In partnership with

It’s time today for another of our popular investor interviews. Today we look at the world of deep tech university spinouts, with a specialist VC in the space.

Let’s meet Johnathan Matlock of Empirical Ventures…

– Martin

Free Notion and Unlimited AI

Thousands of startups use Notion as a connected workspace to create and share docs, take notes, manage projects, and organize knowledge—all in one place. We’re offering 3 months of new Plus plans + unlimited AI (worth up to $3,000)! To redeem the Notion for Startups offer:

1. Submit an application using our custom link and select Beehiiv on the partner list.

2. Include our partner key, STARTUP4110P67801.

Apply Here

How Empirical Ventures approaches deep tech startups at their earliest stage

Who is Johnathan Matlock?

Dr Johnathan Matlock is co-founder and general partner at Empirical Ventures, a pre-seed and seed investor focused on science-led startups.

With a PhD in practical synthetic organic chemistry under his belt, Matlock has his own science credentials.

The firm’s portfolio so far includes names like Silveray, Muddy Machines, and Exciting Instruments, all of whom we have previously profiles in PreSeed Now.

This conversation has been edited for clarity.

MB = Martin SFP Bryant, JM = Johnathan Matlock

MB: So tell me what Empirical Ventures is all about…

JM: Empirical Ventures exist to back the best entrepreneurial scientists at pre-seed and seed, where we act as a domain expert from a fund perspective to try and help solve the problem of ‘who's my lead science investor?’

A lot of the venture capital market often says ‘come back when you have a lead investor’. And there's an added barrier for deep tech companies, which is they say ‘come back when you have a lead science investor or a domain expert investor that can help us validate the technology’.

I can't tell whether that's a classic polite pass or whether it's also really true, because having been embedded in the venture world for a number of years now, you don't often come across that many PhD-trained scientists who are now actively full time investors.

I think that's changing. There's a number of programs that have been set up over the last few years, things like the Newton Venture Program, to try to promote more STEM-to-investor career pathways.

But that was really why Empirical Ventures was set up in the first place. We are scientists by background, myself and [fellow co-founder] Ben Miles have got PhDs.

We've both been operators in a successful startup that went on to be acquired, so we’ve seen all of the structural challenges that come in those really early years, like ‘how do you do chemistry in a lab when all you have is thin-layer chromatography to look at your reactions?’ Like ‘how do I get access to big, expensive equipment? Where does that need to be? Where do I need to locate myself to access the best talent?’

I've done a huge amount of angel investments over the last six years. I've done about 70 investments at pre-seed and seed as an angel investor. I've also allocated capital as an LP into emerging funds because I've got a bit of a thesis from my family side of things that we are interested in backing first-time fund managers, because I think that there's the best alignment between incentive structures and small funds, and I think that they're just easier. They're more likely to return because of the size of funds relative to exit scenarios that you typically getting in exits for businesses.

And so it's that combination of scientists, operators, entrepreneurs, and investors, that positions Empirical a little bit differently. We approach the whole process with the idea of a bit of scientific inquiry, and using that process to form a unique insight about the companies that we invest in.

For example, as a former chemist, I spent 50% of my week doing the same reactions over and over again; not because they didn't work, but often, just because that's how you do chemistry. You always have to do ‘A plus B’ to get to the interesting bit further down the line, which means you're always repeating chemistry.

And so when CheMastery came along and said ‘we're going to build a robot that will remove the need for you to manually do the repetitive tasks’, the insight there, for me, was obvious. This was something could really change workflows fundamentally for a chemist. And therefore, there's a big opportunity here for us to back a company at a very early stage with, you know, an idea and an automated robotics platform that could solve that problem.

MB: Empirical Ventures invests in a wide range of different types of deep tech startups at an early stage. So how do you go about assessing them when you don’t have first-hand expertise in their field?

JM: I talk about this quite a lot: 'what is the purpose of due diligence? And it's twofold. There are obviously corporate hygiene factors, like if they don't own the IP, or the company's been set up in a weird way, or something on those lines, those are just red flags you can't invest in the company.

But then you have diligence related your to your conviction about investing in a deal, and given that all companies have risk, what you're trying to work out is ‘do I fully understand just what the risks are?’, and ‘do I believe that the company can de-risk those things with the funding that we're giving it?’

When you are working in a scientific field and you submit something for peer review, it typically goes to three reviewers, and those people are genuine experts on your sub-field of science, and you can get three different responses from those people.

You can get one saying ‘this is a great paper and should be immediately accepted’, one with a neutral response, and you can get one saying ‘this is an awful paper, don't let it anywhere near the journal’.

So we get that response from genuine experts in the field, that tells you that expert opinion is too often diluted with opinion, rather than factual evidence about where a company is.

So when we created our diligence in process, we said we want it to be a little bit like an onion. We want it to feel like as we get deeper in a company, we’re unveiling layers of detailed information, and that the company is only spending time providing that information that's appropriate to the stage they are in our process, because we're also conscious of founders’ time.

They've got to build a business and develop a technology, and the more time they spend answering diligence questions, the more challenging that is.

So we ask whether we can evidence what's happened. You've told us this has happened. Can we evidence it, whether that's through things you provide in your data room, or can we independently verify it with our own research?

And then we ask what you're planning to do. What core bits of information have informed your thinking? Tell us about it, and then we can form a view as to whether we have conviction in the deal based on that forward-looking plan.

With a standardised process, it allows us to look at the evidence of what's happened or not happened in the company. Are we domain experts, or are we experts in understanding how to analyse information? And it's very much that second one that we're trying to aim for, rather than being an expert in a specific field.

MB: What do you think about the university spinout landscape at the moment? Are there enough spinouts?

JM: I think there has been a huge amount of movement in that sector. In the Bristol ecosystem, Harry [Destecroix] who founded Ziylo, was critical to fundamentally reshaping the number of spinouts that came out of Bristol, and myself and Ben were a part of that.

That comes from creating infrastructure and entrepreneurial clusters that enable people to say ‘I believe I can do this, and I can give it a go’.

I think the UK Government’s Independent Review of University Spin-out Companies is an interesting start on trying to standardise IP license negotiations between tech transfer offices and spin out companies.

I believe that we could more fundamentally reshape how we view university IP in order to encourage more entrepreneurial activity.

I've not yet seen a good reason why you couldn't assign IP to the inventors, and then approach a university as a co-founder and say ‘we'd like to use your equipment and infrastructure, and we'd like you to support this grant application, and for that, we'll give you founding equity in the business as a contribution in kind for your investment in those services’.

And what that means is you leave enough equity on the table to incentivise the people who are actually in the business day-to-day, working seven days a week to make the thing happen. And I very much view startups like an iceberg; the bit that happened in the university is the bit that you can see above the surface. The actual route through to full commercial utilisation, and it being used by people in the world, and exits if you're an investor, is all below the surface. It's all to come.

And so that's my main challenge with respect to equity incentivisation mechanisms - they're not forward-thinking enough.

And that's also true of academic co-founders. You often see this, where people are academics and they'll co-found a business, but they're not really involved day-to-day in driving it forward.

So if I had a spinout and I had a really entrepreneurial PhD student or postdoc who was jumping into the company, taking all the risk personally, I'd much rather them have a much bigger equity percentage than the prof who maybe led on the project originally.

And so I do think that opportunity is open to us in the university spinout sector, because ultimately, these institutions are largely publicly funded, and we want to create an environment that encourages entrepreneurial, risk-taking activity.

The recent review has been useful in trying to create a bit more of a standardised template for these processes and from experience, the feedback cycle of negotiating licenses with tech transfer offices is typically quite a slow process.

And when you've got a startup that's only got 18 months of funding, the only advantage you really have over anybody else is your speed of execution, and so there need to be ways in which you could speed up getting to an agreed licence process, and if that means templating it, which isn't what's being proposed, but I don't, kind of see why that couldn't happen, then I really think you could catalyse more growth in new university spinouts.

MB: Is rethinking of the role of universities in spinouts, such as kind of giving them equity in kind , a viable approach? Do you think universities would go for that?

JM: There's basically two routes forward for IP when you're in a tech transfer office. You can either license to an existing company that can commercialise it, or, typically, if it's earlier-stage technology readiness levels, there aren't opportunities to do that license straight away, so the route forward from a commercial viability perspective is to put it in a spinout company and then raise capital to help de-risk the technology.

I don't know what the split is between revenues that universities make from pure licensing plays, and revenues that they make from eventual exits of equity or royalty streams that they own on licenses and spinouts.

That tension will probably guide a little bit as to how willing a university tech transfer office would be to say ‘for a spin out, we're going to take this slightly new approach. And if you look at with a venture capital hat on, there's a lot of analysis that shows that your portfolio robustness, in terms of not losing capital and performing well, is dictated by portfolio size.

So if I was a university, I'd be viewing the spinout option as saying ‘how do I increase top of funnel and get more companies out there, such that one or two then drive the performance of all of that spinout activity?’ And that's then a numbers game, it’s about how many opportunities can you get spun out and funded.

To do that, you need a talent pipeline to identify the most talented scientists in an organisation, and I’ve not really seen any universities take a talent-first approach across their organisation.

If I took a vintage of all new PhD students that are admitted into a university across all disciplines who are the best entrepreneurial people within that cohort of activity, and how do I match them to the most commercially interesting bits of IP in my portfolio?

MB: What kinds of startups are you looking for at the moment, and what does 2025 look like for Empirical Ventures?

JM: It's going to be a really exciting year. I believe we're still in a little bit of a funding winter, to be honest. Rounds are still taking a little bit of time to come together. People are kind of looking around at all the other investors, saying ‘are you coming into this deal?’ So I think there's going to be an increased number of deals that are syndicated.

From a sector perspective, we're really interested in is the OpEx [operational expenditure] of the AI infrastructure boom. I think the AI hype cycle has got everyone really excited, but fundamentally, it's a very different business model to the low CapEx SaaS era that we've been going through.

What you're going to need to win in AI at the moment is really deep pockets. If you're on the infrastructure side, you only have to look at the announcement from SoftBank and OpenAI in the US; $500 billion to build all the AI infrastructure you need. [This interview took place just before Deepseek opened up a new dimension in this debate]

So how can you potentially win as a small fund? Well, OpEx becomes really important if you're an infrastructure technology. And so any technologies that can improve energy efficiency in terms of the operation of those things - low-carbon electricity to run those technologies, all of these enabling technologies are things that we can look to back and then hopefully implement into large-scale infrastructure plays.

Also, defence tech, and specifically dual-use defence tech, is going to have a big year this year. I think everybody's been looking at the significant amount of conflicts that have been going on around the world. That includes cybersecurity, people developing novel technologies to infiltrate, IT systems of large organisations; all of that sector is going to get a lot of focus this year, so it'll be interesting to see how that plays out.

I think you'll have some funds that will go hard into offensive deep tech, and I think you'll have a lot of funds that sort of take the defensive deep tech approach, because they have sensitivities around the sorts of things they can invest in.

I think we're in that bucket, because we are looking to invest in technologies that help the world. But I do believe that if you want to live in a society that has the values that we do in the UK, there has to be some appreciation for how we defend those values. And so I'm pro-defence tech if it has that kind of dual use.

And then finally, I think space tech will be big.

We are seeing a dramatic reduction in the cost per kilo of getting things into space, enabled in part, by private companies like SpaceX, but what that's enabling is a whole new wave of technologies where people ask ‘what can I do in space that wouldn't be possible on Earth?’

We've looked quite a lot harder in space manufacturing, for example, things like pharmaceutical drugs or diamond semiconductors. We've looked a lot at novel propulsion systems; we've looked a lot at in-space rescue of satellites so if your propulsion system fails, rather than writing off the satellite, how can I do an in space rescue of that satellite to maintain its orbit? Space debris is another thing we've looked at a little bit as well.

So those are some of the sectors that we're quite excited about. Going back, we've got a company that we're looking at the moment that is adjacent to the small modular nuclear reactor space because I think those technologies will be mission critical to the AI infrastructure plays, and they were mentioned in Matt Clifford's report that he put forward to the government around the AI infrastructure play of the UK.

So I think those are the sectors that I think they'll be hot and it will be interesting to see if we can find really exciting teams to back in those sectors.

Back next week

We’ll have a fresh startup for you on Tuesday next week. See you in your inbox then!