Data Aggregation – understanding what it means
Bimini1002016-12-24T14:45:12-05:00In the past five years, one key word has made its way into the farm vocabulary – data. For some it’s a four-letter word in the worst sense of the meaning, given that it’s an item that needs a whole new kind of management. Yet there’s one phrase that often comes along with data – data aggregation. And understanding what that means may help you better understand your options when looking at data agreements, and determining how best to make use of all that information you have on the farm.
First, the idea of data aggregation is far from new. Farmers have been part of programs with aggregated data for decades. The farm business farm management associations across the country strip off identification of data and create useful tools for benchmarking your farm against others of your type across a region from a financial standpoint. Farm Industry News has discussed benchmarking before, and with the right information it has value for a lot of farms. Yet good benchmarking comes from a great database of good aggregated information from a wide range of anonymized sources.
Second, getting a handle on the value of your own data and its potential worth to your operation is important. However, keeping your data locked up may not be the best move for a long-term play in maximizing farm income and management.
Aggregated data – in essence – is a set of information from a wide range of sources. In this case it would be farms across a region. Key information – like specific location, and name-linked data is removed. What remains is yield on specific soil types for key hybrid types, amount of fertilizer applied, even the crop protection products used. All of this in a single farm has some immediate value, but what if you could do some comparisons of your farm to others?
The key is that your information is anonymous in the master “pile” of information, yet you and others can make decisions from that information. The key is that the database has to be big enough for your area (two farms in a county isn’t very anonymized if one is yours). And the information has to be more than basic “stuff.”
While there is the constant concern of data privacy out there, without aggregation into a trusted location, your farm information is kind of like a data island. It’s the bigger picture patterns you need, from actual field information. Does applying a fungicide on a corn crop at V5 to V8 make sense? Look at the returns for farmers across the Midwest who made that decision – that’s possible with data aggregation.
How does population make a difference in final yield? Another question that can be answered through data from hundreds of farmers planting specific hybrids and reporting that information into a centralized data set that can be mined for information. There are farm data aggregation systems out there – Farmers Edge, Farmer Business Network, FarmLink, Granular, Conservis, CropZilla, and others – that aggregate data (maintaining privacy of individual information) to create valuable decision-making ‘engines’ for the farm. And that’s just a few, there are others, but wanted to list a few examples.
Another way data aggregation is at work is with the machine data collected by the major manufacturers. Specific operating parameters of every machine in a specific brand’s equipment is captured for use by the manufacturer to determine issues of warranty concern, efficiency and opportunity for future product development. For example, what is the average idle time of a combine at peak harvest? How many hours per year (on a percentage basis) does a specific tractor go in for service, and why?
Those are data points that could mean something to your individual farm, but for an engineer developing a next-generation machine, they can mean plenty. Individual farm information – so an Arkansas tractor with 3,000 hours isn’t identified as the Jones tractor in Arkansas, it’s got an engine with 3,000 hours and X of those hours were at idle – and that X percentage can say plenty about how the machine is use. So do failure rates across a broad product line across a big geographic region.
These days, credit card companies can tell you what you need next. Your medical history – not personal information – is parsed to see how communities of people differ for different health problems and what can be done about it.
Aggregating data isn’t a violation of your farm’s privacy, yet you should work with partners you trust to keep what’s private, private. For the future, this becomes more important because those that play in the aggregated data space will have a leg up on those that don’t when it comes to fine-tuning management, and capturing profit in this business.