come back successfully. Stay focused on your weaknesses because you can’t know your weaknesses if they’re not being measured.”

Sometimes the data you need is the data you’re not collecting. Hightower has noticed that some of her customers won’t collect data on certain passes that appear insignificant, like when they’re mowing a field. But what if that mowing pass ended up spreading weed seed? They didn’t measure it because they were using their selection bias on the sprayer.

“I’m a bit of a hoarder when it comes to data,” Hightower says. “I’d rather have more than less. As we start looking at new technologies like variable rate applications, we don’t know what data could be important in the future. You may not need the data today, but you will tomorrow.”

Monitor Data in Real Time

Daily check-ins with data on a mobile device can help prevent countless mistakes, Hightower says.

“When one of my customers started to use the remote display access function to do a check-in with a planter, she discovered that it was set up with the wrong crop,” Hightower says. “Checking in every day on the data is what allows a farmer to catch errors, react to external factors and build the audible that will make turnaround plays after the 7th inning stretch.” 

Farmers should also ask themselves if the data they’re collecting needs to be acted upon quickly or if it will require long-term action. Hightower says some customers get frustrated when they collect data and aren’t sure what to do with the information right away.

“Sometimes they’ll say, ‘I have this pretty map. Now what do I do with it?’,” Hightower says. “Be prepared, and be patient if the answer is, ‘You don’t need it this season.’” 

It’s crucial that the data is easy for everyone in the operation to read and get an answer from, Hightower adds. She has a customer who used John Deere Apex data management software almost until its discontinuation in early 2024. His son, who lived 30 miles away, was frustrated because he couldn’t monitor his dad’s Apex data on his own computer.

“Since Apex was shutting down, I finally got them hooked up to the Deere Operations Center,” Hightower says. “Now he has his own login, and he can call his dad and say, ‘Did you see that corner of the field over there? What do we do about that?’ They’re able to have those quick decision-making conversations because they don’t have to be around the same computer to see their data.” 

The real magic happens when the data is used to formulate a plan. Data points give you a book to read, but they don’t read the book for you, Hightower says. It’s up to the customer to interpret the data and figure out how to use the information to solve a problem.

“Data is as much of an art as it is a science,” Hightower says. “Data will not replace a good agronomist. You need to be prepared to take specific actions with the data. Learn how to read the data and find out what works for you.” 

Hightower uses the S.M.A.R.T. (specific, measurable, achievable, relevant, time-bound) model to create goals using data. Farmers should build goals as they go and address questions that arise along the way, she says.

“It’s also important to recognize if that goal needs more than 1 year to be accomplished,” Hightower says. “For example, during the early years of no-till in Washington, we had a researcher who tried it and said it didn’t work. But what we didn’t realize was he was moving the no-till plot every year. So, he didn’t see the long-term benefits of no-till because he was tilling every other year. Keep in mind that sometimes goals are time-bound.”

Data Cleaning 101

At the end of the season, data management is kind of like Christmas lights, Hightower says. Some people never take them down. Other people throw them in a pile and don’t worry about them until next year. And then some people put them away meticulously and can use them the following year without cursing or scaring the cat.

“Sometimes people are collecting data and they’re never looking at it,” Hightower says. “Some people are looking at the data, and it looks like a mess, so they just hide it and hope that it makes more sense next year. And then there are the people who are constantly managing and looking at their data.” 

Hightower adds that it’s important to become familiar with post calibration tools, basic data file rescue treatments and learn how to format and restore corrupted data. Making sure the software is updated in your displays is another crucial step to take before planting season.

"You have to update your software in order to access the new features that come out from farm management software companies that go into displays. When you clean out your display, you might discover all sorts of new toys in there that you didn't know existed before."