It’s 4am on a Saturday morning. You’ve worked 20 hours straight. You’ve done so every day since… you forget when. Time to quit do that start-up you’ve dreamed about. But do you do it? Should you? Could you?
Well, as part of a new series we are interviewing legaltech and other start-up founders working in legal.
We will unearth the highs and lows of beginning and scaling a start-up. As Steve Jobs famously said:
“You can’t connect the dots looking forward; you can only connect them looking backwards.”
And this is very true of start-ups. Having worked at a start-up, and having friends who’ve founded their own start-ups – some successful, some not, start-up life isn’t easy or glossy as it might seem.
Lawtomated: What were you doing before Avvoka? How did you partner up?
Eliot: I was an associate at Slaughter & May in the corporate team, having also trained at Slaughters.
David: And I was a trainee at Linklaters. We both studied law and were firmly on the lawyer track. We knew each other socially and would regularly discuss ideas for start-ups and in particular how we could improve legal processes, which seemed a little dated versus the wider world.
Lawtomated: So what changed – where did the idea for Avvoka originate?
Eliot: Well one of the discussions we’d been having was document automation. At the time, there wasn’t a mature document automation function at Slaughters. After chatting to peers we realised many firms used ContractExpress, albeit to different degrees.
We scouted around and discovered a lighter weight, and lesser-known product, ShakeLaw. Like ContractExpress it allowed automation of document construction via a questionnaire, churning out pdf contracts.
I showed ShakeLaw to some colleagues at Slaughters, and a few of my team began using it to create contracts. It became clear very quickly what worked well and what worked less well. We also began to get a sense of why other legacy products didn’t always see the adoption you’d expect.
But then I went on secondment to Hong Kong and our legaltech start-up ideas fizzled out somewhat.
David: That paused our research. It was simply a matter of capacity. But when Eliot returned from secondment we discovered ShakeLaw had been bought! This spurred us back into action, validating our initial hunch that although document automation was a crowded market (even more so now), that there were significant market gaps we could exploit as a start-up.
Eliot: Yes, and in the meantime, my brother had been building a start-up in an entirely unrelated niche. Through him and his start-up network, we managed to find some initial developers at UCL. These two events really kicked things off.
Lawtomated: What happened next?
Eliot: We knuckled down and committed to working on Avvoka in our evenings and weekends more or less full-time within those hours. We started to define the business models we’d loosely sketched before, and began testing our ideas within our network.
Lawtomated: How did you finance Avvoka and get the first build off the ground?
David: At the start, it was mainly via our savings.
Eliot: But eventually we realised we’d need greater capital to take a run at the opportunity.
Lawtomated: How did you find outside finance?
Eliot: Luck played a part again. I happened to be chatting to a colleague at Slaughters and mentioned Avvoka, which at the time was very much a side project but reaching a tipping point. David and I needed to decide whether to go all-in or not.
Anyway, my colleague was excited by our start-up idea and mentioned a friend who is an angel investor. They kindly put us in touch and we hit it off, leading to our initial angel investment.
David: And to be honest, it was also a big validation that we were onto something worth pursuing.
Lawtomated: Sounds plain sailing?
Eliot: Haha, it actually wasn’t The first thing our angel remarked when we showed him Avvoka’s proposition was that we didn’t have a sustainable start-up business.
Lawtomated: Why was that?
Eliot: Well, and this was perhaps our first pivot as a start-up. We were not only building a document automation platform, we were also creating the underlying templates, precedents and… automating those for clients.
David: Yes, we were more or less proposing to do everything. We wanted to provide the tech, docs and automation service.
Eliot: Based on the advice we received we started testing different business models. We ended up with two pitch decks. One where we offered only the tech platform; another where we sold automated docs and offered a service to automate our customer’s docs.
Lawtomated: How did you go about designing your decks?
Eliot: We did a lot of online research and found lots of examples of successful start-ups, e.g. AirBnB. These helped us nail the tone, content and messaging.
Lawtomated: What sort of feedback did you receive when pitching?
Eliot: Mostly underwhelming at first! We realised early on that making our own documents, automating them and building a tech platform wasn’t going to scale. There also wasn’t the demand for pre-baked automated documents in the B2B market at the time. Early and subsequent feedback suggested users needed a simple, easy but powerful user experience to design, build and maintain their own automations. In part, this was because users perceived nuances between their needs and documents as very significant vs. peers, when in fact this was (and is) often not the case but nevertheless unconvincing for customers in the B2B space.
David: Yes, we were originally creating and automating things like NDAs, employment agreements and real estate docs and didn’t appreciate how much customers cared about making these documents fit their existing workflows and house styles, let alone their own playbooks for conditionality etc.
We also quickly realised we might find ourselves slipping into a slightly grey territory whereby we’d be seen to provide legal documents and therefore advice, making us liable with regards to the legal efficacy of any documents supplied via our platform.
Investors and other feedback recommended we distance ourselves from that risk, which was a further convinced.
As a result, we pivoted from docs plus services plus platform, instead dispensing the first two and expanding the platform element. We still help clients automate their own content, but we’ve managed to scale this via our newly launched Avvoka Academy.
Lawtomated: What do you think would have happened had you stuck to the original model of providing automated docs and automating your customer’s docs?
Eliot: Oh, we would have been killed in our tracks. It just wasn’t scalable. Automating docs, even with a no-code platform like ours, still requires a lot of critical analysis and a deep understanding of the underlying document and its domain. Automation isn’t automated.
We also realised it was hard to automate docs in such a way that appealed to a wide customer base.
Creating a scalable lowest common denominator automation for a document type proved very hard. Especially as a start-up with limited initial resources.
Most users wanted control of how their documents were automated, which is why we leave this aspect to our customers. That shouldn’t come as a surprise – many competitors adopt a similar model, albeit some do provide automated documents as part of their services.
Lawtomated: How did you win your first clients?
Eliot: We made some early inroads with clients such as Countrywide plc and Carlsberg via existing networks, our investors and events. Both clients were dissatisfied with existing solutions and really worked hard with us to help guide our development and invested along the way. We were really lucky to have their buy-in and support for our vision.
David: And our first law firm customer was Allen & Overy (A&O). We applied and were selected for A&O’s inaugural start-up tech incubator cohort at A&O Fuse in 2017. That was a massive boost for us and our profile.
Lawtomated: What benefits did you derive from A&O Fuse?
David: We received a ton of feedback from lawyers at A&O and their clients. It really helped us refine the product roadmap. It also led to a licensing deal with A&O, and many great introductions to A&O clients. We realised early on that both A&O lawyers and their clients were very interested in self-service no code document automation. They could appreciate the benefits of automation but struggled to get past the hangups they’d encountered with legacy tools, that they generally found harder to access and realise value from.
Lawtomated: How did you come up with the name “Avvoka”?
Eliot: I have French family and spend a lot of time there. I always liked the French word for lawyer, Avocat, and thought it would be a good word to use. Unfortunately, avocat.com was taken, so we tweaked the wording to Avvoka!
Lawtomated: Did you guys get some office space prior to setting up at A&O Fuse?
David: Yes we did. We were principally working from home on Avvoka at first, but when we went full time realised we needed a space to work and collaborate. Our angel investor had some spare space, which they lent to us.
Eliot: Yes, a stuffy server room in the basement of a London building. It was a little depressing in terms of the heat, lack of light and the fact we could see ankles rushing by the basement windows. But it did the job and gave us the start we needed. It was also free!
Lawtomated: Sales and marketing aren’t something lawyers get taught and are often a slightly cliched blindspot for most lawyers and law firms. How did you go about building out your sales and marketing?
David: To be frank, we didn’t really know what to do! But we started simple! We collected the contact details for FTSE100 companies, drafted an email setting out our proposition and then figured out how to do a massive mail merge. Once we hit send we waited for the replies to roll in.
Lawtomated: And did they?
David: Nope. We didn’t get a single reply! At the time, and in hindsight, we were super naïve. We laugh about it now, but it wasn’t that funny at the time. Especially having just quit two well paid jobs.
Lawtomated: So what did work? How did you build your sales funnel?
David: Mostly through networking in the early stages. We tried to leverage every connection we and our network could offer. This led to a much better engagement, even if only to get more feedback on our product and proposition.
The first breakthrough of significance was attending the very first LegalGeek event, I think in some basement in London in 2015! At the time it was just LegalGeek’s founder, Jimmy Vetstbirk, and around 12 other people interested in legaltech. But it was a good launchpad for further connections, and ultimately our long-standing representation at LegalGeek events ever since.
Lawtomated: That’s interesting. What was the awareness of legaltech in 2015?
David: It was pretty limited. Sure, legal technology has been a thing for decades. No one is disputing that. But I think the crucial switch is the widening of awareness and appreciation.
Back in 2015 and before, mention legaltech to most lawyers and their clients and it wasn’t really front of mind. Apart from a few notable exceptions, it was more a conversation between IT administrators. The number of lawyers and other business professionals interested in and actively exploring legaltech has significantly increased. There are also many more start-ups than today.
Partly this is due to a wider trend toward ops across almost every industry, not just traditional ops heavy businesses like manufacturing or retail.
The downside is that there is probably too much hype regarding legaltech at the moment, which can create expectation management issues.
Lawtomated: How did the first technical build go?
Eliot: Well, it was mostly trial and error. Our first version of the platform, which you’ll recall was designed around offering an automation platform plus automated docs, offered automated landlord and related documentation.
David: And it grew from there. As we garnered more client and prospect feedback, we deprecated our own content and focused heavily on expanding the platform features and user experience.
Lawtomated: At what point did you realise it was time to go all-in on Avvoka and quit the law? What was the tipping point?
David: We were working our arses off throughout our evenings and weekends. And of course, this was being done around two very busy full-time corporate law jobs at two Magic Circle firms. We realised early on that if the signs were good, we’d need to commit full time to Avvoka or see it die as a pipedream.
Eliot: It was great having something purposeful to do outside our day jobs. A lot of former lawyers say they hated the job, but David and I were a bit unusual in that we really enjoyed it but wanted to challenge ourselves.
David: Yes, starting a company is perhaps the ultimate leveller. We underestimated the graft and grit needed. But it’s been worth it. We would have regretted not taking the chance.
Eliot: Absolutely. I’d not change things.
David: Returning to the question, we were spending a lot of time writing and re-writing our pitch decks, spending a lot of our own money and grinding through product development cycles and testing. Having our law jobs gave us a safety net. It’s far braver to leave Uni and start a company. As a graduate, you’ve got skills, but not necessarily real-world experience or a paycheck and secure job to fall back into.
And on that basis, I think we both rationalised that we were in a very privileged position. We’d got good law degrees, had been trained by two Magic Circle firms and were qualified solicitors. We also had strong networks.
We took the view that if we quit Big Law, the worst case would be that we’d learn a lot of new skills, expand our network, have scratched an entrepreneurial itch, and in all likelihood be able to return to the law. In fact, both firms were very supportive of our decision and left the door open.
Eliot: It was the cost of doing nothing as well. We couldn’t shake the sense that if we didn’t make a go of it, someone else would, whether an incumbent adding new features or a new entrant like ourselves.
David: It’s worth stressing that despite what we’ve said, it was a tough decision. There’s a lot of guilt wrapped into it. You study hard from GCSE to A-Level and through University, competing at every stage to get the best grades and then again to get the best training contract at the best law firm and if you succeed you are rewarded – at a very young age – with a generous salary and prestigious job, in many ways totally out of kilter with the real world. It can feel stupid to give that up, or ungrateful.
But ultimately once we started to get financing commitments we knew it was time to pull the ripcord.
Lawtomated: And when did you decide to hand in your notices at Slaughters and Links?
David: Almost immediately after getting funding.
Lawtomated: That must have been a hard decision. It’s never easy leaving a job you enjoy, let alone one that also pays well and offers a steady career path. How did colleagues react?
David: Well, thankfully! I was actually on secondment to Hong Kong when we decided to push the button. Linklaters were fantastic.
I was extremely anxious – it was, and remains, the only job I’ve quit. I was worried about letting people down – my friends, family and colleagues.
Again, you invest significant time and often identity in becoming a lawyer at a Big Law firm. It’s quite hard to turn your back on that track. But Links were very supportive. The door was left open. The message was “give it your best, stay in touch and let us know how we can help”. They really appreciated the entrepreneurial streak. I also still play football for Links. It’s a great firm.
Eliot: Very similar to me. Slaughters (I think) were sad to see me go, left the door open but encouraging of what I wanted to do and why. My old head of department is now a valued shareholder, which is a great endorsement.
Lawtomated: What tips can you offer founders in terms of fundraising?
David: As Eliot noted, through a colleague of his, we were lucky to have angel investors to help through the journey. That was partly luck and connections.
We did also pitch our start-up to VCs and growth funds. But we realised that people tend to invest in things they understand.
I think at the time there was a lot of misunderstanding regarding what Legaltech does, and a significant scepticism that it’s not really a thing. It probably remains an undervalued market, especially when viewed more as business operations tech and less specifically niche to legal tasks.
Lawtomated: It sounds like you had a few knockbacks. Any advice on how to handle those?
David: Yes, in the early days we had a lot of “nos’. If you are pitching and you don’t get anywhere, don’t be disheartened.
One VC asked their lawyers to kick the tyres on our proposition. The feedback was that we were solving a problem that didn’t exist yet!
The slightly crazy thing was that we had, and had made clear to them, that we’d picked up some significant and paying clients who were already locking into healthy subscription licences. That didn’t seem to matter.
In these situations, it’s worth detaching and not taking it too personally, but instead objectively analyse what did or didn’t work and why to build this feedback into your next pitch. It can help you pre-empt and neuter objections before they arise, or perfect the messaging to make your value land well with investors.
Probably the best advice is to do your research.
Now that sounds obvious, but if you’re in legaltech, specifically research funds exposed to legaltech and adjacents. Look for funds invested in B2B start-ups with long sales cycles. You don’t want to waste time pitching to funds that lean into consumer software with C2C sales and short sales cycles. B2B sales are slow. A fund investing in C2C start-ups won’t have the patience for B2B start-up business models in our experience. Rather than convert, preach to the choir.
Find funds backing boring companies you’ve never heard of! Don’t get tied up in the idea of getting investment from the flashiest fund, which has its place, e.g. as a FOMO draw for other investors. But chasing the most prestigious funds doesn’t always convert to meaningful funding and advice.
All that said, the funding market for legaltech is much better than it was. High profile incubators such as A&O Fuse, MDR Labs and Barclays Eaglelab, widening press / social media coverage and some high profile investments and acquisitions have fueled interest and openness of funds.
Perhaps the biggest game-changer would be a few more unicorns in this or adjacent spaces. To be fair, there are already one or two. DocuSign is perhaps the biggest success story of recent years, not to mention the ever-enlarging platform plays of iManage and Litera. Investors are realising the legal market, whilst hard to sell into, is sticky with a risk-averse change avoidant customer base that, once bought into a licence, tends to stick rather than twist. Done well, that can build a cash-rich business with recurring revenue.
Lawtomated: I know the Avvoka team is investing in content marketing, specifically providing practical how-to guides and demystifiers regarding legaltech and automation. What’s your view on the current state of legaltech media and news?
David: There’s some great content out there. We really enjoy lawtomated and regularly recommend it to interested lawyers and other business professionals, as well as our clients and colleagues. But more generally a lot of legaltech content is largely rehashed and often not very actionable. There’s a need for more practical information, use cases, demonstrable ROI studies and so on. We are producing our own and seeing this trend increase, but there’s more to be done. It’s especially important as a means to help other start-up founders.
Lawtomated: Tech talent is often a blocker for non-technical founders. Knowing who to hire, for what, how to hire and where to find the right talent can be a nightmare. What tips can you recommend?
David: Eliot and I are totally non-technical. Although we’ve upped our tech understanding significantly, we were certainly out of our depth when we began.
Eliot: Absolutely, we didn’t know much at all.
David: Yes, when we began Avvoka we didn’t know who to hire, how to assess technical talent, what we should be paying and where to find the right people. We also didn’t know how to articulate what we needed.
Eliot: Going to start-up and tech meet-ups were great. We stumbled across Tom, our CTO. Tom was, and remains, a brilliant partner. In Tom, we found someone we could completely trust as the expert on different technologies and methodologies and how this meshes with product.
David: Some of the early tech decisions in a software start-up ramp up from 0 to 60 very quickly, e.g. information security, performance issues, scalability and so on.
There’s a huge difference between building something or knowing a bit of code, and someone whose passion is technology and understands product.
Those at tech and start-up meet-ups will be genuinely passionate about tech to attend, hence it’s a great starting point.
Lawtomated: Are developer agencies and similar useful?
David: Developer agencies can be a great way to start, but if it’s your first technical hire – as it was for us – it was really hard to unpick candidate skills, expertise, fit and know what to hire for. Getting your first technical hire right is key.
Having Tom on our team early on made further hiring much easier. Tom is an all-rounder from a tech perspective. He knows exactly what to hire, who to hire and where to find them. He also knows how to assess talent. He is also a great translator, able to turn what Eliot and I and our customers desire, into technical needs and capabilities.
Lawtomated: Is there a supply and demand problem for tech talent?
Eliot: Unfortunately, there is definitely more demand than supply in terms of quality. The salaries for good dev talent in London is astronomical, even for junior skills. In many ways, it’s more and more like the talent market in the US for tech talent.
David: Yeah, our lawyer friends are always in disbelief when we explain how much even junior developers can command, and the much better work / life balance they enjoy! Don’t underestimate this element, both in terms of how much money you will need but also the scarcity and time it takes to find good talent.
Lawtomated: How have you grown the technical team to date?
David: Tom has led its growth. A native of Prague he has a fantastic network there and has pulled together a fantastic remote team through his networks in London and Prague. Our tech team was mostly remote pre-COVID, which set us up really well for the remote working world in which we find ourselves today.
One challenge has been that senior developer talent is perhaps even scarcer than general technical talent. Many of the best senior developers are swept up by huge salaries at the largest tech companies. Motivating them to join start-ups is tricky without strong incentives.
The best advice I can give is this: take your time making your first technical hire, get to know them and choose someone who is well connected, knowledgeable and able to quickly build a team around them. Knowing tech isn’t enough – they need to know what, who, how and when to hire.
Lawtomated: In terms of product, legaltech products have an especially high threshold for what constitutes a ‘minimum viable product’ or MVP testable with prospects. Was this a challenge? If so, how did you overcome it?
Eliot: Yes, the MVP threshold for legaltech is high. We and peers find legal buyers are extremely demanding. Legal buyers are very vocal and eager to provide feedback, sometimes too much.
It can be very overwhelming trying to product manage what is and isn’t a valuable piece of feedback.
Legal users tend to be unforgiving, even with regard to trivial details. Partly this is because their day to day is centred around attention to detail and perfectionism.
Meeting this standard across every aspect of the product is tough. It makes it hard to do the things typical in the start-up world, e.g. agile development and fast feedback loops.
In a lot of cases, we’ve found product versions need to be more or less product ready for the highest common denominator in terms of enterprise requirements, often from day one, before anyone will take you seriously.
This can lead to a tendency of legal buyers to disregard products unless they are perfect and in particular perfect for their particular way of working, which may be very niche despite the underlying jobs to be done being generic or commonplace.
David: Yes, the MVP concept in legal is pretty absent. It’s a little frustrating because the MVP threshold in consumer apps is lower, albeit the quality is very high now – particularly in user experience.
Lawtomated: Why do you think the MVP threshold is high in law?
David: It’s probably due to legal training and the role lawyers play in society. Lawyers are trained from day one to cultivate extreme attention to detail. Friends in other professions routinely joke about how pedantic they find lawyers.
But that’s a little unfair given lawyers are largely paid to manage other people’s risk, so are necessarily on top of every detail, no matter how insignificant. The cost of getting stuff wrong in legal is often disproportionately high. And of course, any software needs to meet that standard.
When testing product with lawyers it seems tricky for them to switch gears from “lawyer mode” to “product” or “user mode”. By that I mean some of the feedback we get can place disproportionate emphasis on trivial details, e.g. the fact a workflow isn’t 100% a facsimile of a traditional one despite the workflow being significantly better in overall terms. This is another issue. Process improvement within the law is still a little immature. Sometimes data-backed ROI gets overshadowed by the trivial.
There is also often a blinkered rush to digitise existing processes, with little thought to whether the existing process is any good. That also distorts feedback on our product. But we’ve learnt a lot about the product and how to pushback and ask the right questions to separate needs from wants.
Lawtomated: That reminds me of the famous Bill Gates quote that “the first rule of any technology used in a business is that automation applied to an efficient operation will magnify the efficiency. The second is that automation applied to an inefficient operation will magnify the inefficiency”.
David: Absolutely. We see this one a lot. Handled the right way, it’s always better to map and then challenge an existing process before rushing to automate it under the banner of digital transformation.
Lawtomated: In some ways, lawyers do practice a type of MVP concept in their own work. For instance, when you draft, the first version is usually a prototype rife with placeholders, footnotes and multiple TBC / TBD items. Through successive user feedback from clients, and adversarial counterparty feedback you end up with a finished product.
David: Yes I’ve not thought of it that way before, but there is an analogy. I definitely see a lot of parallels between the jobs to be done in law and other industries, albeit in other industries they are often labelled at the task level, e.g. project management, product management etc, rather than being lumped under legal-specific terminology as is the case in law.
Lawtomated: When starting Avvoka, what do you wish you knew then that you know now?
Eliot: Haha, almost everything. No, in all seriousness it’s a little crazy looking back in terms of what we didn’t know. I think had we appreciated the gaps in our experience we might have decided to stick with the law.
David: Agreed. It’s quite funny in a lot of ways how we bumbled through the early stages. But it was a great experience and we learned a lot. We still do.
Eliot: I think we’d recommend much of what we’ve already covered in the earlier questions.
David: Yes I’d say key things we’d recommend are the following;
- Do your market research. In hindsight, we did some, but could have done more. We didn’t really appreciate the extent and variety of incumbents when we began. I think had we done so we might have been dissuaded from entering this niche. Equally, don’t spend your time behind a screen doing research in a vacuum. Get out there and get feedback from potential users, and not just friends and family. Show them similar products or prototypes and refine your assumptions and overall hypothesis early and often.
- Find the right type of financing. As we said, approach funds already aligned to the type of idea you have, and often in the legaltech space, this might mean funds invested in boring start-ups you’ve never heard of. Don’t chase funds because of their prestige, or the prestige of their portfolio start-ups. If you can find an investor who is also an advisor, do so.
- Invest in your first technical hire. Getting the right technical hire on the ground will be the keystone to your success building out a platform and a solid start-up team. Without this, you may waste a lot of time and money and be easily misled down the wrong path technically, which may be fine at first but fail to scale later on and / or store up technical debt that puts a brake on growth.
Lawtomated: What tools does the Avvoka team use to stay productive as a start-up team and build amazing software?
David: At Avvoka we use the following:
- Zoom: For calls and VC. We had Zoom before it was cool!
- Slack: For everything, and even more, valuable in lockdown. The ease with which we can interact is great. It’s done away with a lot of email.
- Asana: For task mgmt. Takes some getting used to, but setting up right makes good sense.
- Hubspot: For CRM; so much power you don’t know existed. Fantastic for automating sales and marketing processes.
- Google Meet: For internal stuff. Quick, easy and Slack integrations work.
- Google Workplace (formerly Gsuite): For everything. Drive, GDocs, GSheets, GSlides.
- Loom: Absolutely integral. Easy way to record your screen. Immediately saves as a link for sharing. Used for demos, training, support and bug fixes. 100s a day. Saves a ton of time.
- GitLab: For tech stuff, e.g. sharing and maintaining our code collaboratively and remotely.
- Intercom: Customer support chatbot. People can have a live conversation, suggests relevant help and can triage to a real person if needed. It has helped us scale support.
And finally, do you recommend any start-up or business-related media?
Yes, absolutely. We’d highly recommend anyone keen on start-ups, or with one of their own, to check out How I Built This podcasts (Apple & Android). It has hours of interviews with start-up founders, including many well-known names. They are a great source of ideas and, perhaps most importantly, make it clear that even the most successful start-ups are often “overnight successes” several years in the making! There are many great lessons on pivoting, financing, marketing and the general grit to make a start-up successful.
What is Avvoka?
Avvoka is a document automation, negotiation and analytics tool designed to help law firms, in-house legal teams and businesses draft documents, negotiate them, and leverage data insights from that process to draft better documents, and get to “yes” faster.
Unlike legacy tools, with Avvoka you can rapidly build automated versions of your documents using our intuitive automation builder. This means no more complex coding in Word documents. Our customers reach agreement fast by collaborating on documents with colleagues in real-time and negotiating with counterparties via the Avvoka platform.
Avvoka has a range of sophisticated reporting tools that allow businesses to monitor push-back against their standard contract clauses, track key commercial terms and compare individual negotiator performance.
For more insights into Avvoka, please check out their tips for successful document automation here, their fantastic ultimate guide to document automation or their part in open source contracting here.