Graham Hunter is the head of Segment's Startup Program. Graham joined my first startup in the Wharton VIP (venture initiation program) Techstars alum, the first employee at a Techstars company that then sold to Intuit. Prior to joining Segment, he was Director of Marketing at Patreon Today Ben and Graham talk marketing, no-code tools & startups! Segment offers a Startup Program to enable early startups to track data correctly and easily test the marketing and analytics tools necessary to grow their business. The program is open to any early-stage startup that meets the following eligibility requirements:Incorporated less than two years ago- Raised no more than $5MM in total funding- Located in Google Cloud eligible territory- haven’t previously received other Segment discounts- The Segment Startup Program includes three components:You can find out more about it here:

Graham Hunter - Growth @ Segment - Stories Podcast-MP3 for A...

Sun, 4/26 5:44PM • 29:56


people, building, code, segment, zapier, table, workflows, work, companies, startup, air, alfred, api, big, scraping, glide, tools, craigslist, specialists, projects


Graham Hunter, Ben Tossell

Ben Tossell  00:00

Graham Hunter  00:21

Hi, I'm Graham Hunter. I'm a program manager at segment I manage the startup program. And I used no code now for five or six years to put together everything from little projects to you know, managing the startup program at segment and excited to talk to talk to you all about it.

Ben Tossell  00:47

Awesome. Well, yeah, I'd love to talk to you about Firstly, let's like touch on the startup program and how that sort of managed with no code you mentioned that it was built by is one of the co founders and To me the whole process and we can dive into that a bit, but I think, just really want to say that like everyone points to segments starter program as like epitome of a good program for startups and it's like, really well done so just sort of give kudos there first and yeah, love to hear a bit about behind the behind the scenes of it.

Graham Hunter  01:22

Yeah, absolutely. I mean, kudos all goes to Ilya bull darsky, who designed the program from scratch as well as a lot of tireless effort from you know, segment, the segment team. I I have been there for about five months, and sort of have have, you know, taken the baton from from some people and on some projects and, and, yeah, we we use a lot of no code tools, primarily air table and Zapier and segment at segment, we run the startup program in the startup program gives a lot of benefits to startups that have raised less than $5 million and have been around less than two years. And so the first, the first sort of, like main code bit is the application process. And so you know, we have a website that gives all information and then when people are ready to apply straight from our website, there's an air table form that they fill out where they put in their crunchbase information, when they submit that crunchbase information. There's a sort of like web scraping process, that's not a no code implementation that is, that is, was done by a data engineer, but essentially pulls out all the data from the crunchbase profile, when were they founded? How much have they raised all these things and, and then pre qualifies them and then we manually go in and look at things. We have checkboxes that you know, you know, send little slack notifications and emails so that the right people can go and credit their account. We have all sorts of tracking in place. These air tables also have Zapier, zaps, linked up to them. For example, like when someone completes the air table form in Zapier, we say okay, then fire a segment event that says, application completed. And so this is how we keep track of all of the all of the events that are going on in our analytics funnel and stay on top of, you know, using segment per segment.

Ben Tossell  03:47

Awesome. So do you think that you could actually, you mentioned that that scraping process there's a I was going to a workshop yesterday and we talked about On dash dash, there's a spreadsheet tool that you can like implement a crunchbase API and then pull through company data that way. Even things like parabola wonder if you could now do if you were rebuilding it wonder if you could do the scraping part automatically too.

Graham Hunter  04:17

Yeah, absolutely. I mean, I've always been I started using import IO probably like, seven years ago or so years ago or something like that. And when it was, when it was free, and built a lot of import API's and that was my actually like introduction to no code, where it's like oh, it's just point and click and again, occasionally you'd need to go inspect an element or something like that. And so I've built that since then import has gone sort of like a little more in the in the in the way of like, lead genius and stuff like that. I think we're it's more for for for bigger companies, but still a really big of their product. But yeah, I totally think that we could we could do that web scraping. I also have in the past used import XML within Google Sheets in order to like do some of that scraping. I have not been able to get that to work for six months or a year or something like that. And I just kind of feel like there's so much JavaScript and so much just like happening in the front end now that like most of the times, unless you're scraping a static HTML page that import XML is not really like viable. Yeah, you know, one of those other tools seems great I'm not familiar with dash dash but

Ben Tossell  05:41

yeah, that's that's quite new new company is basically a spreadsheet data database. You can pull things into but yeah, parabola as my love problem. Yeah. Do you get a lot.

Graham Hunter  05:53

I use it a bit. We, you know, mainly for like mapping CSV uploads. into tools and things like that. I think that's the my main use case. I like their vision for like, it seems much less focused on sort of like technical people or even product managers or anything like that. And it seems like your computer can do awesome stuff. Why don't you like, you know, tell us what you do regularly. And we will automate that. And I'm just thinking about like me as a marketing manager 10 years ago, just like editing csvs from AdWords and like trying to make campaign changes and just like all that stuff, that could have been so much easier.

Ben Tossell  06:39

Yeah, I think, like, to me, that's what this whole no code space is about is that there's that job to be done and people are looking just have something, a solution. It just happens to be that, like, all the capabilities are changing the abstraction of how you do that is so much easier. It's funny that you said that employers was sort of your entry into the no code space of playing with API's. Whereas I think mine has been almost the other way around where I started just like building a website with web flow or card or whatever. And then when you start getting into the more complex things, like now make an outsider's profiles that have things going back and for which I'm actually like talking to the web flow API and using different like, web hooks and things like that. And the way I've learned more now is like you do those things you like your must step up, you upgrade every now and then. And then you think, Oh, wait, yeah, I can do something like that. So you're like, do just a little bit more you start with with an API, and then you start recognizing the patterns of, Oh, okay. I know sort of how an API works. Now, I understand why that would connect with that or why this would do this thing. And this would do that, that thing. So I think it's like a really interesting path into Building Products. And then also, like, it's almost the best place to go to start learning how to code like, dude. Yeah, absolutely.

Graham Hunter  08:08

And like you get farther further and further into it, where like, last night I, I bought Alfred, a license for Alfred and started working with Alfred workflows. And it is crazy. Like, you can do some simple ones that are like native to the to the platform like where you just like flip on an integration. But what I did was basically I connected Alfred up to the Asana API, which requires like browsing a bunch of web pages where people have made workflows and put them online, and you're just like, does this seem reputable? like is this do I just download it and like, Where do I do with it after that and yeah, download Alfred, you know, and so like, your, your knowledge just sort of like goes forward incrementally and you know, from Email Marketing and things like that I use like workflow builders and then you know, I'd use like If This Then That and If This Then That turned in turned into more Zapier use, and then that turned into like, messing around with like Integra mat or tray IO or something like that. And then suddenly you're just like, Okay, I think I know what JSON is. Like, I get it, it's just key value pairs. Okay. Oh, and then eventually you run into the thing where something's nested too far down, you have to figure it out. And then so, you know, yeah, I wouldn't say that I'm anywhere close to like being able to code in terms of understanding like, you know, like a tech stack from like, the model and the view and like, all these things, but definitely can deal with structured data and that's like a big first step to me when starting to get into the, from the beginner which is like If This Then That Google Sheets and maybe an air table into the intermediate, which is more about like, you know, like, you know, maybe Zapier multi zap step steps apps and stuff like that, you know,

Ben Tossell  10:04

do you think that this no code building will become one of those skills that we just see quite often as like, there's an integration specialist over a segment. I know what they do, there are no code builder, they like work on building these workflows. And if, if so, yes, how, how far out do you think we are from from that world? If

Graham Hunter  10:26

If I mean, that kind of exists, now, you know what I mean? Like we use tray IO at segment, mainly for, you know, working with Salesforce and things like that. But we, you know, have, you know, all sorts of little logic flows that can be duplicated and modified. And not everybody wants you pop it around in there and doing stuff. Like when you're working with production data and things like that. I think that that what's going to happen is that integrations will enable people to move To a self service world, where the experts provide oversight. And so, you know, even even if you look at like email marketing, or, you know, push notifications, oftentimes, like, you know, you don't go to the bigger and bigger a company gets, like, you know, I've been the first employee like four times, right? And I've been at a company like TurboTax, or segment right? at, you know, and a lot of these companies when you need an email campaign sent to your little group of user research people or, you know, you know, people who opt into this special feature, and you go to marketing, they'll say, Great, let me know if you need any help, you know, the, here's the software, you know, and so like, I think that we're moving much more to a world of self service where there are specialists in every seat, and then everyone else is, is a consumer Have those software's like, customer IO or MailChimp or whatever. And then they're trying to they're trying to do it. And then they're like, Hey, I just need a quick hand. And then the marketing person comes in, and they're like, Oh, yeah, you got to use these templates. They're like, Okay, cool. And then, you know, and then they sort of, like, get it and they say, Does this look good? And they say, Yeah, it looks good. Click Send, and then they send it out. So that's how I think no code is gonna permeate, like, you know, smaller and even eventually, larger companies is through like, it's not like there will be no code specialists, there are no code specialists, and you can hire one on, you know, whatever, Freelancer marketplace to do this kind of stuff. But they're there. I don't think that no code specialist will be a role. I think everyone will be expected to know no code, just like, you know, Google Sheets, you know, no one's grilling you about your Google Sheets skills unless you're building depth financial models, you know. And so I think it'll be like that.

Ben Tossell  12:58

Yeah, I suppose. I probably The same I was speaking with one of the founders of obviously, Ai, and where they help people do machine learning. And it's like, we are taking care of the basic stuff and like the easiest stuff so that you can do. If you're in machine learning, you're a data scientist, you actually, like, the expertise is almost like the top, the top bar then. So if you're in that role, you just focus on the creative, really specific, like interesting expertise of that area, rather than filling your time with all the fluff and all the basic stuff that you have to do over and over.

Graham Hunter  13:36

Yeah, the blocking and tackling that everyone else is doing. Yeah,

Ben Tossell  13:40

yeah, exactly. So yeah, I think I see no codes, like no good skill being sort of a big ubiquitous thing where it's just like, Oh, yeah, I know how to know code stuff. I know how to like do these things. I can connect these things together and build a tool that schedules, emails based on these actions. or whatever. I think that Yeah. It's funny that you say that? Well, I suppose segment, it may be a more unique culture that you say there's people already there who know these things, you know how to do that stuff and like, repeat that. And I think obviously, big customers of trade IO and even Zapier and things like that, obviously, understand that but wonder if you think that there's like more startups these days are using their tables up here, etc, who don't, who aren't quite there. And I'm really looking for more of the generalists. And like, they've got a team of maybe 20 people, and they're looking for people who can do more with with like, multiple tools and things. I wonder if it's more against the smaller companies trying to move fast with, with less people who maybe haven't bought into this whole concept? Yeah.

Graham Hunter  14:55

Yeah, I mean, the hardest. I'll tell you that like when you look at startups, are growing from stage to stage. Like, you get pretty big before you find true generalists, you know what I mean? Like, I'm a program manager, like, that doesn't exist at a 50 person company, you know, like, barely let alone a 15 person company. You know, the, to me one of the like, I won't call it a problem, but like, one of the reasons that no code isn't implemented more at early stages is two things. First of all, like engineers are the earliest employees of startups. And engineers like are in a constant sort of, like build versus buy, kind of like mentality. And founders also see the product that they're building as a piece of proprietary technology. And so like, you know, it let's say that, like, you know, Goldman comes in and they're like looking at your startup, your FinTech, startup, and They're just like, you know, okay, let's see what they're doing. You know, like we want to, like have some early conversations about whether we want to like, choir you or acquire you or whatever, right. And then your your platform is just entirely just, like strung together and it says, Yeah, yeah, totally they're not interested in that they're interested in buying proprietary technology and that encourages founders and engineers to go the route of building a custom and I think sometimes incorrectly, you know, but you know, who the heck am I you know, so, so I think that a lot of times like the path towards product market fit is hard enough and that like adopting these technologies, and then finding the product market fit, and then retrospectively building it, you know, building the product that you would have been using the whole time to replace some of that so that you can customize it and make it proprietary is like a lot of times how A lot of companies should work. I'm sure there are exceptions around like, you know, social networks and things like that you're not going to go launch a social network on some sort of like, WordPress template necessarily, you know?

Ben Tossell  17:14

Yeah, yeah. It's interesting. I wonder, you know, what kind of what kind of companies you think would benefit the most out of no code, then do you think it is actually probably not the tech startup types that are more inclined to be getting the benefit of having their team talk about no code and implementing these things? or?

Graham Hunter  17:35

Yeah, I see no code as like the so you know, you're familiar with like, agile and waterfall and like, you know, like all these things, right. And there's all these big companies that are going through like digital transformations that are trying to that are trying to like move more towards faster execution, and but they don't really know how right like, is like big and slow and right. And so like, I think that no code is like, best for those internal programs at companies. So no big company is going to use no codes for customer facing thing unless it's like sexy, you know? Like, no, they don't want people to be filling out air table forms, you know, the air table doesn't provide enough customization in the forms to, like even ever get it to the point where some brand is going to like, you know, some sort of Nike is going to like, you know, have people fill that out maybe in a customer service situation like I don't know, but but for those companies internally, I think that that these tools can help push people towards executing faster and and also you know, like just speed of execution or speed of you know, like getting Through execution to an outcome, and then and then showing that outcome to people and being like, we should build it, you know, essentially doing the same thing that entrepreneurs do, you know, with MVP, but internally at bigger companies, that's where I think like no code is most like useful. And that's the case in which I would imagine that a no codes person who's a specialist could like, work it into it. And anybody dealing with no code, anything is like coming to this person, because they have, you know, 10s of thousands of employees.

Ben Tossell  19:30

Yeah, yeah. No, that's interesting. Yeah. I'm not like, had conversations with many people who are anyone really who's like, talk these through at different companies stages and where they think so it's been? Yeah, that's a nice, nice take on it. So let's talk about some of the smaller projects that you've, you've worked on it. I'm going to talk us through some of those and

Graham Hunter  19:52

yeah, I mean, these things are I just started publishing on air table universe and love it and very much like upset Like, making it usable by other people just because I'm like, I built these things so much I'll build like little financial models and then it'll, you know, that project will just drift into the background and then I put all this work into it and it's not useful.



Graham Hunter  20:19

So now I'm like polishing these things off and putting them on air table universe. Number one, I am helping to plan my sister's wedding. And you wouldn't believe the number the like intricacies of like, okay, there are people who are allowed a an anonymous plus one and there are people who are not, and there are people who are there with their partner and some are bringing kids. We also have housing accommodations, and you know, groups are being paired together to stay in rooms and you're just like, Okay, we've got a people table, okay, we've got a groups table, we've got an account, you know, just like you the amount of complexity is just insane.

Ben Tossell  20:59

I'm getting married in June. So I definitely know the complexities. TBC Yeah, yeah, we've definitely all those things I'm thinking. Yep, yep. Yeah, yeah. So yeah.

Graham Hunter  21:12

Are you are you doing in an air table? How are you doing?

Ben Tossell  21:15

Now? We use a with joy, which is just a spreadsheet, basically. Yeah.

Graham Hunter  21:22

It's like a service meant for wedding planning, right?

Ben Tossell  21:25

Yeah. I mean, I would, I was gonna build like, a glide app and then have air table, link everything together. But then I realized that I'd be more involved in the actual details I use Julie. And she blocker Yeah, exactly. So yeah, I

Graham Hunter  21:44

love that. So in addition to that, like one of the ones I built recently that I'm most proud of is my girlfriend and I were looking for a new apartment, in the Bay Area. This is before Corona or anything. Right. And so in the Bay Area, it's like a race. You know, I mean, like, you can find reasonably priced things. Oh, yeah, it doesn't have a yard. But it's got this. It's got, you know, it's like everything just falls in the middle, right, really, like, uncompromising, so much. But there are good places out there, if you're the first one to them. And so I'm on Craigslist all the time, just like no hide, you know, like things like that. And so I built a air table that has a combat board for our application process that uses the air table Web Clipper to scrape every piece of a Craigslist page. And so when I'm browsing Craigslist, I'm like, oh, that seems cool. I just click the Web Clipper, add it, it pulls in all the information, the rent, the number of bedrooms, the number of bathrooms, some of its parts with formulas and stuff like that. Yeah. But the content of the page, the email, the phone, Number, you know, all these things and if it's a phone number, then you know, you know, going to SMS through through blocks, I think and then if it's a, if it's a, an email address, automatically sending an email address immediately that, that that's that that's pretty long. Like you want to keep these things short. I know but I think people are screening screening applicants, like, regardless of what fair housing says, like so, you know, it's like me and I'm like, it was so wonderful to see you're posting my girlfriend and I you know, Susanna, you know, it's just like a little flowery like, LinkedIn profiles, you know, just like things like that. And we get really great response rates, people like wanting to schedule times with us. I'm used, I'm using calendly in the in the invite and then there's a calendly to Table Zapier integration that moves them to the next stage once they've scheduled as well as like a follow up email if they haven't, right. So emails, our follow up emails calendly for booking, for booking, for booking showings and then moving moving the stages along of you know, like basically the combat view looks like Craigslist you know, hearing the all the houses big pictures that like oh yeah, that place Oh yeah, I want to know. And so that that's really something something that I enjoyed building and now I'm doing the same thing. I'm building something for like, you know, sort of like Craigslist furniture hunting.

Ben Tossell  24:43

Yeah, but you could do a base for a lot of Craigslist. Things like embedding Craigslist in air table and just hope doing Yeah, it's great.

Graham Hunter  24:54

Yeah, exactly.

Ben Tossell  24:56

Now I used to have to follow you on it on University. Then take a look at some of these bases because I think for Nepal and doing some tutorials or workshops or something on that would be awesome, I think. Yeah, we should definitely do that.

Graham Hunter  25:09

Yeah, absolutely. And it's like, the faster you find a place, it's just so much work keeping track of it. I run a community house now, and you wouldn't believe the number of applicants so I say, hey, we'd like to schedule a time to talk to you and they say sorry, which houses this? I'm just like, I just I have no idea what I'm doing there. What everyone's just texting me and say sure Thursday. You know, it's just keeping track of it's so so hard and being on top of it is the difference between you getting an amazing apartment and not?

Ben Tossell  25:44

Yeah, for sure. That's awesome. I love that is your are your projects mostly around like, mostly air table focused at the moment?

Graham Hunter  25:55

Yeah, you know, I've looked at glide and things like that and it seems really cool. I think it seems a little like, like maybe some of my other projects might be like good for it. But most of the things I'm the only user of my product, you know what I mean? And so like glide seems like if you have a group of people like a wedding, you know, that where they're all other people like, you know, contributing information and things like that, that that would be like better. I have used boundless a lot and I love love boundless. Yeah, in my last role at the head of marketing advisor, basically, visor helps people to optimize their withholdings. And so you know, a lot of people get a tax refund at the end of the year in the US, and that's basically a loan that you gave to the government interest free, you should ideally get nothing Oh, nothing, right. And so advisor helps you to like hit that. And so their application process where you like, apply to talk To a, you know, a CPA and or do it through the app and upload certain documents and things like that, you know, we have a pre application process and that was all in boundless with segment and Google Tag Manager events that are tied to like paid acquisition how far people got in the funnel relative to so it's like, oh, how much are we paying for a person who is pre qualified? Or how much are we you know, things like that. And I really liked using boundless for that, which I think is just kind of like glide but more for the web and mobile is that accurate? glide is primarily primarily like mobile web. I think it's

Ben Tossell  27:42

it's more for just web was was your partner at what we initially saw was like, sort of like bubble, the different user interface, but log still very powerful.

Graham Hunter  28:00

Awesome. Yeah, but but most of my stuff is just for myself and so I use generally air table and air table blocks and Zapier and you know, some some other light tools Alfred, you know, I have my personal productivity workflow that I'm always optimizing. I think the number one thing there that I'm that is going to change my life is like the ability to write down a task quickly. And so my Alfred, when I press Command space, sends things into Asana, and so push a space and then like just a task reminder and then enter and it's gone. And it's in my Asana and I'm back to whatever I was doing. And so that's the difference between be writing it down and just not and be like, Oh my god, I never followed up with Ben.


Yeah. Wait a minute. Hey, Ben. So sorry for the delay, like, yeah, they're off to the races, you know, like,

Ben Tossell  28:55

so cool. Awesome. Well, yeah, I really appreciate you you. run us through some of the stuff you do. segment and personal projects and sound like they're more in the advanced skills type of thing. But I think, yeah, once people get into them, people can really level up quickly in this in this whole space. So I really appreciate your time. And yeah, we can, we can wrap it up there and let us know where the people know where they can sort of follow you and follow your stuff and see some of your projects.

Graham Hunter  29:24

Sure, yeah. I think that Twitter is just the best it's at marketer Graham on Twitter, gra h AM, and I'll send a link to my or table universe profile. And, you know, we'll just go from there.

Ben Tossell  29:37

Awesome. Sounds great. She said that


it makes so much Ben.

Ben Tossell  29:41

Thanks so much for listening. You can find us online at maker or on Twitter at make that we'd love to hear if you enjoyed this episode, and what we do next.


What's your story?  Tell us how you use no-code
Something wrong?