03:33 - VC Landscape and AI Proliferation
04:34 - Structural Changes in Business Building
09:36 - Importance of Domain Experience and Taste
24:37 - Power of Design
33:00 - Customer Obsession in Founders
36:23 - Blind Spots in Design Thinking
43:42 - Balance in Product Design
46:24 - Changing Perspectives in VC

Oliver Darwin:
I think like, yeah, people are shipping very quickly, and like I think you need to do that in AI, otherwise you get left behind, because this is moving so incredibly fast. But I think because it's so easy to build something, people are just building something versus being more discerning about what they've built and the way they've built.

UntilNow Team:
We're a Sydney-based design innovation studio working specifically with startups and scale-ups to build and transform brands and products. In this episode, we talked to Oliver Darwin from OIF Ventures. We spoke to Oliver about how the VC landscape is changing with AI, what he looks for in founders, and the role of design in success for startups. We absolutely had a blast recording this episode, and we hope you love listening to it. Let's get into it.

Megha Sevekari:
Hi Oliver, thank you so much for coming on.

Oliver Darwin:
All right, thanks for having me. Love to start off with how you got into the VC world, a bit of a big question. Yeah. Now that I'm here, it feels like the right place to have ended up, but I can't sit here and profess like it was a master plan type. I wanted to be here and that's where I am. I suppose I was always quite a curious and entrepreneurial minded sort of kid growing up. But being in Brisbane, there was just not a lot of exposure to the startup ecosystem or no clear pathway to get into an entrepreneurial pursuit. So through school and uni, I kind of always just knew I wanted to be in the business world somewhere. Ended up doing a commerce law degree and ended up going to investment banking. Pretty quickly realized after being in investment banking that that was not the long term play. I think banking gets really bad wrapped. There's a lot that I really liked about banking. You get to think super analytically. You get to learn about companies, understand how to pitch their value proposition to potential buyers, figure out what's possible in the future for those businesses. But yeah, you don't get to build a lot of deep relationships with the companies you work with as a junior. You're often pulling cost levers versus revenue levers and the optimizations are not, how do we make this business grow super fast and become a market leader? More, yeah, how do we strip cost out and... you know, they're very mature businesses, very different kind of profile to what you work with startups. So I started thinking, okay, what will let me do the parts of banking that I like being the analytical thinking the, you do a lot of presentation of your analysis, both numerically and visually in the investment banking world. So you actually do learn a bit of like design skills and how to communicate stories visually, which probably not the same as what until now is doing, but a certain level to which I really enjoyed. But I wanted to still do that and figure out how I work with sort of business owners more closely and think about, you what's possible on the top line versus the cost line. So it landed on VC and then just like honestly dumb luck timing friend of mine from banking who was in the OIF network referred me for a role there and. Yeah, the rest is history. I haven't looked back. Yeah, awesome.

Megha Sevekari:
How is the VC landscape changing with the advent and proliferation of AI?

Oliver Darwin:
It's changing a lot. It's making our job a lot harder. It also, I think, makes our job a lot more interesting because there's a lot more thinking to be done. And it's also impacting the different funds in different ways. In the sense of like we're a very high conviction fund. We have a high conviction, high concentration portfolio. And we're investing out of $140 million fund. We write two to $5 million checks. We only have so many bets we can take. Whereas if you look at someone in the Australian market like Blackbird who's investing out of a $1 billion fund, they very much play the power law. They go super early. The way they are able to think about their AI thesis is probably very different to the way we're thinking about it. So it's shifting the landscape a lot for everyone but it's also shifting it differently which then kind of makes a compounding effect of how it's impacting the market. It also changes things structurally in terms of the way businesses are thinking about building. And by that I mean... you know, the cost to build is so much lower nowadays. The cost to get to a V 0 is much lower. you know, engineers can be far more efficient, engineers in particular, but everyone can be far more efficient now in kind of the AI age, even if you're just utilizing basic chat GPT. So, you know, you'll see founders who will go and raise a seed round that'll probably get them much further than it previously would. And, you know, it kind of opens up questions of what happens to the series A market, what happens to the series B market even if you can get to a certain ARR milestone quick enough that you outrun your expenses and you're profitable from there and control your own destiny. So it hasn't fully played out yet, but there's questions around like is there gonna be more seed strapping, I think it's been termed, where you raise one round and kind of that gets you all the way you need to go. Again with the cost to entry lowering. I start asking questions around what do market structures look like. It's so much easier to build a thing now. An example is there's all these note takers, whether they're horizontal or vertical. Probably relatively easy to stand a horizontal note taker up and if you've got good distribution, your product doesn't suck, people are gonna keep using it out of complacency and does the market end up having multiple of the same thing doing five to 10 mil of ARR and no one gets to scale or do those ones kind of fizzle out and the market leader, whoever's built the best product or has the best distribution ends up mopping them all up? Which is what historically has happened, right? Yeah, exactly. But now the costs to kind of build and maintain these things is much lower. you know, often what happens is sort of competition enters market, pricing becomes a thing and sort of you lose pricing power and your margin is compressing and ultimately you can't function anymore. It remains an open question to me whether that happens now with AI. I think it probably does, but it probably happens slower. And it just makes, I guess, survival more important. again, I don't have the answers to these questions, but it does start opening up a lot of questions structurally about how the market actually looks. Seeing a lot of people now thinking about how do I go into an old world business like a... you know, agency of some description or, you know, even something like vets or whatever, something that historically has not had a whole lot of technology implemented. And whether it's AI or otherwise, I think this AI revolution has gotten people thinking about how to put technology into our businesses, whether it's AI or whether it's just more simple, like classic automation. So you've got a lot of really talented, hungry people who are starting to look at that play, like the services roll up and... You know, it's, not sure if VC's the vehicle to fund those businesses, but that's where lot of the talent's heading these days as well, and VC's historically wanna back the best talent, and a lot of the best talent are looking at those opportunities, because there's a lot of money to be made, I think, in that space, if you do it right.

Megha Sevekari:
Can you talk more about this, because I know, what is service?

Oliver Darwin:
I'll use an example of a business that we're invested in, a business called Visory. And they've been going since before the AI days, but essentially they sell bookkeeping services to small businesses. And the reason that works is because bookkeeping is a highly fragmented industry with typically pretty low NPS. And most bookkeepers are sole proprietors, so they don't want to go out, or they don't know how to go out and win their own work. If you can go and inject technology into that industry, which is what Vizor has done, they've built a platform which sits in the middle of the relationship between the SME and the bookkeeper. They can then go and acquire customers using a software style acquisition model, a SaaS style acquisition model. They've got an SDR team which hums, they cold call, they sit demos, they acquire customers, they charge a manual license fee. And they've pulled in the supply of the bookkeepers or the humans in the loop and they've built a platform which sits in the middle which has a bunch of AI in the middle which can essentially deliver the service at a much higher margin. I think people are kind of cottoning on with this AI revolution that there's a lot that can be done within those businesses. If you're able to get your head around the way they work.

Megha Sevekari:
And do you think that changes like you said smart and talented but do you think that's changed what smart looks like? That the way they're thinking about what a business can transform into is also really important in terms of what you're looking for in founders or what their vision is and how they're utilizing AI.

Oliver Darwin:
Yeah, I mean to bring it back to software, think one thing that's always been incredibly important and particularly to us, every fund is gonna say this and it's probably true for most funds, but we're very founder led. Founder is always the north star of our investment. It's always the highest weighting in terms of how we think about investment criteria. But. I that again, now that the cost to build technology is, well least the V0 is going to zero, the founder and their skill set becomes so much more important. And I think like two of the key things that have become really evident is, and they're somewhat related when it comes to building product, is having the right domain experience and having taste. What do you mean by that? Like, I mean, taste is very hard to quantify inherently, right? That's kind of exactly the point of taste. You've got to have taste to know what taste is. Yeah, exactly. But I think like the way it manifests in product design is creating the thing is the easy part now. It's how you create the thing or even should you create the thing. Yeah. I think like... If you're a founder that has taste, you will naturally also exercise discretion as to what you build and how you build it. And having that paired with the right level of domain expertise, I think, is incredibly critical. I mean, just like before AI, software has always been a best-in-class opinionated. Version of a workflow that manifests itself through UI UX. It's deployed on the cloud and blah blah blah. But like, that's what it's always been. I don't think that changes with AI. I think people have gotten a little bit ahead of themselves in terms of like, everything's changing because of AI. It's like I think they're not. Stepping back and realizing like why people have always bought software like Software has been always been bought to remove the ambiguity of how to do a thing and I think like a lot of these Horizontal AI like agent players for example come in and they just introduce more ambiguity into the mix particularly with the unpredictability of like what generative spits out so I think when you have a really strong domain expertise and great taste You can very intimately understand the workflow that needs to be mapped And in building a product to match that workflow, you know where to just use your discretion and stick with the deterministic sort of element of the product build and where to actually inject generative AI for like increased like, just kind of like amplify. Yeah, exactly. Because I mean, there's like things that really matter where you can't get it wrong and you need to have a deterministic output. Like in a compliance context, you can't just have something that's. You know, using a large language model to generate an output unless you have really good evals, which again speak to taste because like you need to know the workflow to know what it looks like. Just sort of set parameters around what the model actually generates. You can't do all of that without having great taste and the right domain expertise to actually intimately understand that workflow that you're trying to build a product around.

Megha Sevekari:
Yeah, that's a really great insight because I think when you're not a regular user in that space, then you don't know what good looks like and that can be really dangerous territory.

Oliver Darwin:
Yeah, 100%. It's like when you ask Chachi Bichay about something that is not your domain of expertise, you're like, this is awesome. Yeah. And then you ask it to do something that you actually know quite a bit about and you're like, hell, This is absolute rubbish. I think people often forget that the stuff that it's telling you about your not area of expertise is probably rubbish as well to some extent. And again, that's where the taste comes in. It's like, what do I think is useful out of this output and what isn't useful in being able to extract that using it as like the companion versus the actual sort of be all and end all.

Megha Sevekari:
And I think at the start when you were talking about like AI muddying the water in terms of how I guess the volume of what people can make, do you think that speaks to it as well where if someone is trying to plug AI into everything but then that... what I'm looking for. Like that disconnect can then just lead to a lot of product with not a lot of conviction.

Oliver Darwin:
Yeah, 100%. I think there's a lot of solutions looking for problems out there. It's pretty fair. I wasn't in VC during the crypto wave, but I think it's probably very similar. AI is awesome. It's so cool. I have no idea where this technology takes us. I know that it's going to make a lot of impact in the world and it's going to change a lot of things, but... Yeah, you've just I think you've got a lot of people with solutions looking for problems and that comes down to the taste thing of like, does this need to be AI? Probably not like you guys probably know the real room guys. I talk to Dan often and without getting too far under the hood of their product, but like why I think their product is so exceptional and is an incredible generative AI product is because they, well first of all, they're designers. They understand the workflow so incredibly well. And secondly, they've really smartly fed in. Honestly like a very large percentage of what they're doing is determinative. They've also got a great piece of data and they've built all the building blocks already for the websites, but they use a lot of determinative logic and then generally I just amplifies what they're already really, really freaking good at. So like, yeah, you see a lot of solutions looking for problems, but unless the problem actually exists, people aren't gonna pick it up and use it. That's the key thing there.

Megha Sevekari:
How does this insight manifest into what you look for in founders when you're going through your processes? Or has your process evolved at all? Have you used AI?

Oliver Darwin:
Definitely using AI. Still figuring out the best ways to use it. There are few tools that are coming up for VC to go through. Yeah. The amount of... I'm probably more skeptical than a lot on those. I'm still very hesitant to outsource or to feel like I'm outsourcing thinking. You can't outsource conviction. So the way I'm using it is kind of using it as a thought partner or kind of just like a... you know, a strong critic of my opinions and just really stress test my thinking, help uncover points I might be missing and stuff like that. But yeah, I think for particularly some of the pre-seed funds that have like incredibly high volume, they are drinking from a fire hose, particularly now with the amount of AI sort of businesses that are coming up. So I can see how some of those triaging tools are super useful for them. And there's obviously a bunch of sourcing stuff that can be done with AI that is super helpful. But yeah, I... maybe I'll be left behind, but I'm a little bit skeptical in some of the outsourcing of the thinking elements of it all. to be philosophical about it for a moment, like, it does kind of make me ask questions about where we're going and, you know, when everything can be done for you, like, why bother with the effort? And I think, I'm very big on, like, the process is much more important than the outcome. So for me, I am... very conscious about what parts of the process I am outsourcing to AI. But this is in no way answering your question about what we're looking for in founders nowadays. again, think it all comes back, I'm probably gonna talk to this a lot, but it comes back to domain expertise and taste. When you talk to the founder and you ask them these questions about how are you thinking about this problem, how are you thinking about why or why not the incumbent in your space would build this, just being able to get comfortable that they have thought deeply about this and they haven't just outsourced the thinking or taken the convenient option. I I listened to the podcast you did with Nick, who's one of our portfolio companies, Clara, and. I really liked hearing him say, how's the product gonna look in five years? I don't know. Being able to admit what you don't know, but having a theory, well don't know, but I think it's gonna look like this because of ABC reasons. That's incredibly powerful because to be able to understand what you don't know actually is an incredibly good indicator of your depth of understanding of your domain.

Megha Sevekari:
Yeah, absolutely. I think that's really interesting because it sounds like then AI really hasn't changed much about how you're looking at founders? Like those two principles seem like pretty consistent?

Oliver Darwin:
Yeah, I think we have always tried to back great founders solving big problems with the appropriate technologies to do so. Yeah, we've invested in businesses in the past couple of years that literally don't use AI at all. Yeah. It will in that sort of outward product manifestations that they're using internally for a bunch of things, as I think most people are. But... You know, we've got a company that helps councils optimize their payment experience. They don't need to use AI. They shouldn't use AI. In fact, the core principle of what they do and what allows them to do what they do so well is that they have such strong trust within the government space and with the ratepayers who the governments use their products to collect payments from. And the second you start to erode that trust by introducing uncertainty into the mix, like you lose your competitive positioning. Yeah, it's almost just like sharpened our focus, I would say, in terms of we've always been very founder led. You know, you definitely, particularly in the good times, which I caught the tail end of when I joined VC, but you could definitely get caught up of this business has awesome metrics on paper. I don't really understand the problem they're solving, but clearly people are paying for it, which is still valid. Founders, okay, the metrics are great and cool, the market's big, let's go. You saw a lot of rounds get done for those kinds of businesses and there's a lot of private unicorns that are no longer growing and will never raise capital at the same valuations they raised at. So I think it's like just sharpened our focus in terms of really drilling down on what's important for us to get comfortable with when we're meeting founders and backing businesses. We're post revenue investors, so we still need to see all of that sort of stuff on paper, which is again, very muddied by AI because there's a lot of experimentation going on in sort of the world of AI products. I think like that combined with the... how much success LinkedIn has had over the past five years and the kind of mass migration of the tech scene onto Twitter or ex-formerly known as Twitter. You're starting to see B2B products go viral in a way that they never previously did, like Salesforce would never have gone viral in years past, but you'll now see someone build an AI product and because AI is sick, It can do like the demos you can build with AI when you control the environment are exceptional. Whether that translates into like a product reality is a different question a lot of the time, but the demos look awesome. And you've now got like so much activity in the tech world on like large social platforms. B2B products go viral now, which is just like not something you've really ever had to deal with before. So you see these businesses that They shoot the lights out in sort of 18 months. They go from zero to five mil in such a short amount of time. Cursor is a great example and that's probably more mature than a lot of the ones you see. But we see a lot of businesses that have gotten to two or five mil of ARR really quickly and they haven't had enough time to even go through a real churn cycle. Digging deeper into the quality of the revenue is really important. then like, you know, to the extent you can't dig deeper into the quality of the revenue somehow, because again, you just almost have to make a gut feel. It just requires you to think really deeply about is this product the right product? You have to think so much more deeply about like the way these products position themselves from both a sort of distribution and like just usability perspective. Because the revenue is much more difficult to pick through these days than it used to be. It's almost like consumer-esque revenue. So it sounds like you look at analytics, you look at the revenue, you look at obviously how the business is going, but there is a lot of attention on how the experience is delivered to the users. How do do it? Because obviously sometimes you are the user of that product and it's easier for you to relate with the problem and make a judgment on it. Sometimes you're not. Yeah, it's quite difficult. And this is where the quality of the revenue being more mixed than it used to be makes things even more difficult because historically as post-revenue, well we term ourselves as post-revenue, post-product market fit investors. Usually you can say, well, I'm not a domain expert. I don't really know this space that well. But you've got 10 customers who've each agreed to pay you $100,000 a year for your product. So that gives me, and I'll go and talk to three or four of those as part of my due diligence. We do talk to them. Yeah, we'll always talk to customers. I can go and talk to those customers and at least get some level of comfort that, well, you've got a handful of people who are in the space who are willing to pay you a substantial amount of money that makes sense from a union economics perspective. And B, I've spoken to those people and got the qualitative feedback that gives me comfort. But. It doesn't make it easy, particularly now as that revenue quality starts to potentially degrade. You get more consumer-esque revenue profiles where churn's much higher, retention's really low, upsell's difficult. You're acquiring customers and they don't stick around, so your LTV's no good anymore. I think I'm a big believer in the power of design. And I think it's just becoming more more important because in the same vein of how B2B products are now going viral, I think there's like this mass. Consumerization of B2B product consumption and like design has always been so critical in consumer products and you don't see a lot of them in Australia because I don't think we have the market size to support the virality that you need necessarily but you know the faster you can get someone from the start of an onboarding flow to their first point of value in a consumer product is essentially make or break for that business you can't exist unless you do that well as a consumer business. And I think you're starting to see that kind of pattern shift into B2B product usage as well now, particularly with, you know, like in the TikTok age where like attention spans are just like cratering, know, products are now going viral. People just pick up, put down products and churn and whatnot. Like the ability to have a truly wonderful design or user experience just becomes... you know, even more critical. What is a product that you experienced that recently that reflects that? Superhuman is a great example. Yeah. Did churn because their calendar functionality is not good enough yet from like an actual product functionality, but from a user experience. It's pretty good. It's incredible and a point I was gonna make as well is you need to be able to design these user experiences in such a way that they're just like magical is the right word and just so intuitive and self-serve. We trial tools all the time in the business and the most frustrating thing is when their sales reps try and book you in for a 30 minute call to get your feedback on how it's going. It's like, I just want to make sure you know how to use it. And I get that they're doing their job, but. If you can give me a product experience like Superhuman where that onboarding is just like seamlessly embedded in you using the product, give me that every day of the week. mean, ReLume's another great example as well. the product that... that team is built. They're not a portfolio company. I just love the guys in the business, so free plugs. I had the same feeling with Granola. Have you used Granola? I haven't. I've been wanting to use Granola. And this is another thing about AI and how it's muddying the waters is we have a native... AI note taker built into our CRM. And it's actually pretty good, to be fair. So as a result of that, I haven't needed to try granola. I actually do really want to try it because I have heard great things, but that speaks to how good the design experience must be. So there's so many AI note takers out there, yet the only one that I consistently see people evangelizing is granola.

Megha Sevekari:
Earlier you kind of talked about virality leading to people picking up a product, but then putting down. So the discussion makes me feel as though the putting down is happening because the UX isn't great, UX UI isn't great. Are there any other reasons why that trend is happening so frequently?

Oliver Darwin:
I think the key thing is that the UIUX isn't where it needs to be. It's either that people aren't actually the right customer. Like because it's so easy to pick these things up and test them these days. You just want to see what it's all about. Like I do it all the time. I'm like I'm part of the problem. I just see a cool tool online and I'm totally not ICP and I'm just like, that looks sick. let me see what that's all about because it might be relevant to a portfolio company or a company I look at or meet next week or whatever it is. So think there's a bunch of values it's really easy to do these days so a lot of people will just pick things up to try it because it looks cool. But I do definitely think like there's a lot of people who are leaking ICP customers because they haven't gotten the user experience right. And that's like pretty damaging because first impressions matter a lot. You know like if you go into superhuman and you don't have a good user experience you turn off. Like it's very unlikely you're gonna, in a six month's time, someone's gonna say, super human's awesome, and you just think, I tried that already, it actually wasn't great. It doesn't matter what they say, no, no, like the product roadmaps come a long way, like they build X, Y, Z. Unless it like really resonates, you're probably not gonna pick it back up. So I think like, yeah, people are shipping very quickly, and like I think you need to do that in AI, otherwise you get left behind, because this is moving so incredibly fast. But. Think because it's so easy to build something, people are just building something versus being more discerning about what they've built and the way they've built it. You know, like, it's like to go back to the first principles of like how you should be building the product or like why you should build a product in a certain way. Like I know a portfolio company of ours that you've worked with, Vouch. You know, they've essentially like made doing short form video testimonials seamless. That's like not an easy thing to do. To get someone to pick up a phone, film themselves, answer a bunch of questions all in the flow. Like that's a huge piece of admin that would probably... you know, take people like hours to even build up the mental energy to decide I'm gonna take that task on and then probably take at least an hour to do. And, you know, the way that... sort of they and with you were able to design that user experience, not just that it was super seamless for the person to do, but the inbuilt virality of that product of, now when you're the person viewing the short form video testimonial, you immediately can see that it's vouched. I super humans done the same thing. You can't miss it when you get an email from someone who's using super human, right? So, like it's the usage and it's also like building distribution into the product as well, which. I think a lot of people don't think about so much as well. mean, it depends on the kind of product, right? You're not gonna build distribution into a heavy sales cycle enterprise business, right? again, it's like about knowing, comes back to the domain expertise of meeting your users where they want to be. I think a lot of AI products have kind of forgotten that people buy software because they don't know what to do, don't know how to do something. So like there needs to be a human in the loop somewhere. There needs to be some taste. I think you're right on that. 100%. Like a lot of people, you know, again, to go back to the visor example, like they've built really great AI functionality within the platform. But. You know, they're not gonna say the bookkeepers, hey, you guys can go now, we've built, yeah, I can do what you do, because inherently, like, people wanna trust something, and I don't think we're ready to trust AI fully yet. It's like when a self-driving car has an accident, even if the incidence of that is, like, so much lower than human-caused accidents, we don't actually care, because we just wanna blame the technology and say, well, this isn't good enough, but. We have much more empathy when it's a human who crashes because we can say, it might have crashed because they were having a bad day and they were distracted or whatever. There's this level of empathy that I don't think people have kind of like begged into the way they think about building AI products.

Megha Sevekari:
Yeah, that's true. In my mind, kind of going back to what we were talking about when it started. It's good if lots of people are using it because then you have the feedback for what design should be looking like, or you have people saying that I want this or I don't want that, and you know why people are falling off. But in that process, if you're losing your ICPs, then you're right, they're never gonna wanna come back. So it's like a byproduct of... having a lot of testers is that you might actually lose someone that you want to retain.

Oliver Darwin:
Yeah, I think that speaks to like a really important trait for founders though is like customer obsession because I mean, you tell me but like when you turn off a product how often you actually giving the feedback of why you turned off, like you're not right? So you need to have a founder that's like customer obsessed and if someone turns off or if someone's using the product, they wanna, you you wanna know that they're trying to find out why, like what can they do to fix that because most people aren't gonna bother, right? Like when I turn off a product, very rarely am I like giving, know, waxing lyrical about, well, I like this part, but this part was no good, and like maybe you should do this, and you know, like sometimes I do, because I'm in the tech world, and I'm just like a weird about it, I'm like, I'll try and help these people out, but like most people aren't like that, right? Like I think we often also, and like I think a lot of people that are building often... forget how close to the coal face we are. Being people who work with tech founders of the earliest stage possible designing their products. In my job, I obviously meet tech founders building all kinds of things. We're always so close to the coal face and we forget, I think, the way that people outside of this bubble think about and consume products full stop, let alone AI products. They do not care how. The thing does what they need it to do. As long as it does. Just have a problem. They could not care less, right? So yeah, think a lot of people have gotten very excited and I get it because people are pushing it down from board level and saying, what are we doing about AI? Yeah, just solve the user's problem. Yeah. Yeah. I'm 100 % guilty of that. I feel like I've only been introduced to how actually things are built, like apps and products are built, like five, six months ago. And I'm like, I did not realize this amount of work goes into it. And I'm way more likely now to provide feedback than I ever was. But nor should you. 99 % of the world is in that camp. they just want... something to remove a piece of friction in their life or improve it in some way. They do not care what piece of technology is doing that for them. AR is great because it deals with a lot of real life situations very well because it's great with unstructured data. So it can do a lot of these things that previously couldn't be done, but at the end of the day, I don't care what tech you're using as long as the experience is easy to consume and that's where the design piece comes in. if you can design your way around needing a power user in an organization, that's the holy grail. Because the internal adoption is so huge for stickiness and expansion for these businesses. So you can design the product in such a way that kind of allows for that to happen naturally without a customer success rep having to put a call in your calendar every second week and be like, hey, do know this feature exists? like. Leave me alone. People don't have time, they do have time, but they're not gonna make the time. It's human nature, you just want something to be as frictionless as you possibly can. That's always been the case in consumer, it's starting to be the way in B2B.

Megha Sevekari:
Are there any blind spots that stand out to you when it comes to founders thinking about design?

Oliver Darwin:
Yeah. Perhaps it's not a blind spot, but. I often it's hard to make it the priority that it potentially should be. And it's like a balance, because I think at different points of the journey, it should be lower priority versus higher priority. But because it's such a difficult thing to quantify, as an investor, it's very hard to say. You need to change your product because of XYZ or your website is no good. You should change it to be like this. Like I don't have the data on that. That's obviously like platforms and stuff you can use to get that data and analyze, you know, where you're getting drop off and stuff like that. But we obviously don't have access to that. But then it's also very difficult for them to think about how that tangibly changes because like you can measure it before and after, but to get the after. Data you need to have invested, I would say, relatively high amount of time and effort and money. So because it's not directly revenue generating, i.e. if I spend an hour on sales calls, can I generate X dollars of revenue? Probably yes. Or can I do something that I would otherwise have to hire a person to do, i.e. save money? It's very difficult for them to go. I really need to like spend all my time and effort on like making sure this design is perfect. So it's like how do you strike that balance of like ignoring it when it's not important but like giving it the right time and attention that it deserves when like you really do need to focus in on that. I think it's like very overlooked. I think founders often miss like how tightly it can be linked to distribution and that that goes to like I think when people think about design a lot of the time they just think it's like that's my color palette and my sort of my typeface and whatnot without realizing that like It's very well, at least I think anyway, it's quite tightly links to your distribution And the way you tell your story So many websites or startups are so bad And like, you know, I'm out here trying to introduce founders to potential customers and you know, they don't even have so much as like a paragraph spiel of what they do and like communicate their value proposition, which I know is not like directly related to design, but that's definitely something that I imagine you are crafting when you're doing the design work, like understanding your value proposition. So you can then subsequently communicate that via your design. And then it's like, you don't, in the absence of that, you say, here's the website. And the website just like, it's two years old, it doesn't say what they do. And sometimes that's because it's a champagne problem and they're selling so much, they don't have time to go and update the website. But a lot of the time, I think like they're missing opportunities. That's like kind of your digital front door, right? Like if you don't get that right, it's hard to help you in a way. And like, One of the sort of lessons I took away from my investment banking time was, like, if it looks good, inherently people will trust it and think it's good. You know, like if you make the deck look pretty, as stupas that is, people will ask less questions about the numbers. you can, you can like almost like, you know, not hide, because you're not trying to hide the numbers, the numbers are right when we've done the deck, but you know, you just end up going down less rabbit holes and you can kind of divert attention to what's actually important. And it's the same thing with a startup, it's a website, whether it's a deck, whether it's a product, through the way you can visually tell the story, you can push people in the directions you want to push them. Versus letting them make their own decisions or have their own takeaways based on the raw data that's presented to them. Ultimately, when you're building a deck or a website or even a product, you actually want to force your... know, read our user, whoever it is, into a conclusion that you've predetermined in a way.

Megha Sevekari:
Yeah. So I was gonna say, I'm definitely guilty of associating like website quality with how much I trust. A company or how established they are. So agreed with if I do hear something like, maybe they're just not as far along where as they could be and they just haven't expressed that.

Oliver Darwin:
Yeah, and it speaks to, think it broadly speaks to taste as well, right? If you have designed what, and everyone has different tastes, right? But if you have designed or built. What I perceive to be a really strong website. Like I am inherently gonna think your product is probably pretty good too. I'm gonna extend that so I think you make good decisions elsewhere. it's like principle of like if you do the small things right, like you do everything right. I think the hardest thing for founders is to decide when design is important or what thing is important. I think, you know, each founder has their own skill set and they have, know, sometimes they're more technical, sometimes they're subject matter expert about something. Sometimes they are designers. What I find is like, founders, are designers. They know when to deploy their design skills better. And they don't over design things, which is quite interesting. If you're a designer, want to over design things. Potentially right, but they don't. And they actually very carefully decide when to design something versus when, this is just good enough for now. So we should move on and focus on something else. I think that's the missing part. Yeah, I think that goes back to taste right? Knowing what level of design to inject is being a good designer right? what? What needs to be designed to this extent and what versus what, can we just sort of like, this is good enough. I think like, Vroom is a very good example, like that both designers, like Linear is a very good example. There are a lot of like, designers, founders that they can tell. They've done a lot of work on the website, a lot of work on certain details, but then certain things are a little bit more loose because they're not super certain that's their approach to go with. But then they will go back to it when it's right time. Yeah, and to really as well, again, to being user obsessed. They've got such great community. I know that they listen to them. They do better testing really well. They just iterate fast and they iterate smart. They're very clever. They're also like the things they do on YouTube to do the, they're very clever. They're very clever. I'm a big fan. A lot to learn there.

Megha Sevekari:
Do you think it also, the balance changes depending on what your product is doing? what's... Problem it's addressing, whether it's a needs base where it's such a highly specialized product where you just have to use it no matter what, I'll just deal with it. Whereas if it's something that you're, it's adding to your life so you don't have to use it. In my mind, the second one requires good design a lot more than the first, the former one.

Oliver Darwin:
Yeah, 100%. I mean, there's products that... I guess it's all design in a way. There is products that are designs that you don't even see them. So like Pable, the company in our portfolio that helps governments optimize payment experience for the ratepayers. The government employees don't even see the product. It sits behind the ERP that they use to administer their rates. It essentially allows the ERP to talk to the payment gateways. But where they have applied design really intelligently is they have figured out like on a psychological level like when you get a text with a link Like what's gonna cause you to open that link because essentially what they're doing is putting an e-commerce Checkout experience on your property That's that's the simplest way to put it that removes a lot of admin for the government's but it also makes it easier for ratepayers to manage and pay their rates and installments and overtime To do that, to put in your common experience, you need to send a payment link. Particularly in this day and age, I don't know about you guys, delete most texts I get that have a link in them. So how have they designed, it's a text message, it's nothing like what do think about designing a website or a product, it's like how do we design? The user experience of getting that text message to build the maximum amount of trust that we possibly can. It's totally designed though. Exactly. You want to get an output and how do you get to that output? Could be language, could be visual, could be whatever media you have. So again, you probably wouldn't immediately jump to the fact that's a very nuanced design process they have to go through to figure out. Well, this product doesn't work if we only get 20 % click through in our links, but it absolutely works if we get 80 % click through, right? So how do we design an experience that has a meaningfully higher click through rate than any other text message you would get with a link in it? So it just expands so broadly, I think it ultimately relates back to first principles thinking. It's like, do want, how do I get there? And that's fundamentally what you're doing in everything that you do as a startup.

Megha Sevekari:
Yeah, awesome. last question for me, So you've been, I think you said at the start, in VC for about three years now. I'm kind of really keen to understand what's something that you've changed your mind about from when you first started to now.

Oliver Darwin:
I think that's probably two things. When I first came in, and they're actually kind of almost in direct opposition. Let's see if this makes sense. When I first joined VC or came into the startup ecosystem, coming from outside in, I kind of just thought, well, the best product is the one that will win. You know, I use an iPhone because it's what I perceive to be the best phone on the market. Like you guys use MacBooks for the same reason, we're using these Rode mics because they're great. And they probably are in the top 10%, 20 % of products on the market, right? But the amount of work to get the distribution right that allows them, this phone to be in my hand or this microphone to be in front of me is like. That's what makes the difference you need to have a good product, but it's very rare I think the truly best product on the market will actually win and like that's something that I had to kind of come to terms with joining the industry The second thing that I think like I've changed my mind on is that You can't worry about who else is out there You know you come in, you meet this business and you think like, well, just bid this or Salesforce will build this or Shopify will build this and you really have to let that go. And it's kind of difficult, particularly coming from like an industry where you deal with mature businesses like investment banking. You've seen how dominant incumbents can be and how valuable those businesses can be even if they grow. Five, 10 % year on year. They just generate a ton of cash and they can essentially just do nothing and stay massive. You come into a world where it's like you almost have to accept the reality that those businesses are destined to fail because they don't innovate well. that's, whether it's something I've changed my mind on, something I've more come to terms with and like. You can ask any of my team, like I am the least receptive to comments around how is this differentiated or like, it's such a competitive market because it's like, honestly, like I've just come to believe that like execution and taste are the only things that matter. And if you have those two things. Then it doesn't matter what anyone else is doing. You kind of like make your own destiny if you are truly elite in those two skills.
Megha Sevekari:
That's great. Love that. Thank you so much. That was really, really great.
Oliver Darwin:
Thanks for having me. Thank you.
UntilNow Team:
It was a super, super interesting episode and we're really grateful for the insights. Thank you all so much for listening. Make sure you follow us on Spotify and LinkedIn to stay updated with new podcast releases and we'll catch you in the next one. Bye.

‍