Every year as farmers we do something really pretty silly; we carry out work in our fields that costs us more money than we make from it.

This mind boggling fact was first pointed out to me by Sebastian Graff-Baker at Andersons a few months ago when I first discussed the idea of robotic farming as a service and it has been nagging away at me ever since. The problem is this; we acquire full knowledge too late. It is only when we have harvested our crop that we know how much we have yielded from each field or each hectare. It is only when we have spent all of the money that we know how much it has cost us to get the crop from seed in a bag to marketable product. It is then, and only then, that we can say with certainty what our break even yield was. With this knowledge, we know that certain parts of each field, most likely the headlands or a patch where the soil quality is particularly poor, have yielded less than our break-even yield and therefore all of the activities we have carried out on these parts of the field have been loss making activities. It is a rare year indeed that every hectare of the farm produces a profit.

The really crazy part of the farmer psyche is that (if we are honest with ourselves) we probably spend more money on those poor bits of the field - we up the seed rate, we might increase the fertiliser dosage.

It is understandable because we are eternal optimists as farmers and we think the whole field will get above that break-even point because it will yield well and the market prices will be favourable. Of course the alternative of spending less on the worst parts of the field, in other words allowing the crop to perform badly, or even in the most extreme scenario not planting the worst parts of the field at all is almost unthinkable to most farmers. Aside from anything else it will look untidy when the neighbour pokes their head over the hedge.

The other issue with the arable farmer mindset as it stands at the moment is that we measure ourselves and each other using the wrong metrics. The first bit of information that farmers try to acquire about each others’ farming enterprises is how much land the business farms, or how many tonnes the farm produces. But this is the wrong measurement. We should really measure each other by the number of tonnes we are able to produce at a profit.

The real reason that we don’t leave patches of the field unplanted and that we don’t currently measure ourselves on this basis is because we don’t have enough high quality, trustworthy data on which to base our decisions. So instead we base our decisions on hope; hope that prices will be good, the weather favourable and the yield at the top end of our budget. But, as I have said repeatedly when I speak to farmers, hope is not a strategy.

There will soon be a better way and Small Robot Company will be part of making it available.

Imagine a world where you had a far more granular understanding of the quality of every square metre of your soil. Robotics could enable this and Small Robot Company will have automated soil sampling available to farmers within three years as part of the service supplied by Tom, our monitoring robot. Tom has the patience to take a soil sample in exactly the same way repeatedly across your entire farm, and to measure every square metre (within a few short years he we will be able to be much more accurate even than that). A human being does not have this patience and so a “digital” soil map such as those made available to farmers today is necessarily subject to an amount of averaging. A soil map created by a robot will be much more accurate and consequently much more valuable; it will be much closer to achieving a truly digital view of the soil.

With this digital view established, the farmer will be able predict with a far greater degree of accuracy which parts of the field are likely to achieve budgeted yield and which are not. They then have a decision to make as to whether they do some pre-planting work to the worst parts of the field, for example to deal with drainage issues or adjust the soil pH. However, the important distinction is that Tom will have enabled this decision to be made on robust, accurate data rather than basing it largely on gut instinct.

Then comes the second wave of this new and revolutionary way of farming in the form of Harry, our digital planting robot. Harry will take this digital view of the soil provided by Tom and use it to vary the seed spacing and seed depth to give the highest possible chance of producing a profit for the farmer. Harry will also hold in his head all of the publicly available market information for the crop that is being produced and therefore be able to make a much more accurate assessment of what the likely break even yield is for this particular crop, on this particular field in this particular year. Harry will not simply repeat the same set of actions he took last year, he will learn and improve every year. If a part of the field has little or no chance of producing a profit (and farmers will be able to control the level of risk they are willing to take - more on this in future posts) then Harry will simply not plant it.

Ultimately, it will be possible for Harry to go back out into the parts of the field that have not been planted with the target crop and re-plant those patches with something else - a secondary commercial crop which is better suited to those soil conditions perhaps, or a non-commercial cover crop that provides a habitat for beneficial species. There is a huge range of options that could be feasible using a digital planter like Harry, although this “permaculture” approach is unlikely to become widely adopted until a fully automated variable harvester becomes available, something that is not currently within Small Robot Company’s initial three year vision.

Of course, there are no guarantees that the days of loss making activity will instantly be banished from farms forever. Our ability to predict the long term weather, the markets and the performance of a crop before it has been planted will never be perfect. But Tom and Harry will bring decision making on farm a good deal further on than it is today.

We are currently recruiting for our Farmer Advisory Group, a select number of farmers who will help us to shape our service and make our vision a reality.

If you are interested in the ideas discussed above and in being part of the next farming revolution, please get in touch with Co-Founder Sam Watson Jones (