Agronomy Consulting Software: Why Your Recommendations and Your Invoices Never Match

Why April breaks an agronomy firm

It is the second week of April. Your six agronomists cover 78 growers and something like 140,000 contracted acres. One is sitting in a truck at the end of a field entering a scouting note into a phone, and the field she is standing in has three different boundary versions: the one in the grower's John Deere Operations Center account, the one in your mapping tool from last season before the grower picked up 240 rented acres, and the one on the FSA map the crop insurance agent uses. She writes the recommendation as a note, texts the grower, emails the prescription to the applicator, and at the end of the month the office bills 140,000 acres of scouting because that is what the contract says, without any way to show which acres were actually walked.

The stack underneath this is usually a mapping tool, a shared drive of soil lab PDFs, two or three manufacturer portals the growers use, a spreadsheet of contracted acres by grower and service, and QuickBooks. Agworld, Agrian and Conservis all exist and all solve real parts of this. What none of them holds is the join that a consulting business actually runs on: a field boundary that is yours rather than borrowed, a recommendation that is a signed document with a professional behind it, the prescription file that came from that recommendation, the application that was actually made, and the invoice line that all of it produces.

The exposure is specific to consulting rather than farming. You carry agronomic liability for advice on land you do not own, given by professionals whose credentials matter, at a per acre price that has to be defensible when a grower questions it in November. A recommendation you cannot reproduce is a legal problem. An acre you cannot evidence is a revenue problem. Firms in this business tend to have both, and they discover it at the same time.

Problem 1: field boundaries are your master record and you are renting them

Everything in agronomy hangs off a field boundary: acres billed, rates applied, zones sampled, prescriptions written, records retained. If that boundary lives in the grower's Operations Center or FieldView account, then your business records depend on an account you do not control, held by a grower who may switch platforms or switch consultants. Every season your team redraws or reimports boundaries because a farm picked up rented ground, split a pivot, or dropped a landlord.

A build must treat boundaries as versioned, dated records that you own, with a clear lineage when a field is split or merged, and with acreage computed rather than typed. Versioning is the part that seems fussy and turns out to be essential, because a 2024 recommendation must be reproducible against the 2024 boundary even after the field changed shape in 2025. Synchronisation with grower platforms then becomes a two way exchange you control rather than a dependency: pull boundaries when a grower has them, push yours when they do not, and keep your version as the record of what you were paid to work on.

Problem 2: a recommendation is a liability document that most firms store as a note

The scouting note says aphids at threshold in the north half, recommend product at a stated rate. Six weeks later there is a yield problem and the conversation turns to what was advised, when, at what rate, and whether it was followed. Your evidence is a text message and a note in a mapping app that has been edited since.

What a build must include is a recommendation as an immutable, dated, professionally attributed document: fields and acres it covers, product and rate, the observation that justified it, the agronomist and their certification, and where required the grower's acceptance captured as a signature. Once written, it is not editable. Amendments become new versions with a link to the original. This is the single feature that changes the character of the business, because it converts advice from a conversation into a record, and it is exactly what a professional liability insurer wants to see when you renew. It also happens to be the thing your billing hangs off, which is the next problem.

Problem 3: prescriptions have to land in the machine, not in an inbox

A variable rate prescription is only worth anything if it loads on the controller in the cab. That means ISOXML for equipment following ISO 11783, shapefiles for a lot of legacy controllers, and platform specific handoff into John Deere Operations Center or Climate FieldView. It also means dealing with the reality that different controllers interpret rate units and zone boundaries differently, and that an applicator who cannot load your file at 6am will spread a flat rate and tell nobody.

Agrian has genuine strength around product labels and compliance, and Agworld handles scouting and recommendation workflow well. The gap for a consulting firm is that prescription delivery is where your advice either happens or does not, and it usually happens through whichever platform the grower or applicator prefers, which is not the one you standardised on. The build has to export in several formats, deliver by API to the platforms that support it, and record delivery: which file went to which operator at what time, and whether an as applied record came back. That returned as applied file closes the loop and is what lets you tell a grower in November what was actually done rather than what was advised.

Problem 4: per acre billing is not hourly billing and it never reconciles

Consulting revenue is a per acre per service model with a service catalogue: scouting at one rate, soil sampling on a grid at another, prescription writing per acre, nutrient management planning per plan. Then the exceptions arrive. A grower dropped 300 acres after the contract was signed. A field was prevented planting. Two brothers farm together but bill separately and share a field. A landlord pays half the sampling. A field was scouted four times and the contract covers three.

This is where firms lose real money, and it is the part general farm software does not attempt because it was built for growers rather than for service providers. Conservis is a capable operations and financial product on the grower side and was never meant to bill a consulting book. The build ties each billable event to the record that evidences it: a scouting visit with a timestamp and a location, a recommendation with an acreage, a sampling job with point counts. Contracted acres versus serviced acres becomes a live variance rather than a November argument. Invoices then reference the work, so when a grower asks what he is paying for, the answer is a list of visits and documents rather than a number.

Problem 5: compliance records are your obligation and your defence

In the United States, certified applicators must retain restricted use pesticide application records, with a federal minimum retention of two years and states frequently requiring more. Worker Protection Standard restricted entry intervals and pre harvest intervals have to be respected and, if you recommended the product, evidenced. Nutrient management plans carry their own state and programme specific rules. Confirm the exact requirements for the states you operate in with your state lead agency or an agronomic compliance specialist, because they genuinely differ.

A build that captures the recommendation properly gets most of this for free, provided the product record is real: EPA registration number, label rate range, and the interval constraints attached to the product rather than typed into a note. Then the system can refuse to produce a recommendation above label rate, warn when a pre harvest interval conflicts with the grower's expected harvest window, and produce a records pack per grower per season in an afternoon. That pack is also a retention tool, because a grower who gets clean compliance records from you every year has a reason to stay that a competitor's price cut does not answer.

What this costs and how long it takes

Across the 2,000 plus projects Digital Heroes has delivered, this is the honest shape for agronomy service providers. A first release, meaning owned versioned field boundaries with grower and farm hierarchy, mobile scouting with offline capture, immutable recommendations with signature, and per acre billing tied to evidenced work, runs $50,000 to $110,000 and ships in 10 to 14 weeks. A full platform adding prescription generation and multi format export, Operations Center and FieldView integration, laboratory result ingestion with normalisation, compliance record packs, a grower portal and accounting sync runs $130,000 to $320,000 over 6 to 12 months.

What drives the price up specifically here: the number of grower platforms you must exchange data with, since each API is real work and their models of a field differ. Laboratory ingestion, because every lab exports a different layout and normalising nutrient names, units and extraction methods across four labs is genuinely fiddly. Prescription export, since controller compatibility has to be tested against actual equipment rather than assumed. Multi state compliance. And a grower portal, which is a second product with its own support burden and should only be built when growers have asked for it twice.

What keeps it down: start with boundaries, scouting, recommendations and billing. That is the business. Prescriptions and lab ingestion are worth doing next season once the record is solid.

Build versus buy, and when buying is the right call

Buy if you are a small firm under roughly 10,000 consulted acres with one or two agronomists. Agworld will handle scouting and recommendations, Agrian covers product and label compliance well, and a spreadsheet handles your billing at that volume. Paying a per acre subscription is far cheaper than a build and its maintenance, and we would tell you so rather than quote.

Build when two or more of these are true. You consult on more than about 40,000 acres and your agronomists spend the first three weeks of spring fixing boundaries. Your per acre billing cannot be reconciled to evidence of work performed. You operate a service catalogue with more than about six billable service types and pricing that varies by grower. You have been asked to reproduce a recommendation from two seasons ago and it took days. Or you have a genuine operational advantage, such as a proprietary sampling protocol or a zone modelling method, that a generic platform flattens into its own workflow and quietly commoditises.

How to choose a developer for agronomy software

Ask them to whiteboard the field model before you sign anything. The right answer has grower, farm, field and boundary version as separate records, with acreage computed from geometry, split and merge lineage, and a season dimension. A developer who models a field as a row with an acres column has never dealt with rented ground changing hands in March and will rebuild the schema in month four.

Ask what they have shipped that generates ISOXML or shapefile prescriptions and whether it was tested on a real controller. This is the difference between a file that loads in the cab and a file that produces a shrug at 6am.

Ask how the scouting app behaves with no coverage for four hours, because that is a normal day in the field, and how it handles a photograph taken in a field with a location that must stay attached to the observation.

Ask who owns the code and the data, and get it in writing before kickoff. You should own the repository, the cloud accounts, the boundary data and the right to hire anyone else to continue the work. At Digital Heroes the code is yours from the first commit. Since the whole point of the build is to stop renting your own field records from a platform, accepting a new dependency from your developer would defeat it.