The AgriNova platform

The AgriNova platform, step by step

How the planned product would work, shown on three synthetic fields: choose a field, inspect the evidence behind its score, see what happens when that evidence is weaker and plan a field check.

Guided website example · Available on this website
Connected field monitoring · Planned · target Dec 2026
AI Farmer's Assistant · Planned
AMI scores · Research

How to read this page

Illustrative demo with synthetic data: three fields, not a real farm.
Fixed example responses. No live farm connection or AI analysis.
The field map is a simplified layout. Not a real farm or satellite image.
Scripted example status: the words Review, Watch and Stable are part of the example. Cause not established in this example: a score change alone does not show why a field changed.
Changes are relative to the previous observation in this demonstration. These synthetic scores are not agricultural assessments or validated performance results.
The assistant conversation uses fixed example answers.

01Interactive demo

Select a field, inspect its evidence and plan a field check

Choose one of three synthetic fields, open the evidence behind its score, see how a score is built, try what happens when an image is cloudy or too old, then plan a field check and print it.

1Select a field

Field overviewMay 22 – June 12

Sample · synthetic data

Status in this example

ReviewNorth Field
StableEast Field
WatchWest Field

Areas marked on North Field

North-east corner (lower score)
South half (typical)

North Field · WheatReview

Agricultural Management IndexDown 11 points

Sample · synthetic data

TrendAMI · 0–100

Area 160.0 hectares

Why this field?

North Field has the largest score change of the three fields.

Context

The example does not establish a cause. Review the observation quality and field context before drawing a conclusion.

Next check

Compare recent field observations and decide whether a focused field check is warranted.

2Evidence for this field

Observation behind the shown score

Observed June 12 · processed June 13 · cloud 5% · usable area 95%

Fresh · 2 days old

Usable observations: June 5 and June 12
Quality note: Both images are usable: cloud covered 12% of the field on June 5 and 5% on June 12.

Evidence records: synthetic — for demonstration.

3How the score is built

A generic example field, not one of the three fields above: four invented inputs, each from 0 to 100 where higher is more favourable, combined with equal teaching weights.

How an example score is built from four invented inputs
Input family	Earlier week	Later week
Crop vigour	75	58
Soil-moisture condition	75	70
Temperature condition	75	58
Water-stress condition	75	70
Score with equal teaching weights	75	64

Teaching weights — not AgriNova's formula. A drop of 11 points is not an 11% yield loss or a diagnosis.

See the teaching explorer for how inputs combine

4What if the evidence were weaker?

As observedWhat if this field's June 12 image had been cloud-covered?What if this field's newest usable image were 18 days old?

This what-if changes only this example.

A what-if applies to the selected field and stays on when you choose another field.

Ask about the example

A question is a useful next step.

Choose a question to read its fixed explanation and the evidence it used.

5Plan a field check

Choose the area that looks different and a comparison zone that looks typical. The plan lists both, with questions to look into and what to record in each.

Affected zone

North-east corner (lower score)South half (typical)

Comparison zone

North-east corner (lower score)South half (typical)

6Record what the field check found on North Field

Choose one once you have looked. A difference whose cause stays unclear is a complete record.

No visible differenceVisible difference — cause unclearVisible difference — likely cause noted

Note (optional)

Your note stays on this page: it is not saved or sent, and reloading the page clears it.

7Take the plan with you

The plan below is what prints: first what the records show, then the questions and notes for the visit.

Sample · synthetic data

Scouting plan · North Field

Prepared with AgriNova AI's guided example — a plan for a field check, not a diagnosis or recommendation

What the records show

Field: North Field · Wheat · 160.0 hectares
Image observed: June 12
Image processed: June 13
Cloud and usable area: Cloud 5% · usable area 95%
Source status: Fresh · 2 days old Review
Score: 64 (June 12)
Affected zone: Not chosen: write it in at the field.
Comparison zone: Not chosen: write it in at the field.

Above this line: what the records show. Below it: questions and notes for the field check — interpretation, not a diagnosis.

Questions to look into (not a diagnosis)

Uneven emergence?
Possible water stress?
Possible waterlogging?
Possible nutrient issue?
Possible pest or disease?
Unknown?

What to record in each zone

Affected zone and comparison zone
Item	Affected zone	Comparison zone
Location
Crop stage
What you observe
Photos
Isolated or widespread

What the field check found

No visible difference
Visible difference — cause unclear
Visible difference — likely cause noted

Note

Date of visit

Checked by

SAMPLE — synthetic data · a plan for a field check, not a diagnosis or recommendation

The interactive example needs JavaScript. The field details and their evidence remain available below; the fixed answers are in the assistant section further down.

North Field · Wheat

Area 160 hectares. AMI 64. Down 11 points. Review.

North Field has the largest score change of the three fields. The example does not establish a cause. Review the observation quality and field context before drawing a conclusion. Compare recent field observations and decide whether a focused field check is warranted.

Trend: May 22: 78; May 29: 77; June 5: 75; June 12: 64.

Evidence: Observed June 12 · processed June 13 · cloud 5% · usable area 95%. Usable observations: June 5 and June 12. Evidence records: synthetic — for demonstration.

East Field · Barley

Area 80 hectares. AMI 82. No change. Stable.

East Field's latest score is unchanged from the previous score. An unchanged score does not prove that every part of a field is free of problems. Continue normal field observation and compare the next usable information.

Trend: May 22: 80; May 29: 81; June 5: 82; June 12: 82.

Evidence: Observed June 12 · processed June 13 · cloud 3% · usable area 97%. Usable observations: June 5 and June 12. Evidence records: synthetic — for demonstration.

West Field · Canola

Area 120 hectares. AMI 71. Down 4 points. Watch.

West Field's latest score is four points below the previous score. A score change alone does not identify crop stress or distinguish it from other explanations. Review the supporting observation and note whether a field check would add useful context.

Trend: May 22: 77; May 29: 75; June 5: 75; June 12: 71.

Evidence: Observed June 12 · processed June 13 · cloud 8% · usable area 92%. Usable observations: June 5 and June 12. Evidence records: synthetic — for demonstration.

Prepare a field visit with the scouting checklist

02Scores

Score first, layers one click away

The three fields ranked by how far their score changed. Each row opens the evidence behind its score.

Sample · synthetic data

North Field · Wheat
64 AMI · Down 11 points
Review
Show what's behind this score
Evidence records for North Field
Observed Processed Cloud Usable area
May 22 May 23 4% 96%
May 29 May 30 9% 91%
June 5 June 6 12% 86%
June 12 June 13 5% 95%
Provenance: synthetic — for demonstration.
Input families the score would combine
- Crop vigour
- Soil moisture
- Temperature stress
- Water stress
West Field · Canola
71 AMI · Down 4 points
Watch
Show what's behind this score
Evidence records for West Field
Observed Processed Cloud Usable area
May 22 May 23 7% 93%
May 29 May 31 15% 85%
June 5 June 6 11% 88%
June 12 June 13 8% 92%
Provenance: synthetic — for demonstration.
Input families the score would combine
- Crop vigour
- Soil moisture
- Temperature stress
- Water stress
East Field · Barley
82 AMI · No change
Stable
Show what's behind this score
Evidence records for East Field
Observed Processed Cloud Usable area
May 22 May 23 2% 98%
May 29 May 30 6% 94%
June 5 June 6 10% 90%
June 12 June 13 3% 97%
Provenance: synthetic — for demonstration.
Input families the score would combine
- Crop vigour
- Soil moisture
- Temperature stress
- Water stress

Evidence records for North Field
Observed	Processed	Cloud	Usable area
May 22	May 23	4%	96%
May 29	May 30	9%	91%
June 5	June 6	12%	86%
June 12	June 13	5%	95%

Evidence records for West Field
Observed	Processed	Cloud	Usable area
May 22	May 23	7%	93%
May 29	May 31	15%	85%
June 5	June 6	11%	88%
June 12	June 13	8%	92%

Evidence records for East Field
Observed	Processed	Cloud	Usable area
May 22	May 23	2%	98%
May 29	May 30	6%	94%
June 5	June 6	10%	90%
June 12	June 13	3%	97%

This sample does not include input values for these fields.

See the teaching explorer for how inputs combine

Sample · synthetic data

Agricultural Management Index

North Field: 64

One number, then the evidence behind it.

03How AMI works

How the field score would work

AMI is AgriNova's proposed 0–100 index for organizing field information into an understandable overview. It is intended to help prioritize review alongside the supporting observations, rather than replace them.

A useful score needs context: the crop, the observation date, the quality of the inputs and what has changed since the previous observation. Crop calibration and field validation remain essential development work.

The four input families

The score would combine four kinds of field information. How they would be weighted and calibrated for Canadian crops is still being researched.

Crop vigour
Soil moisture
Temperature stress
Water stress

Read the field-score guide See the teaching explorer for how inputs combine

04Assistant

The planned AI Farmer's Assistant

Planned

The planned assistant is designed to explain field information in plain language. Each fixed answer below is shown word for word, beside the evidence it used. Growers and advisers keep responsibility for field decisions.

These are fixed example answers written for this website. They are not produced by an AI model and do not analyse a real farm.

Sample · synthetic data

Where should I scout first?
All fields
In this sample, North Field is the first field to review because it has the largest score change: down 11 points to 64. That change does not identify a cause or prove crop stress. Check the underlying imagery date and quality, compare recent field observations and decide with your adviser whether a field visit is warranted.
Evidence used
Score change
North Field down 11 points, West Field down 4 points, East Field no change
Observation date
June 12, for all three fields
Quality note
Every June 12 image usable: cloud 5% (North Field), 8% (West Field), 3% (East Field)
Why did North Field change?
This field: North Field
This illustrative dataset does not include enough evidence to establish a cause. A useful review would examine imagery quality, crop stage, recent weather and field observations. The score alone cannot distinguish crop stress from normal development or a data-quality issue. Confirm the conditions before choosing an intervention.
Evidence used
Score change
North Field: 75 on June 5, 64 on June 12 (down 11 points)
Observation date
June 12, processed June 13
Quality note
Cloud 5%, usable area 95%
What should I record during a field check?
All fields
Record the date, location, crop stage and what you observe. Add photos and note whether the issue is isolated or widespread. Compare affected and apparently unaffected areas, then share the observations with your agronomist. This example does not recommend an irrigation amount, fertilizer rate or pesticide treatment.
Evidence used
None from the fields: this answer relies on no field data.

05Data and imagery

Where the evidence would come from

The planned product would combine satellite imagery with weather and field records. Each would carry its date and quality, because a score can only be as current as its newest usable image.

Satellite imagery

Sentinel-2 multispectral imagery is the intended image source. The observation date, cloud cover and image quality matter to how a score is read.

Weather

Recent conditions shown beside the imagery, never treated as proof of a cause.

Optional soil data

Soil information where a grower has it and chooses to add it.

Your field records

Field boundaries, crops and the notes from each field check.

A satellite pass is not always a usable image

A satellite pass over a field is an acquisition. It becomes a usable image only when cloud, haze and shadow leave enough of the field visible, so the newest usable image can be older than the newest pass.

In-field sensors and machinery data measure things a satellite does not. A satellite view would sit beside them, not replace them.

How the demo treats image quality

Fresh: The newest usable image is 5 days old or less: the score and its status are shown.
Ageing: It is 6 to 14 days old: the score and its status are shown, with the image date.
Stale: It is more than 14 days old: the last valid score is shown with its age, and no status.
Cloud-affected: The newest image is covered by cloud: the previous valid score is shown with its date, and no status.
No usable image: None of the recent passes can be used: the most recent valid score is shown with its date, and no status.

06Setting up

What setting up would involve (planned)

Nothing is set up on this website. In the planned product, starting would take three steps.

Planned
1
Field boundaries
Mark where each field is.
Records that help: a farm map, or the boundary files your equipment or adviser already use.
Planned
2
Crops
Say what is growing in each field this season.
Records that help: this season's crop plan.
Planned
3
Season
Set the dates of the season, so each score is compared within one season.
Records that help: last season's field notes.

07Product status

What exists today, and what is planned

This website demonstrates the intended product experience. A future demonstration or trial should define exactly which capabilities, data sources and support are available.

How to read the statuses

Available on this websiteyou can use it here now.
Plannedintended, not yet available. A date appears only where one has been set, and it is a target.
Researcha method still being evaluated.
Target · not yet achieveda dated goal.

Product status: what you can use on this website now, and what is planned
Capability	Status	Detail
Guided website example	Available on this website	Three synthetic fields, illustrative scores, area conversion and fixed explanations.
Value calculator	Available on this website	Works out the most a field-monitoring service could cost per acre per year for your own farm numbers to pay. Runs in your browser.
Scouting checklist	Available on this website	Prepares a printable field-visit record. Not a diagnosis or recommendation. Runs in your browser.
Pilot planner	Available on this website	Prepares a printable plan for judging a pilot: the question, the window, a baseline, the records to keep and up to three measures. Reserves nothing. Runs in your browser.
AMI teaching explorer	Available on this website	Shows how a 0–100 field score could combine four invented inputs with teaching weights, not AgriNova's formula. Runs in your browser.
Connected field monitoring and AMI scoring on real fields	Planned · target Dec 2026	In development. Live field data, update schedules and supported crops require confirmation for each pilot.
Prairie beta on 15 farms (Canadian validation)	Planned · 2027–28	Initial focus on Prairie wheat, canola and barley. Timing and participation remain subject to confirmation.
Commercial launch in Canada	Target · not yet achieved · Q1 2029	A dated goal that may change.
Subscriptions	Planned · target Q1 2029	Planned for the commercial launch in Canada. Public pricing and features are not yet confirmed.
Weekly satellite imagery	Planned · at launch	The schedule is not yet confirmed, and a usable image depends on cloud cover.
AMI scores and AI explanations	Research	Proposed workflow. Scoring, calibration and assistant behaviour require evaluation against field evidence.
Canadian crop calibration	Research	Models must be retuned for Canadian wheat, canola and barley before they are production-ready.
AI Farmer's Assistant	Planned	The answers on this website are fixed examples, not produced by an AI model.
More frequent imagery	Planned	How often, and from when, is not yet confirmed.
Intended data sources	Planned	Satellite imagery, with Sentinel-2 as the intended source, together with weather data and optional soil data.
Radar (SAR) imagery to fill cloudy gaps	Planned	Intended to fill gaps when cloud hides a field; not yet scheduled.
Offline caching and text-message (SMS) alerts	Planned	Intended for fields with poor mobile signal; not yet scheduled.
Field boundary import	Planned	Drawing or importing field boundaries is a planned product step. This website accepts no uploads.
Mobile app	Planned	An app for phones and tablets.
Alerts and to-do list	Planned	A notice when a field's score drops, and a list of field checks to do.
Shared adviser view	Planned	Your agronomist or adviser would see the same view of your fields.
Notes, permissions and handoff for advisers	Planned	Notes kept against a field and a date, which client fields an adviser could see, and passing a field check on. How access would be granted is not yet confirmed.
Learning from recorded outcomes	Planned	What you record after a field check would inform later scores.
English and French	Planned	Part of the planned Canadian roll-out.
Multi-farm view for organizations	Planned	One view across the farms a cooperative, dealer or agribusiness supports. Which farms, and on what terms, is not yet confirmed.
Organization integrations (single sign-on and API access)	Planned	APIs, integrations and enterprise arrangements are opportunities to discuss, not current service commitments.

Crops and provinces

The Prairie beta is planned for wheat, canola and barley in Saskatchewan, Alberta and Manitoba. Every combination is still at the research stage.

Coverage by crop and planned beta province
Crop	Saskatchewan	Alberta	Manitoba
Wheat	Research — Canadian calibration planned for the 2027–28 beta	Research — Canadian calibration planned for the 2027–28 beta	Research — Canadian calibration planned for the 2027–28 beta
Canola	Research — Canadian calibration planned for the 2027–28 beta	Research — Canadian calibration planned for the 2027–28 beta	Research — Canadian calibration planned for the 2027–28 beta
Barley	Research — Canadian calibration planned for the 2027–28 beta	Research — Canadian calibration planned for the 2027–28 beta	Research — Canadian calibration planned for the 2027–28 beta

08Before you begin

Questions about the platform

Is AgriNova a fully launched service?

No. AgriNova AI is in development. Which capabilities a demonstration or a future pilot would include will be confirmed when enquiries open.

Do I need to install sensors?

The proposed platform would not require on-farm sensors. Optional field data may add context where compatible sources exist.

What does an AMI score tell me?

It is intended as a starting point for reviewing field information. Interpret it with the supporting data, crop context and limitations; it is not a diagnosis.

Is the website demo using a real farm?

No. The fields, scores and responses are synthetic examples. The demo does not connect to your farm or perform live AI analysis.

Can the assistant prescribe field treatments?

No. The intended role is decision support. Field conditions and proposed interventions need appropriate confirmation, including agronomist input where relevant.

How often will field information update?

A confirmed update schedule is not yet published. It would depend on the data source and on cloud; observation dates and gaps would stay visible.

Next step

Continue with the planning tools, the case studies or the planned Canadian pilot.

Explore the planning tools
Planning tools that run in your browser: a value calculator, a scouting checklist, a pilot planner and an AMI teaching explorer.
Explore the case studies
Historical field studies, and the standard a Canadian result would have to meet before it is reported.
Learn about Canadian trials
The planned Prairie beta: who it is for and how it would work.

The AgriNova platform, step by step

Select a field, inspect its evidence and plan a field check

A question is a useful next step.

Score first, layers one click away

North Field · Wheat

West Field · Canola

East Field · Barley

How the field score would work

The four input families

The planned AI Farmer's Assistant

Where should I scout first?

Why did North Field change?

What should I record during a field check?

Where the evidence would come from

Satellite imagery

Weather

Optional soil data

Your field records

A satellite pass is not always a usable image

How the demo treats image quality

What setting up would involve (planned)

Field boundaries

Crops

Season

What exists today, and what is planned

Crops and provinces

Questions about the platform

Is AgriNova a fully launched service?

Do I need to install sensors?

What does an AMI score tell me?

Is the website demo using a real farm?

Can the assistant prescribe field treatments?

How often will field information update?

Explore the planning tools

Explore the case studies

Learn about Canadian trials