DEFRA published the full SFI26 Window 1 breakdown on 9th September - one action is in three quarters of all agreements.
🤓 Once again I've run all the data against SFI24 uptake. Headlines first, then an explanation of the three charts:
📊 7,000 applications (6,900 eligible - a little over the 6,500 estimated previously)
✅ 3,100 were live agreements by 1st September, 3,800 still processing (RPA have been quick so many are probably live by now)
💷 £9,700 a year is the average agreement size (69% of applications came from farms under 50ha)
🌾 There are now c. 48,000 live SFI agreements (across SFI23/24/26), covering c. 38,300 farm businesses
Now the full analysis:
Chart 1 shows the breakdown of actions chosen by agreement. Low input grassland (CLIG3) is in three quarters of them, against two in five under SFI24. Most of that comes down to who could apply, and the changes made in SFI26 - everything that climbed is grassland or boundary work, everything that fell is either arable-only or had its rate cut.
Chart 2 shows the breakdown of actions chosen by total cost to DEFRA, so is weighted towards what the larger farms picked. Hedgerow management (CHRW2) is in 39% of agreements but only 3% of the money. Low input grassland still took 22p in the pound against 9p under SFI24, and herbal leys (CSAM3) went the other way, 17p to 8p.
Herbal leys is the odd one out, and the one I would carry into Window 2. It fell on land these farms could perfectly well have used, and about two thirds of that is the rate cut rather than choice, so expect the same again.
Chart 3 is hectares per agreement, which says more about the 30% who applied with over 50ha. The grassland actions came in at much the same size as SFI24, but the precision ones did not - variable rate nutrients (PRF1) averaged 146ha here against 243ha under SFI24. Even Window 1's larger farms are small by SFI24 standards, and Window 2 brings in the ones that set those averages.
🛰️ Variable rate nutrients is the one I would watch, it comes back when the big farms do. We are building it into Soil Benchmark off existing soil maps and satellite imagery, so scanning does not eat most of the £27/ha.
📌 Rough maths on the queue. 6,900 eligible applications at £9,700 is about £67m chasing a £60m window. Window 2 is £230m, so at Window 1's average that is roughly 24,000 agreements, at the SFI23/24 average of £21k it is 11,000.
My guess is nearer the bottom of that, as it is bigger farms applying this time picking dearer actions. Budget counts as allocated when you submit, not when you get an offer, so waiting on DEFRA's 75% post is waiting too long. Worth getting it drafted before the 22nd.
🗓️ Benjamin ran five bitesize sessions on SFI26, half an hour each, ending with planning an application in SFI Shepherd. They are all recorded - email me and I'll send them over.
All 71 SFI26 actions are written up here: what changed in SFI26. And if you want help drafting an application, SFI Shepherd works out what each field is eligible for, flags actions that clash, and lets you compare a few drafts side by side.
