Blog Post – SoilStack

Using design-based soil sampling to monitor soil health

Dan TerAvest on April 21, 2026

Soil sampling is a crucial part of sustainable land stewardship, helping producers understand what is happening on their land. The most common reason for soil sampling is to make effective nutrient management decisions. Increasing adoption of sustainable, regenerative, or soil health building practices over the past 10-15 years has created demand for a different type of soil sampling: soil health monitoring. Soil health monitoring helps support data-driven land management decisions and assess the effectiveness of management practices, such as reducing tillage or implementing diverse rotations or cover cropping.

Many soil testing labs now have specific soil health packages, measuring a combination of biological, chemical, and physical soil health indicators not included in traditional soil fertility test packages. (Check out these USDA-NRCS bulletins for in-field and lab-based soil health assessments.) At the same time, there has been a proliferation of national and regional programs promoting soil health and soil health testing including “Farmers for Soil Health,” American Farmland Trusts “Soil Health Stewards Program,” Pasa Sustainable Agriculture’s “Soil Health Benchmarking Study” and California’s “Healthy Soils Program,” just to name a few. While farmers may conduct soil health sampling to meet program requirements, the benefits go beyond a box that needs to be checked.

Soil testing and sampling strategies

Most producers using soil sampling to support nutrient management decisions will collect one composite sample, containing 5-15 cores from different areas of the field. Soil labs offering soil nutrition packages often provide guidelines for how to collect and manage quality soil samples. Farmers who have adopted precision agriculture frequently use grid sampling to collect one sample for every 1, 2.5, or 5 acres in order to generate precise nutrient maps.

Advances in GPS technology and specialized machinery have led many farmers to scale up farm soil sampling methods to develop high resolution maps to be used in precision farming. Precision Agriculture uses GPS and other technologies (sensors, drones, AI, etc.) to manage crops with high accuracy. Variable Rate Technology (VRT) targets inputs to where they have the most impact—for example, reducing planting density and fertilizer applications on sandy hilltops where the yield is expected to be low. Using precision agriculture practices profitably requires significant investment in both specialized machinery and digital prescription maps to make it a reality.

Grid vs. stratified soil sampling

If your goal is to accurately monitor soil health changes in response to management, neither single-field composite or grid sampling is a great option. Composite sampling may be good enough for small fields (< 25 acres), but soil conditions can vary widely over small areas. Small differences in where soil samples were collected across different years can affect the results and make tracking changes over time difficult. On the other hand, grid sampling requires many samples at significant expense, making it a less-than-ideal option.

A more suitable option for how to design a soil sampling plan, then, is design-based soil sampling.

Best soil sampling practices for farmers

Design-based soil sampling, aka stratification, divides fields into smaller, more uniform units (i.e., zones) based on known differences in soil type, topography, yield history, or previous management. This allows producers to collect fewer samples while maintaining confidence in the value of the soil data being generated.

Design-based soil sampling in practice

The figure below shows a 74-acre field. Using either 5- or 10-acre grid sizes would require 8 or 15 samples, respectively. In contrast, using publically available digital soil maps of organic matter, clay content and slope generated 4 zones with similar conditions. Now a producer can collect a composite sample from within each zone, sending only 4 samples to the lab, to get a realistic average for each zone.

Cost-effective soil sampling for farms

When described this way, design-based soil sampling sounds like a simple process, but it’s not something most farmers have the tools or experience to replicate on their own. Not only would a producer need to generate sampling zones, they also need navigation tools to accurately guide them in the field to confidently collect samples from within each zone.

Simple digital tools can enable farmers to stratify and sample their own fields at low cost. The sampling plan above was generated using SoilStack’s automated stratification algorithm. SoilStack is an open-source and flexible platform Our Sci designed to simplify soil sample design for agriculture and conservation purposes. SoilStack helps farmers create smart soil sampling plans by interpreting all available data about your plot of land—whether that’s 5 acres, 5,000 acres, or 50 different plots of 50 acres each. The SoilStack design-based soil sampling platform includes an intuitive front-end app that pinpoints exactly where in the field you need to collect a sample and uses GPS to guide you to each sampling location accurately. When you’re managing multiple fields or multiple farms, SoilStack’s administrative tools help keep track of your soil sampling projects.

How to use design-based soil sampling

Create your own account at SoilStack to get started with your own design-based soil sampling plan, or contact us to talk about precision soil sampling design for your farm or project. Soil health, agroecological land management, and conservation practices are among our team’s areas of expertise.

Taking the Next Steps, Can SoilStack Help Producers and Agronomists?

Dan TerAvest on February 28, 2025

10,000!

That’s how many “smart” soil sampling plans have been generated by SoilStack since it was launched in 2021. We’ve reached that number because SoilStack makes it really easy to apply a template for a design-based soil sample planning to tens or hundreds of fields at a time.

What is design based soil sampling?

Collecting soil samples from predetermined locations, using known sources of variability, to best represent the field. 

SoilStack uses digital soil maps (clay and organic matter content), digital elevation maps (slope) and satellite-derived vegetative indices (NDVI) to understand in-field variability. Then it uses sampling algorithms to optimize the locations for collecting samples based on that variability.

Where are we now?

When we first launched SoilStack, we wanted to address key bottlenecks in the soil sampling process:

Reduce costs by leveraging publicly available data to reduce the total number of samples needed for each field while providing high quality data.
Enable anyone to generate “smart” sampling locations by automating the processing of publicly available data and applying scientifically rigorous approaches to generate sampling locations.
Simplify the in-field sample collection process with tools that guide soil samplers to each predetermined sampling location and capture critical sampling metadata.

In its current state, SoilStack allows a user to draw a field boundary, generate unique zones based on variability of key properties (see box above) and collect samples at locations within each zone optimized to represent the variability of field conditions. Users can analyze each soil sample separately, or can aggregate samples from each zone into composite samples to reduce lab costs.

Users pay a one time fee ($10) per “smart” sampling plan created. Once created, the user has access to all of the detailed metadata about the sample plan design and that plan persists for as long as they want to use it. That means they can use SoilStack’s front-end app to navigate to those sampling locations, collect samples at those points and then access the sampling metadata as often as they wish in the future with no additional costs.

Where do we go from here?

SoilStack is developed and maintained by Our Sci LLC, whose goal is to support community research, turning citizens into researchers, researchers into community leaders and communities into solvers.

What should we, as a small, bootstrapped, open-source company do with SoilStack to continue developing it as a tool for shared community knowledge?

The simple answer is as much as we can. The more complicated answer is that we need to determine how we can leverage the existing technology stack to provide value to a (hopefully) growing user base that includes producers and agronomists.

What else is out there?

We decided to do research into some existing platforms that support soil sampling and improved agronomy support services.

There are a lot of platforms out there! They tout their support for farmers, ranchers, agronomists, consultants and supply chain actors to reduce costs (“reduce fertilizer use by 20%”), increase yields (“up to 15%”) and improve environmental stewardship (reduce water and fungicide use). Their tools benefit big groups with features like:

Proprietary models
AI insights
Advanced Sensors
Customizable views
Satellite imagery and remote sensing technology
Real time crop monitoring
Big data analytics

These platforms offer their customers real value, but the tools listed above only work if you feed in a lot of data. Which means you have to buy and install sensors, pay for soil mapping using NIR technologies or digital soil cores, or buy remote sensing products.

Another thing we noticed when reviewing the platforms’ websites is that most of them do not list prices for their various services. Granted, digital agronomy support isn’t the same as paying for soil analysis at a lab (analyte x = $y, or $z if part of a package). But, companies with affordable prices tend to mention that, a lot.

Is the digital agronomy space becoming a zero sum game? Either you go all in on a digital agronomy platform, and buy all of the bells and whistles to get the AI insights and pretty maps, or you’re on your own?

What do we want the future to look like?

We’d like to invest in a future where producers are not priced out of advanced analytics, but rather have more options to engage at levels that meet their resources and goals?

We want to hear from you!

If an open source platform like SoilStack can provide a la carte options, what features do agronomists and producers need to derive value from more targeted use of soil sample data?

Below are some of our thoughts, what do you think? What features would provide value to you or your clients?

Easily visualize the sampling zones and the data layers used to generate sample plans?

Import sample results to SoilStack for each sampling point, which would enable generating heat maps of soil properties?

Import sampling locations from external services so soil samplers can use the front-end app for in-field sample support?

Integrations with other platforms? For example, pull in field boundaries from farm management software and push sampling locations/metadata into that same software?

Connect with us at info@soilstack.io