With Matthew Hodgson and Amandine le Pape.
The journey to world-class decentralised, secure communication.
What got you started on your journey with Element?
Matthew Hodgson (MH): I've been really interested in online communication for as long as I can remember and I'm a geek (if it wasn't abundantly obvious!). I spent way too much time as a kid playing with computers that my dad would bring home. I can remember at one point getting obsessed with some pretty obscure computers made by a company called Torch, slightly pre-dating the PC. A friend of mine also got a computer, and we were trying to call up each other and figure out what we could send with both of our modems. I then started to think about video - which was crazy at the time as it pre-dated voice and video - and the computers just didn’t have enough graphics to do it. So I was thinking, what if you could do animated vector imagery, a bit like flash cartoons, as flash cartoons could be teeny weeny but could render quite prettily. So basically, what if you could have a device that would take a video and do edge recognition and produce a version of your face - that could be compressed down to be tiny - and then the other person could see you on the other side of the screen. I must’ve been 8 or 9. And it’s all gone on from there.
I started doing physics at university but got moved over to Computer Science because I had started a startup; Cambridge didn’t want me running a business from university, so they gave me the choice to leave or to move course. Met some guys on that course that I did a group project with, which was to build an instant messenger called ‘Project Foxtrot’. Our supervisor took one look at our work and said it looked a bit like MSN messenger and that he’d wanted us to do something different, like build a decentralised communication network. We’d never thought about doing decentralised communications before and wished he’d told us that three months earlier when we started the project, but it was something we took away for future times.
After university I started working with one of the guys from that group project on a tech startup called MX Telecom, which was doing SMS aggregation: APIs for sending and receiving text messages. I joined on the condition that I didn’t have to do anything with SMS because it was boring and ancient - that was 20 years ago and it was already boring and ancient then. I said I wanted to work on voice and video and some other messaging stuff. They let me do that, running their black ops communications team. We had some fun projects though, like interactive video streaming. Back in the day, Big Brother used our service and technology, and then we got acquired by Amdocs in 2010, a big Israeli telco supplier. And we continued working on voice and video technology.
The iPhone had just come out, so we were wanting to make video (the ones we’d done for Big Brother) work on phones, rather than running it on the server. We were the first people in the world to do voice and video telephony on the iPhone, but unfortunately we never released it publicly. That was before iOS came along. And then in 2012 everything changed.
Amandine le Pape (ALP):
My interest in communication came a bit later. I started studying electronics and realised that I found the telecoms side of it pretty interesting and started looking into 3G and 4G networks and wanted to really learn more. I wondered how we could manage to get people to actually share anything via their phones. At the time, 3G was the main way to do so. I decided to pursue this further and worked for a startup in Brittany after various detours on the way. The startup was doing mobile app development, to share video or anything you wanted on your mobile. It was way before iOS and Android existed so they were developing mobile apps for sharing things on Nokia phones for example and all sorts of crappy Java phones, and making it easier to do so. This company was acquired by Amdocs a month after Matthew’s company did.
So that’s around the time everything got started because there was a new Division President, Gary Miles, who at the time was leading the Digital Services entity, who came across Matthew’s work in London. He was looking for all the exciting new stuff coming out, as he said “we have this voice messaging team, but we’re not doing anything with them” and “we also have this mobile team back in Brittany” who could assist with voice messaging - and together we could build an equivalent to Skype or WhatsApp and sell it to telcos. The project ended up in my office, so there was me and my colleague, and our boss came and said “hey guys, I need a business case on this project” and I said yes and jumped on it. So that’s when Matthew and I started working together, building messaging apps that we were selling to telcos in Singapore and Brazil, and we had quite a lot of success.
MH: It was quite fun and we shifted into entrepreneurial mode in 2012. Before then I was more of a tech lead and a project manager, and Amandine was doing pre sales for the company in France. But once this opportunity came about to put together this Unified Communications business case, we were very lucky to be given the remit to run it as an incubated startup. So despite being part of a 25,000 person company, we had sufficient confidence from the powers-that-be to carve out about 12-15 people, to run as a completely independent business unit with us calling the shots. We basically were learning how to run a startup. Traditionally I don’t think this has worked out very well and we were lucky that we had the sponsorship and the commitment from very high levels at Amdocs to make it work. We were the first time they’d ever done an incubated startup or incubated innovation or intrapreneurship, but we got to benefit from being the canaries in the coal mine. And this is important to the backstory of Element, because the team at Element now is the original team from back then. I started building the team back in 2003 at the old company, so we’ve been working together for 15-16 years, which is pretty unusual for a startup.
ALP: After doing that, that’s when we realised that even if this small incubated business was successful, it wasn’t really worth it, as we were never going to take over the world. Whilst our app was having good market penetration with the telcos, having Whatsapp against us was huge. It was 10-12% market penetration for our telco app vs 98.6% for whatsapp. So that’s when we sat down with the team and said “if we really wanted to make a difference, what would we do to fix this industry that’s a bit broken, because it’s managed by the big corporations, who own the entirety of real-time comms?” And that’s where Matrix comes in.
Has the vision of Matrix changed since that point or has it stayed pretty consistent?
MH: The vision has stayed freakishly consistent in the grand scheme of things. I think we first seriously started talking about it in November or so of 2013, where we were wondering if we could take what we’d done with unified comms at Amdocs and open it up to a much wider audience and become much more disruptive, rather than just being a rebranded WhatsApp or a clone of WhatsApp for a telco. It really just started as an idea, but then we saw it as a light at the end of the tunnel and something to work towards. We fairly seriously started pushing it with the powers-that-be and tried to convince Amdocs that we should pivot this successful, profitable, revenue-generating business unit into a moonshot to create Matrix. There was a lot of validation needed, going to partners and to MWC in Feb 2014. We sat down with as many mobile networks as we could and as many competitors of Amdocs. We asked them if this product would be useful for them. We spoke to Oracle and all sorts of other people in the telco space and the feedback was pretty positive. The response was “this sounds pretty crazy, but if you guys can do it, why wouldn’t we get involved?”.
By the time we got to May 2014 we had a pretty crisp pitch that we'd been delivering to the CEO at Amdocs, which was basically here is our 3 point plan over the next 18 months to go and create this protocol, to liberate instant messaging and here’s why it’s the right time to do it. People will be able to run their own stuff and we’ll build a flagship client to push the technology forward; it’s not just going to be an R&D exercise, it’s going to be a concrete project and something that end users can use. The whole idea was that it’d be an ecosystem replacing various things such as email. We were pitching them for that funding to monetize it, so they could have a front row seat and be part of a very disruptive technology. And now when we look back on that six years later, it’s possibly the only deck I’ve ever seen where it was executed on and we kept to the timeline we proposed, possibly to the detriment of Matrix because it was quite an aggressive timeline but we got through it anyway. We got the green light in May 2014 after the final presentation of that vision and we got the whole team working on it frantically. By September 2014 it was basically the thing that we have today, but we’ve spent the last 6 years debugging it, because we really rushed to get that initial cut out of the door. But it’s really strange to be able to roll back the code and look at where it was on launch day and the amount of features that we had then are basically the things we still have today.
I think we’ve ended up with a reputation in the industry for having had this crazy moonshot vision that sounded totally implausible, and then bashing our heads against the wall for six years trying to make it a reality and then, eventually, getting more and more traction. And those are the crazy Matrix people for you - that had this crazy idea that they eventually executed on.
ALP: The only thing I can think of, in terms of that original vision, is the evolution of our realisation that this is going to be successful, not only if we actually deliver on secure decentralised communications, but if we also have some moderation tools on top of it to manage the network. But that’s more of an evolution because, in terms of the value proposition, I think that’s exactly what we had in mind.
MH: The ‘what’ we are building hasn’t changed, but the problem we have to solve has changed. We had an ‘oh shit’ moment a couple of years ago when we got it working well and lots of people were piling on and using it to communicate and we were starting to see, not just moderation problems, but also social interaction problems.
You could have communities of people popping up using it to communicate, you might just not want to have any contact with them. It could be because they're nasty people doing illegal things or it could be because you just don't like their politics or whatever it might happen to be - so you might just want to dial down their content.
At first we just thought that’d be a nice feature to have and then it rapidly turned into us thinking this thing was never going to work unless we nailed the concept of controlling and filter and empowering users to control what they can see in it. And this is an unsolved problem. And it's a really big problem because it's basically the roots of the disinformation, propaganda and abuse that we see on the general internet today, nothing to do with Matrix. So we realised that for this to be successful, we essentially have to solve the problem of fake news and online disinformation. That's probably been one of the biggest changes, and wasn’t in the original pitch.
What’s your experience been of hiring Game Changers?
ALP: It’s hard, but on the tech side especially we’re super lucky that we're working on an open source project, as it makes it much easier to find exceptional people, and something we try to do is really keep our bar high. The team from the get-go was made up of really great people, so we’ve been trying to stick to that, because we believe that the best way is to build an org which stays a bit lean then continues moving forward quickly, particularly on the technical side.
MH: The nature of the project means that there's a huge community of developers out there who are building on top of us and it encourages good people to get involved. And it's a no brainer for us to then approach the best of those people and say “hey do you want to come and work on creating this?”. Netscape in the early days of the web was the place to work, if you wanted to work on this crazy web thing, as there was nobody else who was building and investing in solutions for the web in the early days, and it’s the same here for Element. It creates a halo effect, because we have a reputation for really good developers running around. Open Source credibility is then appealing for other roles, whether it’s marketing or sales or anyone else who wants to be a part of that. For a long time we were only developers; we got up to about 36-37 people or something crazy, without really having anyone outside of the engineering org. More recently we've been hiring up: marketing, sales, finance, HR and all the other bits of the company. We’re building on those strong foundations with the existing team, particularly one where lots of people have been working together for ages and there's a lot of trust, so hopefully we can get more people into the game - without being too cliquey!
ALP: Ultimately we’re all very mission driven. We’re not just trying to build an “Airbnb for dogs” or whatever, we’re trying to completely change the way in which the world communicates. It means that, for example, even in the sales organisation, they have to understand that the priority is the ecosystem, and the bigger the ecosystem, the more money we’re going to be able to make, as we’re a smaller slice of a bigger pie. And it means that people really have to understand this idea of building an ecosystem, they have to really have to understand the mission and that also comes by pitching the mission to customers. So I think being mission driven has been helping a lot in finding good people.
And so what have been some of your biggest challenges as founders, how do you kind of cope with the stress and the toll of setting up a big business?
ALP: Just touching on the hiring theme again, when you’re raising money and trying to find people who actually understand that we're trying to build an ecosystem and we're talking about a long term vision rather than a two year stint - this can be a bit tough. The other thing we struggled with was hiring sales and marketing leads, because we both come from a technical background and had never managed any sales or marketing people before. Despite reading many books and interacting with sales and marketing people across the ecosystem, it’s so different when you’re trying to hire for your own team and company - that’s been quite a complicated process.
MH: Having to raise money when you’re doing something a bit unusual is hard. The “Airbnb for dogs” is an easy concept to understand, and I think there’s a lack of understanding that we want to fundamentally change communications. If we were just cloning Slack it would be easier. The fact that we've got a much more complicated story of open ecosystems, moonshots and terraforming new civilisations means that investors often glaze over. The good news is that it ends up being a self-fulfilling prophecy, as we end up talking to the people who actually get it, as Notion and Jos did. And then it ends up being a stronger partnership, because everyone is on the same wavelength. It’s a bit like dating; you find yourself constantly wondering if you’re going to find the right one and then you do find that match, but the process is very stressful. The spinout of Amdocs was very stressful. We ended up passing in such a way that we had to pool our resources together, without any funding, and persuaded everyone on the team to just jump off a cliff. We had three months left to find sufficient seed funding to support the team, otherwise the whole thing would fall over and be dead before it began. It was a very stressful period, as we had to persuade this team we’d been working with for ages to take this massive risk and to put together their own resources to allow us to pay the rent and salary before we actually got funding.
ALP: But we only lost three people, and one of them ended up coming back!
What’s surprised you the most about your journey?
MH: For me it’s been the broadening of skills. I’ve spent a long time as a software developer and I had focused on honing in my skills quite narrowly on that relatively small domain. Then when we started the unified comms division at Amdocs, suddenly we needed to build out the P&L for it and get it passed by the finance processes. We suddenly needed to work with the Amdocs sales team, provide them with marketing material and build out sales compensation plans. And it was like “what other random new disciplines are going to be added to the menu today?”. Then we had to design a logo for Matrix etc. And that’s what surprised me the most: I never thought I’d be sitting looking at excel, looking at P&Ls, figuring out headcount etc.
ALP: I’d say the same goes for me, but also the fact we’ve been networking with all these different people, from VCs to ministers. It was never something I expected to be doing, but it also ended up being easier than I imagined. But on the other hand, there are parts of this role that are still very complicated to me, such as managing people. Obviously I still have to do it, but overall the difficulties have come where I wasn't expecting them to come from.
If you could go back to when you were first starting out, what would you say to your younger self?
MH: Something that’s always bugged me is whether I should have started this sooner. At the time, I convinced myself that I really wanted to understand the technical side of things as deeply as I possibly could - not just theoretically, but actually in practice. I wanted to know what things go wrong when you write software and basically have real industry and life experience before going and doing my own thing. I always did want to do my own thing from the outset, but in the end that process took longer than it probably should have.
We started Matrix in 2014 and we were running it as an independent company by 2017. As I started working in 2003, there was a 15 year or so run up to that, whereas now I think maybe a 6 or 7 year run up would’ve been better. You have much higher energy levels in your 20s, in comparison to your late 30s, as well as the timing with having a family. I’ve got 2 kids now so I’m trying to be a vaguely responsible parent, as well as running a startup - my eldest is pretty much the same age as Matrix, so I’ve seen her grow up over the last 6-7 years at the same rate as Matrix. In hindsight it would’ve been less stressful to do this as a startup earlier on. But that being said, we wouldn’t have been able to do Matrix at that point, as the whole structure is the result of the additional maturity and experiences we’ve had across the team and ourselves across the 14-15 year period. It also allowed us to get into a position where we could do it as an incubator startup - it’s not a business where we could have bootstrapped or started a conventional seed business.
So it did have to happen at the right time. If we’d have done Matrix 10 years earlier, it would’ve totally bombed, as the world would not have been ready for it at all. I’m hoping we got the timing right, especially with everything that went on with Cambridge Analytica, an increased focus on privacy and geopolitical shifts over the last few years. It feels like the world has converged with the path we were on already.
ALP: I think I would’ve been involved in younger startups earlier on in my career, in order to learn faster from them. Also when people say that, when running a startup, one day you’re on top of the world and the next you’re at rock bottom - it’s so true, so be ready for that!
What advice would you give to an entrepreneur that is considering taking VC funding?
MH: I’ve seen it both ways. The first company I joined early on, on the tech team, was interesting, as it was entirely bootstrapped and we went through a $100m exit. It was a point of massive pride that we never took funding and it was really interesting to see the positives and the negatives that it came with. The positives were that we just go to make up our own playbook, our own rules. We had a huge competitive advantage over other people in the industry, by just being able to do our own thing. On the negative side, we often had massive blind spots which, had we been able to talk to anyone else in the industry (such as investors or people in their portfolio) we’d have benefitted from a shared wealth of knowledge. And if we’d had that knowledge, we may have had even bigger outcomes. And in retrospect, if the company hadn’t been bootstrapped, I think it could’ve been way more valuable if there were other investors pushing for a more valuable outcome, rather than it getting to a point where the two founders were happy to just split the money and retire in their 30s.
So if you do want to get VC funding, I think it’s critical to find investors who you get on with, feel comfortable with and who are on the same page as you. They have to be people you want to work with - it’d be a disaster if not. It’s a bit like choosing the wrong employee; we have a company credo that it’s better to leave a role open than hire the wrong person into it. We’d rather limp along undermanned than have the wrong person, and it’s exactly the same when it comes to investors. Hold out for the right investor, who clicks with you, that you can work with and that you can trust.
ALP: Find out what they can bring to you, as if they’re just sitting on the board it’s not actually very helpful to you. If you have to give a bit of your company to someone, the best is actually making sure they can deliver too either by providing general advice, technical help, help with hires, advice on how to structure your org, networking help and intro to potential customers - all these sorts of things can make a massive difference.
What’s been your experience of working with Notion and have we lived up to those expectations?
MH: Yes definitely, and we’re even out of the honeymoon period now, coming up for a year in the Notion portfolio. I’d genuinely say it’s been very positive working with Notion and the other investors who came in on our Series A. Part of it was probably the self-selection process, as those we didn’t get on with didn’t want to invest anyway. And those that did get the vision and were willing to work with us, are the investors we now have. The different funds all have very different flavours in terms of their areas of focus: Notion has the Platform team, operational experience and the network that goes with that, FirstMinute is all about the incredible contact lists that they have, and then Dawn has a lot of people in their portfolio in a similar space, particularly around public sector and go-to-market, to compare notes with. In all instances, taking VC funding has been a genuinely positive experience. And I do hope that’s not unusual.
Something worth acknowledging is that the investors and the startups are playing fundamentally different games. I think a lot of people come into it with the assumption that it’s like hiring an employee, who you expect to be playing the same game. Whereas instead, recruiting an investor means they're going to want a return on their fund, they will have responsibilities to LPs and they’re just a completely different set of motivations to that of the entrepreneur who, for example, might be aiming to build a product or trying to become well known or might be trying to get rich. There’s obviously an overlap of goals there but it’s important to articulate early on exactly what motivates the various different parties and figure out where there is, and is not, alignment. There’s no point pretending it’s always hunky dory all the way through, because there are always going to be areas of conflict, for example when it comes to valuations, or funding rounds dynamics, or the governance of the company. Having a relationship where you can be transparent about the places where one needs to collaborate and the places where one is optimising for different things, is super important. It’s the sign of a healthy relationship.
ALP: It has been great to be a part of the Notion portfolio and to have the shared experiences that come with that. It’s been really useful to chat to everyone in the portfolio, about any topic we may need to address. It’s a shame we can’t connect with everyone at the founder retreat this year but we are already looking forward to next year!