June 5, 2023
Ravi Parikh is a 2x founder with over a decade of experience building and scaling B2B SaaS startups from the ground up. Ravi is currently the Co-Founder and CEO of Airplane, the developer platform for building internal tools. Prior to co-founding Airplane, Ravi co-founded Heap, a digital insights platform, in 2013 with a former classmate from Stanford. During his time at Heap, he led go-to-market efforts and helped scale the company to a $960M valuation and 350 employees. Before building Heap, Ravi started his career as a musician with several published tracks.
Julian:Hey everyone. Thank you so much for joining the Behind Company Lines podcast.Today we have Ravi Parikh, co-founder and CEO of Airplane, the developerplatform for building internal tools. Ravi is so excited to chat with you. Notonly I was, I was talking to you before. I have intimate experience with yourprevious company being that I, I had a couple roommates who worked for, yourprevious company Heap, and really just excited to kind of dive into your mindon how, you, you think about enabling teams and really just offering toolsthat, don't necessarily replace what people do, but really optimize and evenkind of add additional value to, and we'll go into what, you know what I meanby that in a second here.
But before we getinto Airplane, what you're working on currently, What were you doing before youstarted the company?
Ravi:Yeah, for sure. Yeah. Thanks so much for having me on today. So yeah, before Istarted Airplane, you alluded to this I previously co-founded a company calledHeap. So, Heap is an analytics tool that basically helps people measure userbehavior on websites, mobile apps, things like that.
So, I co-foundedthat company back in 2013 with a friend of mine from college. Both of us hadbeen a couple years out of college at that point, but he had sort of had theidea for the product. And was kind of working on it and he sort of pitched meon the idea and was like, Hey, we should work on this together.
And I was like,yeah, that sounds great. And so, I did that for about seven and a half years.We started the company towards the beginning of 2013. Went through YCombinator, raised a few rounds of funding. Grew the company to, I think about200 people before I finally left the company in 2020.
And so Heap isstill around. It's I think roughly doubled in size since I left. So it's stilldoing quite well. But since then, I moved on and started a new company calledAirplane.
Julian:Yeah, it's so exciting. Thinking about kind of how the different experiencesbuilding companies. I'm sure it was back then going through kind of thetraditional ecosystem, right?
Like you, you gotplugged into like this operating engine and essentially helped you scale to, toa fairly sizeable amount. And then what's so different about now's environment,being that, funding landscapes changed there's more available tools andtechnologies. What, what, what else have you seen that's so different frombuilding company now, obviously experience, but what else?
Ravi:Yeah, I think it is really different doing this a second time around. Almosteverything feels different, but the core things feel the same. Like at the endof the day, you still have to find product market fit and build somethingreally excellent that solves a key problem for people and improve some aspectof their lives by like a huge margin.
Like those thingsare still the same. Achieving startup success is. Really, in some ways, not anyeasier or harder than it's ever been. You have to build a sort of outlierproduct, but the things surrounding it are super different. And so when westarted Heap in 2013 it was like, I think today in 2023 there is this sort ofreality that like tech is the most financially successful industry in theworld.
The, the biggestcompanies in the world are tech companies. Google, Facebook, whatnot. Yeah.That wasn't true 10 years ago. They were definitely big and they were up and,but they were still sort of seen as up and coming. I think probably the biggestcompany in the world and create their team is probably like ExxonMobil orsomething.
And so you, youstill sort of had a little bit of this narrative that. Tech was like an up andcoming industry in that it was a little bit more of an underdog industry andthat narrative was changing, I think when we started Heat, but it hadn't fullychanged yet. And so as a result, I think today when you start a company there,there's like a lot of like literature and advice and institutions and peoplewho will tell you how to do it, especially if you're doing something like B2Bsoftware where there has been so many successes over the last decade.
But 10 years agowhen you started Heap, it was like actually a little bit more unknown. And sowhen we started to keep, I, I, I didn't even know the word SaaS or B2B softwareor cloud, like, I didn't even know what those words meant when we started thecompany. We just thought, this is a cool problem to solve and this is a cool,tool that we can build and put out there.
And so, we juststarted building it without really a, a ton of clue about how to monetize it orscale it. I think, Matine, my co-founder at Heap, when he actually startedworking on it before I joined him on it, when he was like first hacking on it,I don't even think of, he, he thought of it as a company idea.
I think he thoughtlike, Hey, this would be a nifty app I could build to solve a personal painpoint that I've experienced. And then it kind of, he kind of pitched to somefriends and they're like, Hey, I would actually buy this if he turned this intoa product. And then, You bitch it to me and he's like, we should work on thistogether.
Maybe turn into acompany. And, and so like that evolution happened a little bit moreorganically. Whereas I think today people set out a bit more and say like, Hey,I have like a clear B2B startup idea. I'm gonna talk to a hundred customers,invalidated, and all that kind of stuff. So it's a little bit more structurednow.
I think that's thebig difference.
Julian:Well also thinking about the category you were kind of defining and, and, andmaybe people don't consider as much what I do in, in regards to how you'reenabling teams and essentially it's like, it's this, it's right as when teamswere starting to purchase tools for the use of, of the individual kind of user,like the product teams or if you think about sales teams getting, these, thesesuper robust tools to be able to actually execute these strategies.
It was, it wascompletely kind of nuanced. In the time you were doing it, I just remember allthese things coming about and thinking about how has it become like a standardto like enable your teams? Are we, did you see that adopting early on or isthat something that grew as, as you grew the company being that, people arestarting to actually purchase auxiliary tools for for their team?
Ravi:Yeah. It's a good question. And, and so basically like, yeah, so, so you'rekind of alluding to I think this concept of. Buying a tool to enable a team ofpeople to do something is kind of now seen as like a pretty common concept, butwasn't always that way. At the time I think like, I mean if you, if you rewindlike 30 years or 40 years back, way, way back before I was even in thisindustry or even old enough to be in this industry I think like the waypurchasing decisions were done around software were just really different thanthey're done today.
And you had somelike centralized IT department who would, who would try to like, Figure outwhat the entire company needed in terms of the IT stack and you would just likeuse whatever they bought and stuff like that. And then you evolved to thisthing where people could buy their own software and then now you have this sortof thing where people can sort of like enable their team with software andstuff like that.
So that hasdefinitely evolved over time. And I, and I, and I would say that like, both he,my previous company, my current company Airplane had sort of like, Flavors ofthat in, in terms of like the way we sort of approached the problems that wewere trying to solve. And a lot of products I think, have sort of mirroredthat.
I don't, I wouldn'tsay that we're unique in that respect in any, any way, shape or form, butyeah.
Julian:Yeah. What's interesting about, kind of building tools in the space that, I guessyou, you're not necessarily selling to the decision maker or adding value tothe decision maker directly, but you're doing it through.
Whether it'sconduits like developers who are building tools internally or in Heap Yeah.Product teams that are driving deliverable outcomes. How is that just, justoverall different and what does that change in terms of how you kind of,interact with your consumer?
Ravi:Yeah, for sure. So, yeah, I can, I can dive into that a little bit.
So Airplane is akind of unique product. It's kind of a interesting to sell product as youalluded to, because what we do is we're selling a platform with which you canbuild tools, right? So we sell to develop, we sell to developers, they buyAirplane, and then they build tools like an admin panel or a dashboard or some.
Billing console orsomething like that. And then they give it to a different team, usually like a,a support team or a sales team or a customer success team or someone like that.Then those people can use that tool to actually do their job. And so, like yousaid, sometimes the actual main stakeholder of that tool is not even the peoplewho are buying Airplane, right?
They're, they'resomeone like the, maybe the head of, head of solutions engineering or somethingat a company. And so, that is, is often. It depends a lot on the company. We'revery inbound heavy in terms of the way we sell right now. And so what thatmeans is like Norm, normally people are going to us and they already have somesort of like budgeted, project that they're gonna tie Airplanes to.
Yeah. They're,they're, they're coming to us and saying like, Hey, I'm, I'm on the engineeringteam and I'm getting pressure from our sales team to build this console thatdoes. X, Y, z, whatever it does we need to show some customer data and allow peopleto extend the free trial and whatever. And then, so I'm looking for solutionsto help fix that issue.
And it's already ona roadmap for this quarter. The Edge team is gonna spend some time on it thisquarter, Airplane looks like it can help us get there faster. So like some,that's, that's the best case scenario if someone comes to us with somethinglike that. Sometimes a little bit different though sometimes, someone will cometo us and say like, I think Airplane is really cool.
I'd like to be ableto use it to sort of build things for other people on my team or other peopleat the company. But we're gonna need their buy-in too. And then you gotta kindalike talk to them as well. And then you have to kind of prove the value oflike, Hey, if you had better tooling around this thing, things would run moreefficiently.
So it can be kindof a complicated sell sometimes. For sure.
Julian:Yeah. And, and what is kind of the relationship with. The, with like in regardsto why these companies are not purchasing other SaaS tools to be able to buildout a CRM or sales tool. Is it to create other products within their own, or isit to kind of replace other things that aren't being serviced by these SaaSproviders?
What, what are someof the use cases you're seeing?
Ravi:Yeah, it's a great question. I, I think, like, the way I think about it islike, if there's something that's common enough at every company, it's gonnaturn into a SaaS tool. But there's always some things that are specific to yourcompany, right? So you might buy Salesforce as your crm.
We're not gonnareplace that. You might buy NetSuite as your, e r P or whatever. We're notgonna replace that either. But even if you bought Salesforce, you see around,you're always gonna need other things around it. You're like, well, we boughtSalesforce, but you know, the way our company does.
Billing is reallyspecific. We have a really specific pricing model and all that stuff. So we'regonna build a really custom billing panel where you can sort of like go in andsort of see the data in a very specific way and it pulls data from ourproduction database in order to populate some charts and graphs or somethinglike that.
Like that's theexact sort of thing you might build in something like an Airplane. Versus goingto a dedicated SaaS provider. And every company has its stuff. And this islike, we didn't put this study together, but I've seen studies out there thatsay things like, as companies scale, especially at really big companies, up to likehalf the time that engineers spend writing code is writing code that is gonnabe for internal tools, right?
They're not evenbuilding customer facing software or new features. They're building these kindsof internal systems. They're building things that, so the sales team or thesupport team or the marketing team or whoever has some admin panel or consoleor workflow or something like that, that like makes some sort of internalprocess more efficient and companies build more and more and more of that asthey scale.
And so I thinkthat's what Airplane really tries to accelerate. Can we get those developershalf their time back, essentially?
Julian:Yeah. Yeah. And, and is part of this, kind of this huge movement, at least I'veseen, I run a company that's, recruits for, for startups and what one thingwe've seen is just this huge influx of low code, kind of no code types of toolsutilized by developers and, and, and from, from the, from face value from theplatform.
It seems similar inthe respect that you can kind of plug in almost like a white label service. Butcreate kind of a customized system for your team. What has, what advancement intechnology recently has been able to kind of increase that? Is it like, is it,is it like o o open kind of code platforms, being able to congregatecommunities?
What in particularhave you seen that has been able to grow?
Ravi:Yeah, I mean, you're right, there's a lot of like low-code, no-code platformsthat are trying to solve similar problems to, to Airplane. One example is likeZapier, right? Very different product. We don't compete with them at all, butit's like, okay, there's all kinds of people trying to connect Salesforce toSlack or, GitHub to their email or, whatever.
And so they canjust use to like kind of glue those things together and, and quickly create,workflows with, with a lot less code than have you do it internally. That's theexact sort of thing that like, used to probably take an engineer writing somecode and spending a couple days on the problem, now takes 10 minutes in theirplatform.
We're trying to dothe same sort of thing with other types of internal tools. I, I don't know thatthere's like any specific advance that sort of led to all this. I think there'sreally two things that have kind of happened in the last like 10 years thathave led to this. One is like, there has been an explosion of SaaS tools and soif you think about tools like Airplane and Saper and all the rest of theselow-code, no-code tools.
They kind of fillin the gaps that dedicated SAS can't fill in. And so 10 years ago when therewasn't a ton of SAS tools in some sense, like you were a lot less productive,but you also didn't have as many of these gaps that could be filled in bysoftware, right? And so I was describing like a, a billing dashboard that youmight give to your support team or your sales team or something.
That billingdashboard kind of probably lives alongside something like Salesforce, right?But if you don't have Salesforce at all, then you're probably doing everythingin a manual way anyways. And so building an internal tool separately from thatprobably isn't your first priority. So now that we've bootstrapped all theSaaS, I think there's now this bigger landscape for some of this internaltooling.
I think the otherbig thing is that. Almost every software product over the last decade hasbecome a cloud product. And so I don't just mean like, SaaS companies, but likeif you think about how do you watch movies, you watch them on Netflix, how doyou interact with your bank? You don't go to the branch, you log into chase.comor the app, right?
And so everything'sa cloud service, right? There's a client being served up to you via website ora mobile app, and then in the background there is some server farm somewherewith a bunch of backend software that's serving up that experience. And thenature of cloud software is, it needs sort of.
Intervention pointsthat are not the user facing software. So like, let's say you're logging intoyour bank account and you hit some problem, right? Like you notice that your Byou made a transfer but it didn't go through whatever. Right? Now you have tolike call up your bank and then when you call up your bank, you get Grepresentative.
That representativeneeds to be able to fix the issue. When they wanna fix the issue, they need beable to like pull up your record and be able to click a button to like, makeyour transfer go through or whatever. That software has to exist there to letthem do their job. This is the nature of all cloud software.
If you have a cloudservice operating, there's gonna be have to be some sort of team on the otherend sales support, customer success that serves that software and like fulfillsthose requests and they need software themselves, internal software to be ableto fulfill that. That wasn't the case 20 years ago.
So if I ship you acopy of Microsoft Word on a CD ROM and you install it on your computer, there'sno like debugging I'm doing over the, over the internet, right? Like, you mightcall the support line. I might tell you, well, sorry, or like, click thisbutton and it'll fix your issue.
But I don't need aninternal tool to fix your issue. Right? Yeah. But now when everything's in thecloud, I do need an internal tool to do that. So I think that change has causedlike a huge need for things like Airplane.
Julian:Yeah. I also think about just like the standard at which customers expect to,to have certain experiences or certain relationships with products, certainuser experiences.
It's definitelycaused a lot of this as well. I'm, I'm assuming just being able to have amodern cus consumer business facing, aspect of your business, whether or notyourself, selling SAS or not. Yeah, like almost like table sticks. You, youhave to have kind of modern feel and thinking about just like the changes and.
Ob, obviously, interms of changes for founders, like the velocity at which you can go to marketis, is as the time to go to market is decreased so significantly. What are, inyour opinion, some of the challenges now that founders are facing that are moreso amplified? Being that, that go to market timeline?
I feel like thatwas the biggest issue and then now that's kind of almost closed. What, what,what are some of the bigger challenges that people are seeing now?
Ravi:Yeah, it's a, that's a really good question. I, I would say honestly, there'sone of the biggest changes from when I started Heap in 2013 to Airplane todayin the 2020s.
The, like, you'reright, distribution is easier than ever in the sense of like, you can justpublish a website and you can put things out there, and there's all these,Reddit and Hacker News and LinkedIn and Twitter and all these places todistribute your ideas. The flip side of that though is it's easier for everyoneelse to.
Right. And so in2013, I remember when we launched teep we just put it up on like Hacker Newsand a couple other places, and it like immediately went kind of viral a littlebit and got like a ton of attention. We literally had over 3000 people sign upfor the product on like day one of launching it.
And it, inretrospect, it was a really bad landing page with a really bad demo video. Likeit was like not high quality. Whereas with Airplane, like we did not have thatat all. We did a launch, we got a couple hundred people to try it out and stufflike that, but we had to really work at it to get people to try it out becauseeven though it's easy to publish a tweet or even it's easy to put something outthere it is much harder to break through the noise cuz everyone else is doingit too.
And so just havinga cool idea. It was, if you had a really cool idea 10 years ago in tech atleast people would just pay attention. People would, I remember people would goto techcrunch.com and like just read every new funding announcement cuz thereweren't that many of them. And learn about cool new companies and try outproducts that way.
No one does thattoday cuz there's a billion funding announcements every single day. You can'tkeep up with all of it and it kind of gets sort of boring after a while. And soif you, you have to build something truly special to cut through the noise inthat way or you just have to do the, good old go to market techniques that,that have always worked around like sales and marketing and whatnot to, to getpeople to pay attention.
Julian:Yeah. And you know what's interesting also in thinking about both products,heat and and Airplane, is this idea of like, When you gain more AC or want moreaccess to the platform, it's like you increase the, the, tiers you have to pay.I've always, I've, I've never run a company that, has that kind of pricingmodel, but one of the challenges for me is when I think about it, is like, howdo I make sure that everyone is getting the most out of the current tier theyare, and how could I drive them to want more into the next evolution?
So my, my questionto you is, how, what strategies and tactics have you been able to actually getpeople to start utilizing not only heat, but also Airplane now and continue togrow up, up upstream?
Ravi:Yeah, I mean, honestly that's one of the hardest things in, in sas. There aretons of, of companies where, they have these like tiered pricing models and.
They eitherstruggle a lot to like drive people up through those pricing tiers or, they doit really successfully and it's very good for them. But it's like a huge, kindof unsolved problem that every company struggles with and every companyiterates on. I, I think I, I, I wouldn't say we've cracked the code at eitherHeap or Airplane.
I think companiesthat have done it really, really well, I think Datadog is one that does itreally well. I think the way Datadog does it so well is you can start out usingDatadog for almost for free. It's, it can be really cHeap on day one. You canget a lot of value out of it. And then it sort of uses this usage-based pricingthat scales very quickly as your company scales as well.
So you might beusing it on day one as a startup, paying almost nothing, thinking this is agreat tool, and then all of a sudden you start getting users. And your Datadogbill kind skyrockets and also sort of, it's, it's usage based, so you keep likeinstrumenting more events and things like that in Datadog and you realize yourbill sort of explodes, but you are getting value that whole time as your bill'sexploding.
So they've kind ofdeveloped this very nice sort of thing, which is very annoying for you as abuyer, but it's, it's, you kind of have to pay it in, in some sense wherethey're delivering so much value. That you naturally want to use more of it andinstrument more of it, and the pressing model is tied to the ways you use it invaluable ways.
And so when itexplodes in cost, you like get angry, but at the same time you're like, well,it's kind of worth it too. So, that's the thing. Really clever of them.Salesforce, I think they do a great job of, like, they have a seat basedpricing model, which is pretty simple. But they also like, have tons offeatures that they've built over, 20 plus years of existing that now they, theyknow exactly what each type of business needs.
And so you're asmall business. They'll sell you this exact group of things and then whenyou're like, oh, well, but I need more API credits, well I gotta upgrade overhere. Oh, now I need this thing, I gotta upgrade over here. They sort of walkyou up into their highest tiers. But part of that is because they've just beenaround so long that they have so much to sell you, right?
They have like somany different features and functions and they've tested it so much that theyknow exactly what you're gonna end up needing at each stage. You just gottaiterate on this stuff, I think is a real answer, right? So I think withAirplane, what we've done, we've kept it really simple. Like if you look at apricing page, it's not complicated.
What you get in eachtier. We've tried our best to be pretty generous with all the tiers. There'sdefinitely customers out there who are paying us only $10 a month. Per user andhave told us like, Hey, this product's kind of a steal. We would pay more forit, but we're getting all the value we need out of this set of features so wedon't feel a need to upgrade.
And on one handthat's a bit of a bummer, but on the other hand, we're an early stage company.And so we're happy to sort of get that learning, get that like evangelist forour product, who's gonna go tell 10 of their friends to go try an Airplane too.And then later on, when we're much larger, we can figure out how to optimizeour pricing so that that customer would've upgraded more.
Julian:Yeah. And in for, to answer like consumers questions, I was like, oh, why don'tyou just offer everything? Just, I guess, briefly describe the amount of workthat it just takes to offer these features. Things that break that you have tofix. Like there's so much. Like dead and a sweater that you put into these toactually justify the, the cost of what you're offering.
It's not just likea a, it's not just like a pre-packaged thing that, that runs on its own. Right.It, it's, it's a to maintain.
Ravi:Yeah. And, and, and to be honest, I think like most customers of a certainsize, they don't want. To pay nothing for the product, right? They want to paysomething for it because they want things like, like you're, you're mentioningthey don't just want a product that's gonna like fall over and break all thetime or something like that.
They wantguarantees, it's gonna be high quality. They wanna make sure there's a supportguarantee, all that kind of stuff. And so, if you look at that, like, wheneverwe talk to a cust company of a certain size, they're not even consideringanything that's not like the enterprise tier. And they're usually happy to paythe price for it because they understand that they're gonna use Airplane forsome mission-critical stuff.
For some reallyimportant pieces of, of functionality in their company. They're gonna wantguarantees that it's gonna stay up a certain amount of the time. They're gonnawant guarantees that if they run into a problem, and they, they can get supportin a certain amount of time, they're gonna want guarantees that, we're gonna besolvent as a business to be totally honest.
Right? Yeah. Like,if, if all of our customers use this for free, we're not gonna be around verylong and, you don't want to bet on a piece of software and invest heavily intoit. Only to find out that they go out of business in two years. So I, I think atthe end of the day, if you have a fairly priced product you can charge a prettyhigh price and people will be not even just accepting, they'll be happy to payit because they'll, they'll be seeing the value in it and they'll find, they'llfeel that that money is getting something real and tangible.
It's not just anarbitrary number. For sure.
Julian:Yeah. That's one thing I think I, I, I have appreciated about sas, kind of, Iguess. It's, it was around, but almost just like amplifying, which is the needto service customers and being completely customer-oriented almost, and almostwalking through their shoes versus like transaction or commodity selling.
It's really, it'sreally like a long year journey. What can we contribute to the success of you, eitheryour job or your business? Which is extremely exciting and I think it's, it'sone of the positives that, that that's come out of, businesses being built. Butjust to obviously, talk about Airplane a little bit more, tell us a little bitabout the traction you've seen thus far and what's pretty exciting about howthe next phase of growth that you're you, you're looking forward to.
Ravi:Yeah, for sure. So we're still pretty early. We're, the company's been aroundfor a little over two years now. And so we've raised two rounds of funding, alittle over 40 million total in, in funding. The team's, about 25 people today.And so yeah, we have, couple hundred paying customers, a couple thousand free,free customers, things like that.
We have a pretty generousfree tier, so you can get pretty far not even paying for the product. And thenwe, have a few hundred people who are paying either on our self-serve tiers or,a fair amount of people on our enterprise tier as well. And so it's kind of allacross the board.
There's sort ofsmall startups using the product. There's larger companies as well. Most of ourtraction has been maybe like growth stage tech companies, so companies withmaybe. 200 to a thousand employees, that kind of range where they're largeenough where they are starting to see some growth pains around internal toolingand they need to build some of these internal systems but still small enoughwhere they maybe haven't overly invested into this stuff already.
Sometimes we'lltalk to a company where they're like, yeah, I wish I'd learned about Airplanetwo years ago. Something like that. So that's kinda the sweet spot for us rightnow. But we do have some of those larger customers as well. So yeah, thetraction's been great. It's been almost all inbound.
So we, we launchedthe product almost two years ago, so July of 2021. And, and since then yeah,it's been, it's been mostly inbound growth. We did initial launch, we had alittle bump in signups, and then we had, a couple each day. And then it sort ofhas trickled up over time via word of mouth and then content as well.
So a lot of peoplecome discover us through our blog. We, we blog a lot about sort of, thingsrelated to our product. So Airplane is kind of, it's kinda like you mentioned,it's sort of a unique product in that an engineering team will buy Airplane tosolve a. Issue for another team. Like they will build like a sales dashboard orsomething like that.
And so we writeblog posts that are kind of. About other ways to solve those problems, likeother pieces of technology or open source solutions or things like that youcould use to build these kinds of tools. And then people will find those blogposts via like seo, like via search. And then they'll realize like as they'rereading this, tutorial on how to use Django admin or something like that, thatthey'll see at the end like, oh, by the way, this Airplane thing seems like itmight be a more interesting way to solve the same problem.
So we get peoplethrough that as well. And so that, that's just worked pretty well for us sofar.
Julian:Yeah, I love walking through the funnel. I, I could see it, I could, I couldsee how it kind of leads into kind of the overall not only customeracquisition, but it's interesting to think about how your product serves forinternal teams and thinking about how that decreases probably product attritionwithin teams and within companies.
And I'm, I'mcurious, like what are some ideas that you have around like partnerships? Icould see like, accelerators being extremely exciting to partner with a companylike yours because, there's a lot of work that needs to be done when you'rebuilding that mvp, but not necessarily needs to be foundational when you're,building and scaling your company, right.
To find productmarket fits a completely different ballgame than, kind of, kind of scaling. Butyeah. What kind of, what do you think about, or do you get excited aboutpartnerships or ways that Airplane could kind of partner or attach itself to avehicle and kind of grow it with with something else?
Ravi:Yeah, we started to explore that a little bit actually. So, we've tried a fewangles. The one that sort of worked the best for us so far is The, there are anumber of, like, so, I, I mentioned before, something like 40 to 50% of thetime that engineers are spending is building these internal tools.
And so what a lotof larger companies do, they actually outsource some of that work. And sothey'll outsource it to agencies that do like, tool building or in some casesagencies that do like, DevOps work or, or, or things like that is an outsourcething. All kinds of stuff. People will have sort of like outsource agencieshelping them.
So we've found thatsome of those agencies like using Airplane to accelerate the value they'redelivering to their clients. And so we've only done that maybe three or fourtimes where we've like brought on customers through that sort of channel. Byvirtue of like finding, like having an agency who, who tried an Airplane, likedthe idea, pitched to their client, and then was able to sort of like, implementsomething at the client using Airplane.
That's happened acouple times now, but it's something we've seen a little bit of success with sofar, so it's something we're starting to double down on. So yeah, that's beendefinitely pretty effective for us.
Julian:It's, it's exciting. I've been big on kind of thinking about partnerships andthings like that.
Also, just likeeven could, probably partner with some universities to even, get people onAirplane and, and maybe even learn some real skills and maybe when they take itto the other jobs. I, I love tools where you learn and, and it enables peoplebecause. I feel that when you find that right channel, then it, the growth canreally excel and, and it's exciting to kind of see what that how that plays outwithin the right channels.
But thinking about,kind of external and internal what are some of the biggest risks that you thinkthe company faces today?
Ravi:Yeah. I mean, I think like there's it, it's always hard to predict the future.I think like ultimately There are certain risks that are like, running outtamoney, all that kind of stuff.
I don't thinkreally have those risks are actually really well funded. We have a lot ofrunway, all that kind of stuff. For, for us, I think like the most. Likely riskis we have a certain approach that we're taking to helping people with internaltools. We're not the only ones helping with that problem.
We have a very specific,sort of unique, differentiated approach. We think that's, that's the reasonpeople have come to us over more mature solutions and stuff like that. Andwe're so far the only ones really doing what we're doing in, in this specificway we're doing it, being very developer first, all that kind of stuff.
I think the risk isthat, some up and coming startup. Comes and does like, has an angle on theproblem that is just better than our angle on the problem. And so I think thisis what you kind of see in various markets. Like if you look at a company likeslack. Before Slack, there were a bunch of other chat solutions and things outthere.
People use like hipchat or IRC or Gmail chat or things like that to run their companies. But whenSlack came out, it sort of just wiped out everybody all at once. And some ofthose companies had real traction, some of those products had quite a bit oftraction. And, and really what the, the difference there was like somebodyfound the right approach to that problem that was better than everyone elsebefore, obviously, if I ever knew that they would've just been doing the Slackthing from the beginning.
And so I think weare, we feel really confident in our approach. We have really happy customers.They're very well retained. We've actually never had a single enterprisecustomer churn from the platform. Every single one has actually increased theircontract sizes on renewal. We've only been around for a couple years, so it'snot a ton of data, but we, we have like, All that stuff is great, but ifsomeone just thinks of a fundamentally better way to solve the same problems,we're not gonna be around for very long.
So I think that'sthe thing we think about all the time. And so we're always brainstorming like,Hey, is what we're doing the best way we can solve this problem? And we havetaken things that we've built even a year ago and sort of tossed them out andsaid, Hey, there's a better way to solve that now because we thought of abetter way to do it.
Or some technologycame out that enables a better way to do it, or something like that. So youhave to be kind of disrupting yourself at all times. And if you're not, someoneelse is gonna do it to you.
Julian:Yeah. Yeah, well said. And if everything goes well, what's the long term visionfor the company?
Ravi:For sure. So yeah, I mean, overall, the whole point of Airplane is to savedevelopers time when they're building these kinds of internal tools. And so,right now there's a lot of stuff you can build an Airplane, but there's a lot,still a lot of stuff. You can't build an Airplane. And so when we sell to acompany, we're solving some of their problems.
We're not solvingall their problems when it comes to internal tooling. Over time, you want thatproportion to grow. Like, and the way we are gonna grow that is we're sort ofbuilding more and more types of primitives and building blocks that let you buildmore and more unique types of internal tools.
So for example,today if you have like a script or something that's like a database query, youcan model that an Airplane. But if something that requires like databasestorage or something like that, you have to sort of. Build your own. You haveto sort of bring your own database and attach that to Airplane.
But eventually youmay have our own storage options sort of built into the platform. That's oneexample, but there's many other things like that. And the other thing is reallyjust making that developer experience better and better so that when you'rebuilding stuff in an Airplane, it should be as easy to do as possible, as fastto do as possible.
If you're takingsomething that would be maybe 20 hours of work to do. Yourself without Airplaneand maybe Airplane can get it down to five hours of work. That's a big win. Butcan we get that down to two hours or one hour or can we like have a templateoutta the box that just does it for you?
And so these arethe types of things that we're starting to think about is like, how can we cutthat time down more and more and more so that the proportion of the amount ofwork that we can automate for the sort of like backend internal tooling workthat you to ultimately. Isn't moving your business forward?
Can we automate asmuch of that as possible? So like the long-term vision is like, can we actuallyautomate away all this internal tooling work that's not differentiated so thatyou can focus on the stuff that really matters to your business?
Julian:Yeah. And also just enable, like, it, it keeps you like what you have to, interms of like hiring, like it can be so lean and so efficient with the rightperson who can, I was looking at some of the technologies like type script andreact, things that people are using so commonly nowadays where, it, it, it's,it's.
It's needing thatmuch more gives you that much more ability to look like a mature company whenyou know you're earlier on in, in your, your growth as a startup, which is socool. I always like to, I always like this next section I'll call my founderfaq. So I'm gonna hit you with some RapidFire questions and we'll see whatwe're get.
So, first questionI always like to open it up is what's particularly hard about your job day today?
Ravi:Yeah. I mean, probably the same thing as it is from any other founders, whichis just time management. There are a million things to do and only a few hoursa day to do them. So, being able to divide my time between people at thecompany, customers recruiting various back office tasks, whatever that stuff.
Julian:Yeah. Yeah. And thinking about just like what your, Airplane's able to do, whatare some of the common things that, companies are using or building withintheir organization? What's like, almost like the standard of. If I was thinkingabout like a, like a startup a starter kit, what am I building internally thatyou see kind of most common?
Ravi:Like the things people build with Airplane, like most commonly? Yeah. A fewthings. I'd say like, number one, admin panels. So like you wanna have somesort of place for your customer facing teams to like, look up customer data,edit customer data, extend someone's free trial, edit their billing data, allthat stuff.
Number two would beoperational workflows around like data management. So like a GDPR deletionworkflow for example. If you need someone to be able to like click a button anderase a user's data, that's the type of thing you might model an Airplane. I'dsay like number three, like sort of like onboarding workflows and stuff likethat.
So like, let's sayevery time you onboard a new customer, you have to do these five things. Youhave to spin up their tenant, import a bunch of data, whatever. You can sort ofmodel that step-by-step workflow, multi-user workflows, permissioned workflows,all that in in Airplane really well. That's maybe a top three, but the natureof Airplane is everything people build is very custom.
So there's tons andtons of stuff that people are building on the platform. Has anything surprisedyou? A little bit, yeah. I mean, like, I would say like a lot of people useAirplane for things that I wouldn't necessarily even have thought of as like,internal tooling until people started building things that way and then westarted supporting it more.
One big thing, oneof the big use cases for Airplane is like scheduled operations. So if you'relike, oh, every. Every day I need to like, send out an email blast to my usersor every, every hour I need to sort of, run a payroll or something like that,that kind of stuff people will use Airplane for.
We didn't evensupport that functionality in the beginning. People kind of hacked around itand we're like, okay, well let's, let's add scheduling. And now it's actuallyprobably the biggest use case by volume of, of usage. On Airplane is sort ofscheduled operations.
Julian:Yeah, it's funny. Nobody's really ha like really kind of completed that.
Like I, I, I use itfor like, I use like Mixmax even for sales, and outbound, right? Yeah. Use,like Mix Max, use Apollo. There's so many other tools that you can use, butit's the timing of it. It's, it's super important and the customization. Soit's interesting to, to hear that people are still looking to solve thatproblem.
To thinking about,your journey as a founder up to this point, what's something that you're reallygood at now that you wish you were better at earlier on?
Ravi:Yeah. I mean, I think I'm, I like to think I'm better at hiring than I used tobe. When, when we started Heap, I'd never hired anybody for anything.
And, we had to sortof made a lot of mistakes along the way. And also had made a lot of really goodhires, probably more by luck than by skill along the way as well. And I thinkat, at Airplane we've been able to be a lot more. Skillful and sort of likeassessing people and making sure we're doing so in an objective way and reallygetting a good assessment of their skills and things like that.
In ways that wasjust a lot more naive about 10 years ago.
Julian:Yeah. Yeah. Is there anything you would kind of change or do differently in, inyour past experiences as a founder?
Ravi:Nothing major. Honestly. I think we were really lucky to, to get where we didwith Heap. There's obviously individual decisions that in retrospect are, arewrong and, and things like that. I, I think like the big thing is like if Icould go back and really like, spend more time with our customers andunderstand them better and build features that were more customer oriented, Ithink that would've probably saved us some time and gotten us to grow a bitfaster.
Julian:Yeah. On a personal, what, what's something that you really enjoy doing? Is itcustomer calls? Is it building product? Is it talking to advisors and orfundraising? What's, what's something you enjoy?
Ravi:Mostly customer conversations. I think that's the main thing. I like spendingtime with our customers. I think you.
It's never a wasteof time to talk to your customers. You always learn something from it, evenwhen you think like, oh, this customer's already really happy. There's notreally expansion potential in the account. Even when you start thinking in thatway and then you actually spend that time with them, you realize, wow, there'sactually a lot to uncover here.
And there's justlike interesting use cases and case studies and stuff like that that you canlearn and use it to inform the rest of what you do.
Julian:Yeah. What, what's one habit that you feel most founders should start building?Like today?
Ravi:Yeah. I would say like, have a really clear sense of what the most importanttwo or three things that you're trying to accomplish, like right now.
And like, write itdown somewhere, whatever to-do list on a piece of paper or something like that.And then just like, look at that thing every so often and ask yourself ifyou're actually doing that cuz it's so easy to get pulled in a hundreddifferent directions and realized that you've spent a whole week.
Working really hardand spent zero time on the things you thought were most important. So, I thinkjust like have those things in the back of your head and like pull yourselfaway from getting distracted by other things.
Julian:Yeah, that's a great point. That's a great point. Always like this nextquestion because I love how founders extract knowledge out of anything thatthey ingest whether it's early in your career or what books for people have hada lasting impact on you.
Ravi:Yeah, actually, honestly it's a bit of a tough one for me cuz I, I don't readmany business books. I don't think I've read a single business book in a decadeor something. I think at one point I read High Output Management cuz someonetold me to, and it seemed fine, but like, honestly, I think I get a lot moreout of having conversations with people who've done it themselves and beingable to ask questions.
I just don't. WhenI read a book cover to cover, I don't extract the learnings in the same way.Sometimes I have really insightful things to say. It just doesn't stick with mein the same way. For me, it's more about like, when we were early at Heap,being able to talk to the people who were growing in scaling segment or Lookeror these other companies that were a couple stages beyond us and being able totalk to them and say like, Hey, how did you solve this particular problem orwhatever.
And so that was thestuff I found really insightful for, for me. Yeah.
Julian:Yeah, so, so good and, and so fun about the, the startup ecosystem. Being afounder and, and grabbing advice and getting people's honest feedback. I feellike it's such an encouraging network and relationship that, that people have.
And it's excitingfor sure. I think we're all just wanna build cool stuff and as long as you'renot trying to build the same exact thing, I think we're all down to, to chitchat. For sure. Yeah. Yeah. Well, Ravi, I know we're closing to the end of theshow here, but last little bit, is there anything left on the table?
Any question Ididn't ask you that I should have? Obviously will give you a chance for yourplugs, but yeah, anything I didn't ask you that I should have?
Ravi:No, I thought you did a great job kind of going over, what, what I've learnedover the last 10 years and, and things like that. Thank you so much for havingme on.
Julian:Of course, Ravi. It's been such a pleasure learning about your early experienceas an entrepreneur and learning about what you've learned, kind of growing andscaling, keep how you've taken a lot of that into Airplane, but also just howyou've continued to enable and, and keep people using tools that increase theirproductivity or deliver on things that are common problems and, and really it'sexciting because it not only gives you a level of intelligence, but also helpsyou build more productive, more mature companies.
So really excitedto see what you're building and such a big fan. Last little bit, where can wefind you? Give us your plugs. Where can we, is it LinkedIn, Twitter, where canwe find to be a supporter and a fan?
Ravi:Yeah, for sure. So, if you, if you wanna check out Airplane, it's justairplane.dev dv.
If you wanna, lookme up I'm just on Twitter at @ravisparikh that's my handle. And then if youjust like search for my name, an Airplane or something, and then LinkedIn,you'll find my LinkedIn over there. But yeah, you can, you can reach out to mewherever.
Julian:Amazing. Ravi, it's been such a pleasure chatting with you. I hope you enjoyedyourself on Behind, Company, Lines, and thank you again for being on the
Ravi:Thank you so much.