Can The P2 Solo Make Oxford Nanopore Profitable?
I really don’t mean to keep talking about Oxford Nanopore… but unfortunately it. just. keeps. coming. up. in. conversation.
Today we were chatting on the Discord about Albert’s awesome mergers and acquisitions sheet. I’m not much into trading, only holding stock acquired through prior employment, and in general selling all tradable shares at the earliest opportunity1.
But interestingly, Albert has put the probability of Illumina acquiring Oxford Nanopore at 0. There are many solid reasons for this, not least of which is that Oxford Nanopore has so called “anti-takeover” shares.
The anti-takeover shares are interesting… and seem to expire some time in October 20242. The suggestion was that by this point ONT could be profitable, saving them from a potential acquisition.
Analysts don’t see ONT as reaching profitability until 2026. And ONT management seem to have been selling off equity too! If we can make a convincing argument that ONT will be in profitability by the end of 2024, this would be pretty interesting and might give us more faith in the long term prospects of the company as an independent entity.
So let’s explore the issue…
On discord we discussed the P2 Solo in particular. And if this instrument could help drive the company toward profitability. In H1 2022 ONT lost 30.2M GBP. Oxford Nanopore's 2021 net loss was £167.6 million, 2020 £61.2 million. Overall, I’d estimate they are running at a loss of about $70M/y.
There are obviously a few different ways of looking at this. Albert discusses it in terms of how many P2 units at 50% utilization on Discord. But let’s look at it in terms of how many PromethION flowcells you’d need to sell here.
PromethION flow cell pricing varies by pack size. Taking current pricing from the ONT site, the following graph shows the number of flow cells that ONT would need to sell at 50% margins3
So between 100000 and 200000 flow cells. How many do they currently sell? Not sure, but let’s try and guess.
2021 revenue from “S3” (PromethION) customers was 55.7M GBP. I can’t find full accounts for 2022 yet (I believe they should be available later this month). But if you exclude a payout from the NHS, revenues seem to look fairly flat for 2022. So I’d estimate about $66.5M USD for current PromethION revenue.
They could therefore be selling up to 200,000 flow cells currently. I suspect the reality is that this revenue is heavily skewed toward instruments however. But in any case they’d need double this to reach profitability.
Now the good news is that between 2020 and 2021 they did grow the S3 segment 213%. The bad news is that this growth was “mainly led by the EGP”, and that there has been a “decline in EGP consumable use”.
In any case, to drive growth, the P2 Solo is going to need to attract large numbers of users. Currently the S3 (PromethION) segment is composed of a few (about 50?) large customers.
How many P2 Solo customers would you need? Well we need to sell between 100,000 and 200,000 extra flow cells a year. I’d estimate you could do 4 runs a week of a P2 at essentially full utilization. ONT themselves also quote ~200/y. This would be at least 500 users at 100,000 flow cells. But given that these users would likely be buying in bulk packs at lower margins probably closer to 1000 users…
That’s at 100% utilization. Which seems very unlikely. If you’re doing that much PromethION scale sequencing, it seems more logical to buy a PromethION 24 or 48.
I suspect a 25% utilization is more realistic, this would be ~2000 instruments (similar to the current Ion Torrent install base). And at 10% utilization (which seems totally reasonable for a small lab, doing a run every other week) you’d need ~5000 instruments (similar to the current NextSeq install base).
To me, selling this many instruments seems extremely challenging. Particularly given the current market size.
But it will be interesting to see what happens and perhaps the 2022 full year accounts will provide more insight!
I haven’t verified this, best statement I found in a quick google search is from this document, some time in H2 2024 seems right.
The latest margins I’ve seen are in the region of 50%.
Interesting post, as always. Naive question: what does EGP stand for? I googled briefly but was swamped by all kinds of irrelevant hits..