Where the biology genuinely runs out
Qualifying work is where the outcome is not readily deducible. Engineering a microbial or mammalian host to express a target at viable titre, where pathway flux, toxicity or genetic stability is uncertain. Developing a bioprocess where yield or product quality cannot be predicted, resolving oxygen transfer, shear sensitivity, foaming or by-product inhibition. Novel downstream purification for hard modalities, viral vectors, mRNA and lipid nanoparticles, exosomes, where recovery and aggregation are not solved by known methods. Synthetic-biology work, engineered enzymes and genetic circuits where function is genuinely uncertain. And cell and gene therapy manufacturing process development, which is distinct from the clinical trial and often the real R&D: making a scalable, consistent process where none existed.
The work that feels like R&D but is not
The routine end of a biotech is large and skilled, and mostly outside the definition. Analytical testing and sample analysis to established protocols, routine QC, batch-release and stability testing, standard cloning and off-the-shelf gene edits, and routine production culture at set conditions are all testing or production, not the resolution of uncertainty. Regulatory and CMC documentation, GMP paperwork and trial submissions are administrative; the underlying process development may qualify, but the write-up does not.
Scale-up: the boundary that decides most claims
Scale-up is where biotech claims are won and lost. Moving from shake-flask to pilot bioreactor to production only qualifies where the scaling itself creates real technological uncertainty, the point where mass transfer, mixing time, heat removal or altered metabolism make performance at scale genuinely unpredictable. Where standard correlations carry the process across, it is routine engineering. The honest claim isolates the uncertain window and stops when a working process exists.
Sector pitfalls we would check first
- ERIS before the merged scheme. Check SME status, the loss position and the intensity test first: for many loss-making biotechs the enhanced route is worth materially more, and defaulting to the merged credit leaves money behind. The gates are on ERIS explained.
- Platform versus project. “We claim the platform” does not survive. Discrete projects seeking an advance do. Re-applying a proven platform to a new target is often routine.
- Contracted-out research. Heavy CRO and CDMO use makes entitlement fact-sensitive: the merged scheme generally gives the claim to the party that decided on and bore the risk of the R&D, not the contractor, and each contract has to be read.
- The overseas restriction. For periods from 1 April 2024, overseas subcontractor and worker costs are excluded unless the conditions genuinely cannot be replicated in the UK. Cost and workforce availability are not qualifying reasons, and this bites hard where vector or fermentation capacity sits abroad.
- Consumables apportionment. The largest cost line is also the easiest to overclaim: only the portion consumed in genuine R&D counts, mixed runs are apportioned, and output sold in the ordinary course is restricted. The rules are on qualifying costs.
Frequently asked questions
Does scaling a process up from lab to production qualify?
Only where scaling introduces genuine technological uncertainty a competent professional cannot readily resolve, non-linear effects on mass transfer, mixing, heat removal or metabolism. Adding capacity or applying standard scale-up correlations is routine. R&D ends once a working process or pilot plant exists; ongoing manufacturing is excluded.
We are pre-revenue and loss-making. Which route?
Very often ERIS. Loss-making, R&D-intensive SMEs whose qualifying R&D is a large enough share of total spending can claim through ERIS, which is materially more valuable than the merged credit. It turns on a strict intensity test and your SME and loss position. See ERIS explained.
Can we claim for the whole platform?
No. Claims are project-based. The qualifying work is the discrete project seeking an advance, not the platform or its commercial exploitation. Re-applying a proven platform to each new target is often routine, and is judged on the facts.
Our reagents and consumables are our biggest cost. Do they count?
They can, and in biotech they are often the largest line, reagents, media, cell lines, single-use bioreactor bags, chromatography resins, consumed or transformed in the R&D. Mixed R&D and production runs have to be apportioned, and output sold in the ordinary course is restricted.