Bad data means bad insights. This can lead to bad business decisions and, in turn, lost revenue and missed opportunities. To avoid these potential losses, you need to ensure you have the highest quality information possible when you evaluate your farm data.
Agribusiness professionals are chomping at the bit for data to inform their business decisions. But getting high quality data doesn’t just happen. It requires a concerted effort on your part to ensure the data you’re investing in is worth it:
- The data is accurate both at an aggregate and grower-by-grower level
- The data has maximum coverage of the market
- The data is detailed enough to provide helpful and relevant insights around prospects and customers
- The data is recent enough to tell you what’s happening on the farm and in the life of the grower right now
The better information you have, the better decisions you’ll be able to make. That’s why you should consider all four of these criteria when you evaluate your farm data.
The number one thing you should look for when choosing data is accuracy. If the data isn’t right, it won't be helpful. In fact, it could be detrimental.
The most effective way to evaluate your farm data is through its sourcing. For instance, if they used primarily surveys to collect their farm data, that would be a problem. Here’s why:
- Surveys incentivize inaccurate reporting – both under- and over-reporting – because of the incentives associated with them
- Surveys are subject to selection bias, where only those interested in the survey end up reporting, leading to a skewed and limited number of respondents
- Most surveys can’t account for overlapping acres, or the distinction between land owned versus operated
Any collection methods that rely heavily on surveys, polling or anything similar aren’t going to deliver reliable results.
That doesn’t mean that surveys are bad or don’t offer value. When put into the proper context, surveys can provide unique insights that typical objective means of collecting data can’t do. But if you’re just leaning on surveys without any way to verify that the information that’s been collected is correct, then you’re going to run into problems.
Remote sensed and ground truthed crop data--like the cropland data layers (CDL) provided by the USDA--help to determine what’s in the ground and, when intersected with attributed farm field boundaries, can tell you who’s farming what. Publicly available financial and tax records can give you insight into the grower’s bottom line. And public information can also provide ways to determine the connections and relationships among various growers.
An example grower showing CDL intersected with attributed farm field boundaries.
There’s no reason to lean entirely on surveyed data when there are other sources that give you accurate information and insights.
When it comes to data, both breadth and depth are important. That’s why coverage, as well as detail, matters.
If you’re engaging in high-level, aggregate market analysis, you need to have as much coverage as possible. Otherwise, you’re going to be analyzing a sample that’s not representative of the market and could, thus, lead to inaccurate interpretations of the market.
But even if you’re using the data for marketing or sales purposes, having less than stellar coverage could mean you’re missing out on some key opportunities:
- Maximizing opportunity. Plain and simple: if your data provider has more coverage, that means that you have more opportunities to find new customers.
- Identifying new markets. If your data provide only provides partial coverage--say, 60 percent--how do you know that additional 40 percent doesn’t contain a new target market you could go after?
- Matching to current customers. Wallet share expansion is a key part of any ag revenue growth strategy. The less coverage your data provider has, the likelihood that you’ll be able to append their data to your current customers--thus giving you more insights to aid in expanding wallet share--is going to be lower.
Even if you’re going after a highly targeted segment of the market, the wider the coverage of your data, the greater chance that that data will get you closer to where you need to go.
Of course, coverage alone isn’t enough. The USDA has near-total coverage of the market and provides aggregate planting figures on a state-by-state basis. But what they don’t make publicly available is detail down to a grower-by-grower level.
The more detail you have on growers and segments of growers, the better able you’ll be to make strategic decisions, create custom marketing campaigns and engage in enticing sales conversations that drive farmers to close.
Here are just a few types of data you should go after:
- Basic information. Start with the basics: contact information for multiple channels--email, address, phone--as well as the name of the grower and their business entity (if applicable).
- Crop rotation & history. It’s important to understand broadly how many corn, soybean, and other types of growers are available in the market, but also the specific rotation patterns.
- Acreage & farm size. Large farms have different needs than smaller farms, and vice versa. It’s also important to know how many acres of different types of crops a grower operates, because a more diverse operation will respond differently than a homogenous one.
- Owner vs. operator status. Just because someone is listed as a grower doesn’t mean they’re the primary decision-maker. It’s important to know not only who owns the land, but who’s operating it.
- Related growers. Whether they co-operate the land with a sibling or cousin, or they’re a child who operates the parent’s land, there’s a web of relationships that you barely scratch the surface with when you look at a single grower record.
The more detail you have on individual growers, the better you can aggregate that information to look at specific segments of the market. When you only have estimated or projected aggregates, or your grower lists stop at name, address, crop type, and acreage, you’re missing out on so many applications and insights.
Data from ten years ago tells you--well--what happened ten years ago. While historical agriculture data has value in determining the patterns of farming over time, it doesn’t give you the most up-to-date picture as to what’s happening on the farm right now.
In the midst of a tumultuous planting season, updated data is absolutely critical.
A great example of recent data is our in-season data that we offer at Farm Market iD. We build that data from a variety of sources:
- Historic and In-Season Satellite Imagery from Landsat and Sentinel Sensors
- Historic and In-Season Weather Data Observations
- Historic planting information derived from the NASS Cropland Data Layers
Imagery collected in-season is the primary driver of crop identification. Historic and in-season weather information is used to predict crop growth stage and to fine-tune identification. Data driven from the NASS datasets are used to create spectral profiles unique to crops of interests.
An example of in-season data within one of our applications.
Information from satellite imagery is used to distinguish crops early in the growing season. Imagery from previous seasons, pertinent weather data such as growing degree days, soil temperature, and precipitation are used to create the baseline model against which new satellite imagery is compared.
The reason we provide in-season data is to make sure that our customers get the most up-to-date insights possible.
The more current information you have, the better able you’ll be to ascertain what’s going on in the farmer’s head, and the better able you’ll be able to communicate with and, eventually, sell to them.