For farmers who want to graze better.

A feed budget is a month-by-month plan that shows whether your farm has enough grass and forage to meet the needs of your animals throughout the year. It balances two things:

• Animal demand — how much your livestock need to eat, broken down by group (e.g. cows, growing cattle, ewes)
• Feed supply — what’s available, both from grazing and from housed feeding (hay, silage, concentrates)

A good feed budget helps you plan ahead, spot problems early, and make confident decisions about stocking rates, supplements, and grass management before you run into trouble.

Feed budgets in ruumi

Go to Planner and create your first feed budget. You can name feed budgets and choose when a feed budget should start or end. We recommend a budget for every year - or every breeding cycle.

One budget can be set as active by clicking the o in the dropdown on the left of the name. This is the one that drives your stocking rate calculations and grazing days ahead shown on your dashboard.

Stock classes

When you set up a feed budget, you assign animals to stock classes (e.g. suckler cows, weaned calves, finishing cattle, ewes). ruumi includes presets for common stock types with typical daily intake figures built in, so you don’t have to start from scratch.

You can adjust these to match your own animals. For stock that are nursing — like cows with calves at foot, or ewes with lambs — the mother’s daily demand is increased to account for milk production rather than tracking the youngstock separately.

If you enter a mating date, ruumi can automatically project feed demand changes across the year as animals move through pregnancy and weaning. After saving the boundary, ruumi will prompt you to fill in some details that help it generate accurate grazing plans.

Housed vs grazing periods

For each month, you can tell ruumi whether your animals are grazing or housed. This matters because:

• Grazing months — feed comes partly or fully from grass, and ruumi calculates the area of land needed to meet demand
• Housed months — animals still need feeding, but land area isn’t required; demand is met by hay, silage, or other supplements

You choose the supplement types from a dropdown (hay, silage, concentrates, forage crops), and ruumi tracks planned vs actual usage.

Understanding your budget: surpluses and deficits

Once your plan is set up, ruumi shows you a graph of animal demand versus grass growth across the year. Where these lines cross, you either have a surplus or a deficit.

If you have a deficit (not enough feed)

You have a few options:

• Reduce Average Farm Cover (AFC) — graze the farm harder to carry more animals through the tight period
• Feed hay or silage in the field — supplement grazing animals to bridge the gap
• Reduce animal numbers — sell or move stock before the pressure hits
• increase grazing area

If you have a surplus (more feed than you need)

• Increase Average Farm Cover (AFC) — let covers build up as a buffer
• Cut silage or hay — take surplus grass off as conserved forage to use later (i.e. reduce grazing area)

Scenario planning

Because you can create multiple budgets, it’s easy to run “what if” comparisons — for example:

• What if we have a dry summer?
• What if we keep the calves on rather than selling them?
• What if we increase stocking rate by 10%?

Duplicate an existing plan, make your changes, and compare the outcomes side by side. Switch your active plan at any time.

Every field in ruumi shows Total Cover (kg DM/ha) and a colour-coded map of how that grass is spread across the field. This article explains where those numbers come from, what data we lean on, and what you need to keep up to date so the estimates stay useful.

The two things we estimate

1. The field-level number: current average cover in kg DM/ha. One number per field, refreshed every day. This is what drives the grazing wedge, grazing days-ahead and growth rate.

2. The within-field map: the colour pattern. A picture of how the cover varies across the field. High covers in dark green, lower covers in lighter shades. The map shows you where you might want to put fences, where grass is growing well and where it isn’t.

How we build the field-level number

Think of the field-level number as a curve through the season: growing day by day, dropping sharply when you graze or cut.

We build it like this:

• Starting point. We anchor the curve to the most recent grass cover reading you’ve given us before the start of the grazing season. If you haven’t given us one yet, we start the season at a sensible default (1500 kg DM/ha) and let your first reading correct it.
• Daily growth. Between events, the curve grows each day using a grass growth model. Growth typically is highest in May and June and tapers off through autumn. We don’t allow grass to grow above 6000 kg DM/ha.
• Grazing and silage events drop the curve. When you log a grazing event, the curve drops to your post-grazing residual on that date. A silage cut drops it to a low residual (around 800 kg DM/ha by default).
• Your DM readings reshape the curve. Whenever you record a reading, we re-adjust to make sure the curve passes through that exact value.
• Satellite imagery proposes events. We watch each field with Sentinel-1 radar (sees through clouds, picks up biomass changes) and Sentinel-2 optical imagery. When we spot a sustained drop in radar backscatter and a fall in greenness around the same time, we propose a grazing event for you to accept or reject in the Rotation Timeline.

What you should keep up-to-date

The model can only be as good as what you give it. These are the things that matter most:

• Field boundary. Draw the field accurately. If the boundary is wrong, we’re measuring the wrong patch of the earth — pulling in tracks, hedges or a neighbour’s crop.
• Field type. Make sure grass fields are marked as grass (not arable). The model treats them differently.
• DM readings. A reading at the start of the season anchors the whole curve for that field. Readings during the season pull the curve back to the truth — handy after a stretch of weather we couldn’t fit, or when growth has been faster or slower than the seasonal average.
• Grazing and silage events. Log them with accurate dates and a sensible post-grazing residual. The post-graze cover is what the curve drops to — if you put 1500 when you actually grazed down to 1200, the curve will sit ~300 kg too high until your next reading.
• Animal group and post-grazing target. Set these in your group settings so we use your number rather than a default. This also helps detection know which fields are due for grazing next.
• Accept or reject proposed events. When we propose a grazing or silage event from satellite data, please confirm or reject it. A missed real event leaves the curve climbing; a confirmed false event drops it for no reason.

Why we sometimes get it wrong

We try to be honest about this. Common reasons the field number drifts away from what you’d measure with a plate meter:

• No recent reading. The longer it’s been since you gave us a number, the more weight the seasonal growth model is carrying — and weather rarely matches the average.
• A wrong post-graze cover entered on a past event. Every event after that one builds on the wrong starting point.
• A grazing or silage event that was never logged. The curve carries on growing as if the field is still standing.
• A proposed event that’s still sitting unconfirmed. Until you accept it, we can’t be fully sure it happened, and the chain reflects that uncertainty.
• A boundary that was redrawn recently. We rebuild the satellite history for the new shape, which can take a couple of days to settle.
• Long cloudy spells. The within-field map will look stale, and we have less satellite signal to detect events with — we lean harder on the growth model and your inputs in those weeks.

If a field looks wrong to you, the fastest fix is almost always to add a fresh DM reading. The curve will rebuild around it within a day.

If you spot something that doesn’t add up — a number that’s too high, a map that doesn’t reflect what you’re seeing in the field — please tell us. Honest feedback from your fields is what keeps the model improving.