Philippe Gamache  0:07  
Music. What's up, guys, welcome to the humans of martech podcast. His name is John Taylor. My name is Phil Gamache. Our mission is to future proof the humans behind the tech so you can have a successful and happy career in marketing. What's up everyone today, we have the pleasure of sitting down with Simon Heaton, Director of growth marketing at Buffer. Simon started his career in the agency world at Banfield in Ottawa, Canada. He later moved over to Shopify, where he would spend nearly seven years, first as a content marketing manager, and later as the senior growth lead for all of acquisition Simon's also worn a part time teaching hat for over five years. He was an instructor with the Telfer School of Management at the University of Ottawa, as well as a professor at Algonquin College. He's a startup mentor for founders that are part of the Singapore based equity fund at antler, and today, Simon is director of growth marketing at Buffer, the world renowned social media management platform, Simon, thanks so much for your time today. I'm really pumped to chat. Yeah, thanks,

Simon Heaton  1:09  
Phil, excited to be here too.

Philippe Gamache  1:11  
This episode is brought to you by our friends at Knack launching an email or landing page in your marketing automation platform. Shouldn't feel like assembling an airplane mid flight with no instructions, but too often, that's exactly how it feels. Knack is like an instruction set for campaign creation, from establishing brand guardrails and streamlining your approval process to nack's no code drag and drop editor to help you build emails and landing pages. No more having to stop midway through your campaign to fix something simple. Knack lets you work with your entire team in real time, and stops you having to fix things mid flight. Check them out@knack.com that's K, N, A, K, and tell them we sent you. This episode was brought to you by our friends at customer IO, oversold on a legacy marketing automation platform that is still struggling to update its user interface. I've done a tour of duty with all the major marketing automation platforms, and many are definitely similar. But customer IO is the most intuitive and beautiful platform. I'm talking about the industry's top visual workflow builder to design and implement your unique messaging strategy, powerful AB testing features inside your workflows, not just on subject line sends, hold out testing functionality to see the incremental impact to your messages queue draft mode, so you can QA messages and conditions in production with real users before anything is sent. Copy workflow items, so you don't have to repeat the building process again and monitor campaigns tests and keyless membership growth from your personalized dashboard. The icing on the cake, marketers using customer IO have seen a 20% increase in conversion rates from strategic messaging, so stop using clicky old tools and adopt a multi channel approach that creates joyful interactions with your customers. Start a free trial without a credit card@customer.io and tell them we sent you. You made my first question, really easy for me. Recently on LinkedIn, you posted about all the tools that you have in buffers, martech stack, and you know, you have some pretty well known products in there, customer, IO for automation, Ahrefs for SEO, Typeform, figma, segment, Mixpanel, buffer, obviously eating your own dog food there. Not sure if you had a chance to check out Ascot breakers. State of martech report that dropped, I think, couple weeks ago. We're recording this in the middle of May, but there's one really interesting question in his report that he asked a lot of survey responders. What would you say is the center of your marketing stack? Kind of the orchestrator? And the most popular answer amongst recipients or respondents was the CRM. And what I found really interesting from your list that you shared on LinkedIn is that buffer doesn't list a CRM. What would you say is the center of your marketing tech stack, and does buffer even have a CRM?

Simon Heaton  3:57  
Yeah, it's a really good question. Phil, so we Yeah at Buffer, we don't traditionally use a CRM, like HubSpot or Salesforce. I'd say there's two kind of main reasons why we don't do that. So we're very much a product led growth first company, we don't have a sales motion at Buffer, so there's no sales team, there's no account managers, and so the traditional requirements and from a commercial or, like, a revenue perspective don't really apply to buffer, lead, routing, scoring, those sort of things aren't as prevalent in our user flow as some of the more enterprise style organizations that have a robust revenue org in place. And then I'd say, from like a marketing perspective or gross motions aren't as overly complex enough to require like a CRM to orchestrate their delivery. What I would say is our closest tool in the stack to a CRM, it's probably our data warehouse, you know, as a very product first and data first team, our CDP, which is segment as you mentioned, is truly the brain of the operation. Of what we do where that information is surfaced for the team could be something like a product analytics tool, mixed panel, everyone can get in there and use that and find information about a user or customer. Or it could even be the marketing automation platform, customer IO, which has pretty much all the columns that our users receive. And so I sort of go to those two to understand usage patterns, trends and find details about a user. And I would say Mixpanel in particular has been an awesome tool for us at Buffer. It's it's been set up really well by our data science team, and it's loaded with pretty much all not only product usage data, but also marketing data as well. I think I mentioned in that post that we implemented some tracking on on the server side to facilitate this. And we're also, you know, piping data, not just from the website into Mixpanel, but also from these tools as well, AdWords, customer, io, Pendo, you name it. All these sources are coming into Mixpanel, so we can sort of get that broader story of our users and stitch together how marketing efforts are really impacting product lifecycle, the user lifecycle, you know, more downstream than, let's say, website metrics,

Philippe Gamache  6:06  
very cool, makes a ton of sense. Yeah, I was assuming that there probably wasn't a big sales team at Buffer, but you can actually see, like salaries. I was like creeping on there, and I didn't see like a sales development role or anything like that. So the full true, like product led motion there. I think that in the report that Scott Brinker did there was probably, I think he breaks it down by B to B and B to C after. And what I found interesting by chatting with folks in the industry that are in B to C, and even, like overseas marketers that are still in B to B, this like term of CRM and marketing automation, there's actually, like, a lot of, like, misnaming and overlap in there, like in Europe, a lot of folks that use marketing automation, they call it a CRM. Like, it's just like a customer relationship management tool. I'm sending emails, I'm engaging with my customers. They call it a CRM. But like in North America, especially in B to B, the CRM is really like, we have a sales led motion, and leads come in for marketing automation. They're synced into Salesforce or HubSpot, and we have like, all these like sales statuses, and we need to, like, go through pipeline and try to predict, like, who's gonna close, who's not gonna close. So it's, it's, it's an interesting world. And, yeah, I thought that was really cool.

Simon Heaton  7:23  
Yeah, absolutely. I would even say, you know, in the craft of marketing automation, you really have that dichotomy as well, where there's a huge portion of individuals that are really marketing ops from a revenue perspective, of like lead scoring, routing, you know, nurturing towards SQL and MQL type metrics, where, for Buffer, you know, some of those concepts do apply, like qualification, but, yeah, like you said, we don't have an account team. We don't have a sales team. And so really, the product is 100% self serve. And so marketing automation for us is sort of the supportive motion of getting those users through the product lifecycle. Yeah, very cool.

Philippe Gamache  8:01  
I was like, thinking of what my answer would be to that question, like, what is the orchestrator in my tech stack? And it's a bit of a confusing question, because, like, the orchestrator is maybe the one like activating that data and doing something with it. But really, you can't do anything with the data if you don't have the data in the first place. So like, what is the center of this stack? I think a lot of folks just went to the CDP or the data warehouse, whether you're composable or, like, you're all in on a CDP platform. Now you guys are using segment. I've used it in past companies. Also for me, like the orchestrator of my stack, I think, is always going to be the marketing automation platform, especially from like, a marketing point of view. But like, everything that matters, like, should end up in that automation tool to be able to, like, power growth messaging, right? And I know buffer is, like you said, a big fan of customer IO as your marketing automation platform, maybe walk us through, like, what you love most about it, and maybe share, like, some of the cool workflows that you've created with your team so far. I don't know, like, how hands on you are in customer IO, these

Simon Heaton  9:03  
days. I know you guys are still pretty small, so maybe you're getting your hands dirty still, yeah, and that's one of the beauties of working at a smaller kind of team, like buffers. Yeah, I'm able to get my hands dirty and involved in the work of a day to day, which, you know, I particularly love to do. So yes, I'm deep in customer IO. Actually, I've used it a few times in my career. The first time I came across it was at Shopify, actually. So in the earlier days, it was the automation tool that we use, and it was already in place when I joined buffer. Eventually, Shopify grew out of an event, I think, switched over to Salesforce marketing cloud to facilitate some more of these enterprise motions that we were speaking about. But at Buffer, we still use it. We really like it. We use it largely to facilitate our email marketing and push notifications for our mobile apps. So that's set up via their SDK, and we also run both our marketing programs. Communication, lifecycle, blog, newsletter, those kind of things through the through the platform, as well as our transactional programs as well. So you can think of like billing, product notifications, account setup, security messages, those kind of things. Personally, I love it. I find it an incredibly powerful platform, particularly for that scale up sized org that maybe has outgrown, maybe a more accessible email tool, and they're looking for something a little bit more robust that can have more segmentation, personalization opportunities based off of the data that they have. It's an overall an incredibly complex and comprehensive platform I really enjoy, you know, out of the gate, you know, the journey mapping functionality the WYSIWYG editor makes it really easy for the team to pick up, especially if they're not overly technical, super user friendly. But they also have some of those deep technical capabilities, and you can make everything from from scratch, from HTML through the API, things like that, which we require. I think I mentioned this. But, you know, the transactional element of customer IO is really interesting, because I know in a lot of organizations, we split that up. You'll have, you know, your marketing comms coming from one thing, and then maybe your product comes coming from SendGrid or some sort of other, you know, distinct platform we've really put the effort into have these all living customer IO so that we have sort of a singular view of all of the communications that users getting. Because regardless of whether it's like a marketing communication message or, you know, something from a product to the user, it's still buffer, right? It's the same person that's sending it for them, the business, at least, and we want to just make sure we have full visibility over that experience, and we're able to take into account those different touch points, not only the marketing ones. So that's been super valuable to us. And the beauty of that is that also has, you know, dedicated IP tools for the transactional service. So it's it's very reliable. We're able to safeguard those emails and the domain reputation from marketing sends and more broadcast based messaging that we do. And I would say there's a very I mentioned earlier to the segment integration. So we use segment as CPD for buffer, and the integration with customer IO is like integral for us all. Like I said, all of our data is in segment. And, you know, I treat marketing automation platforms as an equal to like a data or product analytics platform. We need to make sure that it has the right data coming in, you know, formatted accurately, and the right data coming out in a timely manner. And so being able to rely on the integration with segment that is how we pipe up everything else and buffer has been an incredible unlock for us, and has allowed us to leverage user data, product data in really unique and interesting ways within the workflows, whether that's snippet personalization through liquid within the emails themselves, or for more robust kind of workflow customization based on certain actions that users take or do not take, we've been really able to light up some really cool and interesting lifecycle programs to support our users successful the product thanks to the power of the tool itself.

Philippe Gamache  13:13  
Very cool. Appreciate the detail and all the stuff you shared there. I actually just recently got back into customer IO. After the they decided to sponsor the show. We migrated our newsletter for the podcast from for MailChimp to customer IO. I used it heavily when I was at close I spent like, a full year and a half getting really deep in the product and fell in love with, like, the the visual interface is like, the first thing that's like, striking, like I came from a world of Marketo, and so okay, I was, like, logging into customer I own, building visual workflows, and I just fell in love with that. And I told myself, like, I'm never gonna use Marketo again, but not closing that door on it forever. But I was like, surprised, like, it's like, a year, year and a half or no, probably a bit more to two and a half years since I used it. So logging in again, some of the cool things that jumped out to me, that I know you guys are using is, like, hold out functionality. Hold out testing functionality. In the past, I've had to, like, manually stitch this together, like being able to do incremental testing. Like, it's one thing to be able to say, we sent out this campaign and we saw, based on UTM codes, how many people went to the landing page and registered from from our emails. But it's another thing to be able to say, because of all these other touch points that are going on, I'm gonna hold out 235, percent whatever from this email audience, and I'll send it out, and I'll be able to compare the conversion rate from the test group versus the holdout control group. Yeah, maybe walk us through, like, how you guys are thinking about that. And the other thing that was really cool is, like, in marketi at the time that I was using it, maybe they added this now, but like, you can't do AB testing within the work. Flow itself, like in customer IO, building a workflow and a journey like, you can have 17 plus different actual AB tests all within the same flows, like all as different branches. So, yeah, just curious how you guys think about that.

Simon Heaton  15:14  
Yeah, it's a really good question. And I would say, you know, experimentation is a cornerstone of our team's approach and methodology being very data focused, and yeah, customer IO does make it really easy. I'd say there's a few versions of the tests that we run within customer IO. So the holdout test that you mentioned is a really great functionality that's within the platform in that whether you're, let's say, trying to launch a new program and you're trying to just validate, you know, will having something make a difference against, you know, not having anything, or you're just trying to compare, like, the performance of how a particular campaign influences, you know, the broader user action or lifecycle, the way that customer IO set up the holdout, it works really nicely in that it actually counts a delivery for the holdout group. So in some cases, you might end up with, you know, no stat or no data point for the holdout group at all that you didn't send the email to, and that can make tracking that cohort over time a bit more challenging, or at least a bit more manual for the team to do. So what the holdup functionality and customer IO allows us to do is essentially based on the split that we've decided allocate one group of the cohort into the recipient button, they'll get the variant, and one in the holding customer, IO will trigger a delivery for both of those. But of course, the people in the holdout don't get a subsequent metric for, let's say, open or click or anything like that. Because we have this deep integration with segment, we actually surface all of our customer IO platform data in Mixpanel. So of course, you can get it in customer IO, but Mixpanel, we find, is a lot more user friendly for us. So we can set up unique reports or dashboards to support experiments because we have that delivery event that has been fired for the holdout group. It makes cohorting the two groups exceptionally easy within custom within mixed handle story, so that you can actually track, you know, the differences in behavior between both of those by just the the action ID, if they receive the email or not based on that delivery event, which is super cool in some platforms, you actually don't get an event associated to the holdout group, which can make that downstream tracking a lot more challenging, and so we use that a lot within the platform. We do AB testing, both on the message level as well as the workflow level. Message level is really easy. It's all within platform. You know the scale. You can slide in to see how many people receive each variant. And you just select two different versions within the workflow. They have randomization logic. So you can essentially have a branch after your trigger event, and you could say 50% here, 50% there, and they can each receive, you know, a different email, but the same timing, the same, you know, whatever it is that you're testing, I guess, and that makes it also super easy as well for us to have, you know, ongoing tests pretty much in all of our programs, more or less all of the time, whether it's on the content, the sequencing, or like, or just in general, like a holdout test.

Philippe Gamache  18:09  
Very cool. Yeah, it's interesting to hear how you're you're pushing that into into Mixpanel and being able to analyze that further, because I feel like that's like having the functionality to be able to do holdout tests is one thing. Like, you can figure it out in your tool, but like, what do you do with that data after and remembering to just, like, go back to those experiments, because it's not as quick as an AB test. And a lot of these tools are you setting this up and it's like manual. When I was@wordpress.com we had an internal tool for our CDP, and for a lot of this, like, uh, internal platforming with data. And so when a user started a trial, we would have a daily sync that would just take a randomized group of those folks and put them into a holdout group, and those folks would just never get any email columns from us. And just so over time, we would have this, like, big cohort of people that, just like, didn't get any message from us, but like, there was no functionality to just like, see real time. Like the results of that, we'd constantly have to have a process to go back to look at that list and compare the conversion rate versus like, other groups. It's interesting here, how, how you're using that. The other feature not to, like, spend a full episode on customer IO there, but I feel like we're jamming on this a little bit. So I'm curious to hear if you guys are using this too. One thing that I fell in love with when I was using this at close was this, they call it QA draft mode, where you can essentially, like, build a workflow, super complex, bunch of steps. Sometimes you're, like, triggering off certain, like, data points, and you want to, like, test this out and like in production, almost like before you press go on it. And I've have yet to find a tool that allows us to do it in the way that customer IO does it, and they've been doing it for, like, multiple years. And I'm shocked that, like, other tools haven't caught up to this yet, but you can build your workflow in. Like your production environment with all the data triggers and everything, and you can press go have it live, and you can see real users qualifying for different branches. And all of your actual emails aren't sent yet. They're like, sent in draft mode. So you can see what they would look like, and you can make sure that, like, yes, this person would have qualified for that, and then you look at your liquid code and how it actually, like, ended up looking like, I just find that fascinating, because, like, I don't, I haven't found another tool that allows you to just, like, see what the result of your automation is in production with real users, but just not sending out messages quite yet.

Simon Heaton  20:37  
Yeah, and I, what I really love about that feature too, is like, once you've confirmed too, it's like, just a click to actually send that out to the vested users. If you're doing just like a short term holdout to validate stuff, right? Yeah, it's really, really valuable for us too. And we've done, you know, several, let's say, new iterations of onboarding or some product notifications, and we just want to validate before we turn it on just due to the frequency or volume of the sentence for those programs that things are showing up, you know, as they're supposed to. And like, you can only see so many examples as you're scrolling through preview with data that, like, once you start getting real ones in, that's when you sometimes catch errors and things like that. Yeah,

Philippe Gamache  21:16  
yeah. Super cool. Maybe we can talk about some of the underrated tools that you mentioned on the LinkedIn posts. I wanted to ask you about one name, particularly that jumped out to me that I wasn't familiar with, and that's a redash I think most folks are familiar with, like the tableaus and the lookers for like the bigger like bi names. One of my first startups was called folio in Ottawa, in the BI space I'm familiar with, like, the the tools for SMBs, like data box and get code board and grow and and all those other tools. But yeah, I'd never heard of redash. So, um, why? Obviously, like, checked it out since then, open source tool kind of like playing in that, in that bi space. So, yeah, just curious, uh, your thoughts on that so far?

Simon Heaton  21:58  
Yeah, yeah. Well, first of all, say, buffer. We love open source tools. We try to be super open. Most of our business, our code is open as well, if people want to go check it out, and so we always are in favor of supporting businesses that kind of follow along that path. Rehash is a really interesting tool, because I mentioned we have mixed panel and mixed panels, primarily our BI tool, that's what the majority of the team uses. Redash is especially useful for more complex queries that require us to join data from multiple databases. Buffer as any tech company that's been around for over a decade, we have some dust settling. We have a couple data warehouses set up a couple databases that we pull from. And so redash really lets you know, myself or our data scientists and analysts write custom queries and visualize the results when data needs to be stitched across these multiple databases. And so really, that's the primary usage for redash. We also use it for exploratory work in that regard as well. So when we're doing growth analysis or anything like that, or data scientists will go in there and use redash to do those queries and modeling as he's exploring. But it's really sort of a secondary data resource for us, second to mix panel.

Philippe Gamache  23:18  
Gotcha? Yeah, it's a it's interesting to hear that you're, you're using Mixpanel in kind of like a bi perspective. Also, we use amplitude at my current startup. Super similar tools. We're using Mixpanel at clubfolio, so I'm familiar with both. But we also use Looker at my current startup, and we use looker@wordpress.com Also, one of the few third party tools that we were actually paying for and not building out in house. But it's, it's been interesting to see, like, the difference between this, like product analytics, World of amplitude and Mixpanel, which aren't really just product analytics tools anymore. Like we just had Adam Greco from from amplitude on the show. The episode dropped last week, and he was kind of like talking about how, you know, they're using web data now, combining that with product data, and it's essentially just as powerful as a BI tool, but you can do way more deep dive analysis and like conversion rate funnels, like you can't do that in looker. I can't build, like, a funnel of my registration flow. Where are the drop offs on the steps? So I spent a lot more time in amplitude than than I do in looker. So it's cool to hear that like, yeah, Mixpanel almost plays that like, B roll, a bi role for your your team as well,

Simon Heaton  24:33  
absolutely. And it's a highly accessible tool as well. Like, I think once the setup is done, you know, you don't need to know SQL or be a data scientist yourself to then go in and make your own dashboard for the product or program you're running and to us at Buffer like that. Decentralization of data to the rest of the team, being able to arm the team of insights for the product they're owning or the program they're working on is really important to us, and mixed or really facilitates that. Like I said, Where redash Looker Tableau, you know, you really need either a data scientist or analyst or someone who's at least comfortable technically in SQL to model out some of those. It's not, as, you know, drag and drop in modular something like amplitude or Mixpanel. Yeah, yeah. Definitely. I

Philippe Gamache  25:16  
was looking at like, some of the looker competitors, and some of them have like on their comparison pages, like you don't need to know look ml like look ML is like their their SQL language for for building anything complex. This episode is brought to you by our friends at revenue hero. I can't think of anything worse than finding out a lead waited a week for a response from sales. That's why we recommend revenue hero. It's the easiest way to qualify leads based on Form Values or enriched data and route them to the right sales rep. Their product is packed with a bunch of behind the scenes superpowers that ensures qualified leads are assigned to the right reps, following your custom round robin rules and sending key data back to your CRM. That means more qualified meetings for your reps. We all know they want more of those, but more importantly, no more waiting time for your potential customers. They back all of this up with the best product support out there, offering 24, five support on Slack, connect for all customers No matter your pricing plan. So if you want to 3x your conversions with the same traffic, go to revenue hero.io and tell them we sent you. Your Sales Team will thank you for it. This episode is also brought to you by our friends at census, the number one data activation and reverse ETL platform loved by Activision, Canva, Sonos notion and more. As you might know, I'm pretty opinionated that the future of martech is composable, and that the single source of truth for your marketing data should be your data warehouse. Census helps marketers solve an age old marketing problem, getting real time complete access to your customer data without needing to write a line of code. Also, if you want your own face as a humans of martech style image, we're doing a fun monthly raffle with census for a personalized t shirt. Enter to win at getcensus.com/humans but you mentioned server side tracking at the start of our conversation, and like some of the comments on your post, on the tech stack, posts were just like, but Simon, where is Google Analytics on your list? You're not using GA? And as a marketer myself, who has not used GA for or Google Analytics in like three plus years, because we've been on this, this amplitude train, and the server side also, so I'm familiar with that kind of journey there. But maybe you can unpack the decision from from your perspective, moving buffers, web analytics to server side instead of client side, like tools like GA I'm assuming you're, like you said, just using a combo of Mixpanel plus redash to analyze all that stuff.

Simon Heaton  27:52  
Yeah? And honestly, I'm not, I'm not surprised that that sort of feedback, like, yeah, you know, you know it best as well. Like, Google Analytics is a monolith in the analytics game, right? Like, almost been for years. And, you know, 10 years ago, it was incredible, like when I was starting out in in my career, you know, the robustness of what was available then for marketing, you know, felt, you know, really incredible. It's free, it's easy to implement, more or less, it's pretty much the standard. So, you know, I'm not surprised that there's a lot of teams still relying on it. Buffer also used GA forward to primarily their web analytics for quite some time, actually, and actually, before I joined buffer, you know, they were still very much using it as the primary way to measure marketing performance and web performance. The problem that I see with Google Analytics and other client side tracking is, while they're easy to use and easy to implement, which I think is the benefit for a lot of teams, I don't find the data comprehensive enough or raw enough. Let's call it to use it in a more strategic way to inform, you know, data based decision making for marketing, you try to get more granular in Google Analytics, you run into sampling errors, right? And how can you, you know trust or exactly say how much you know your emotions drove if it's being sampled, right? And then you're just sort of guessing. And you know, you're also more restrictive into reporting options without going into full like custom reports as well. And I think the biggest limitation to me is you just can't as easily connect the marketing and web data to the product data. Sure, you could create conversion goals and events and that sort of thing and instrumented that way. But it's not the raw product analytics. It's almost like a proxy for those those events and actions? Yeah, and I think that's where tools like Mixpanel and amplitude really came in and and have created this, this ability for us to stitch those two together through server side tracking. So, so the type of work that we're doing and the experimentation we're doing is like, we're not really just looking at how many. Page loads? Can we get? How many signups can we get? We're tracking full funnel with a lot of the work. So we want to see, you know, are we acquiring signups for buffer that are actually doing something of value in the product? Are they converting? Are they activating, retaining that sort of thing? And so, you know, server side tracking, you know, really solves for most of this, I would say we're very fortunate at buffer that our product and marketing teams work super close together. Naturally. We're very collaborative, and actually like data reports in into me. So that's just saying something in that, you know, marketing isn't treated as like a second class data area that it can be in some businesses for whatever reasons. And so when I joined, you know, I started to see these limitations. We knew GA four was coming because I joined back in 2022 so that was sort of the year that transition was happening. And, you know, we really wanted to make this push to server side tracking. We had to set up Shopify before I left. And it was extremely valuable. So the team and I, we implemented this through through CDP segment. So which that segment captures all of our on site events, like page loads, we have button clicks instrumented, you know, whole bunch of really interesting events. Some of these are out of the box with segment like page loads, you really once it's like, loaded up onto the website segment, just capture those others. We've had to define and instrument ourselves like, you know, individual nomenclature for different buttons and CTAs on the site to track their effectiveness. One thing we did have to do as part of this was develop a couple models on our own so that we could measure sessions on the site in the same way, as, you know, we traditionally do in marketing. You know, we capture events on a page load level with segment, and that's sort of how mix panel showcases data as well. And so through some custom modeling, we built out like a session session model that strings together page load events into full sessions based on a bunch of different inputs, and that'll be a whole other conversation if I get into the intricacies of that, which allows us to really categorize groups of page loads together, map out full user sessions on the website, and then into the product itself. And then with that, we were able to build on top of it a custom attribution model to measure where are these users coming from? Are they coming from organic Are they coming from paid ads, etc? There's a ton of different signals and layers that go on to that. You know, we use the refer property. We use gclid to track all of our conversion ads through Google ads, our conversion, conversion events to Google ads. Sorry, so there's a bunch of different signals that layer into that, and then the front end experience is really powered by redash and mixed panel, right? So the team can go in there and they can really, you know, build reports for, you know, a blog or for an experiment, or for a particular launch that we have where we've introduced a new page, and then be able to tie all that back to, let's say, the launch objectives of adoption or retention for that product case.

Simon Heaton  33:02  
So it's been an incredible addition to our marketing stack, having server side tracking and like, like, what I would say for any team that's, you know, has the ability to implement something like this and move away from client side, I would say, go for it, because you know, you're going to end up with boundless amounts of insight and data that then you can model and use to ideate, to run tests, to do a whole bunch of, you know, really impactful data informed work that Google Analytics and these rituals I just, I just don't think they give you in the same way. 

Philippe Gamache  33:32  
Yeah, yeah. Definitely agree. I had to go down the the technical rabbit hole of server side versus client side when I was like, close, because we were doing part of the same that we saw. We were using segments, and we're comparing segment web activity data to GA and we just, like, couldn't or like, figure out the discrepancies between the two. And then, like, obviously, through time, we discovered that, like, most of the discrepancies were because of client side events firing versus server side events. So got really deep on there, but yeah, the refresher was a conversation we had with someone from meta router, Michelle, who is a Director of Product Marketing there. They're like a CDI tool, essentially. That's like making this easier for marketers and data team that are building this, like composable stack. So instead of maybe, like a CDP and a CDP housing all of that data in like their own kind of storage spot, this composable route is kind of like leverage your own warehouse, and so meta router, like, collects all of that data, pushes it to your warehouse, and you would use something like census to reverse ETL and connect all that stuff and throw it into customer IO or something. But I want to talk about one thing you mentioned that I thought was really fascinating, data reports into growth marketing. One of the things from my synth at a bigger company@wordpress.com was this, like internal versus third party tooling debate that we had all of the time. Time with a lot of resources. So our CEO and founder, Matt Mullen, was really big on this like dichotomy of like, Why pay a third party tool for something when we have really smart people internally who can potentially build something just as good and then we can go to market and sell that after also? So we have those battles all the time. And I know you mentioned that the the internal custom experimentation platform is something that's built out internally at Buffer, and we had the same thing@wordpress.com but it was supported by a pretty big team of data engineers and data scientists, and it was, it was really good. And like, when we had issues with it, like we almost had, like, a support team to kind of go to, or whatever. What I found shocking from knowing that buffered also does this internally, obviously, like the smaller team size, but like the salary benchmarking shows, like the team makeup, and I was just like, what, there's only one person on the data team at Buffer, who's supporting all of these things, what's, what's the story? They're really curious.

Simon Heaton  36:05  
Yeah, really good question. And I would say it's probably generous to call our in house platform a platform. It's maybe more of a methodology, you know, being able to to have, you know, server side tracking and everything in mixed annal, you know, it's really this exercise of how you tag the variants, and that in product, we are using something like split, which is the out of out of house platform, to do sort of product testing. But yeah, on our end, we do it all in house. Yeah, the data, one's really interesting question. I think, you know, one thing I really admire about buffer is, you know, we're a small team by design. We're a bootstrap company that, you know, at one point there was some investors, but the the team bought them out over time, and with that, you know, the goal has really been, you know, let's create a sustainable size company that doesn't have to continuously scale up, both in terms of growth and people. And I think that's a really, first of all, unique type of business in in tech. What this means for the composition of the team is that, you know, we're always in flux, I guess you could say so, you know, in some in some areas or seasons of buffer, you'll have, you know, greater investment in certain areas and teams might be a lot bigger than they are in the credit state and and vice versa. And I would say, you know, data today is is in a much different place than it was. I would say, you know, three years ago or five years ago at Buffer, I would say, for a few reasons. So one of the the big tenants that we have for the data team at Buffer is this concept of decentralization. I mentioned this earlier, the idea that the team, a team armed with data insights relevant to their area, whether it's a product manager, engineer in their little pod on their products, or it's, you know, content marketers on the area of the website, they're focusing on having that data really accessible and really be really intimate with that data can create a lot of level of ownership and a lot of impactful work for the team there. And so over the year, our data team has changed in size for a variety of reasons. Some people have left, some people have changed roles, that sort of thing. But during that transition, we've made the choice to keep keep it small for now and really focus on leveling up the data expertise throughout the org. So data engineering as an example, we had a couple data engineers in the past. How we're handling this now at Buffer is that the engineers and engineering managers within product pods are now responsible for the data engineering components for their product. So rather than data acting as like a service line, which it does in many organizations, hey, you guys figure out the issue related to the data pipeline. It's now on the onus is now on the engineering team itself to to make sure those are built properly. They're instrumented properly. There's a tracking plan. And each team, you know, has an individual or two within their engineering group that you know takes on that responsibility, and that, you know, we do have a more centralized team, the one that works with me that covers more of our horizontal use cases. So, you know, our data scientists and analysts largely does the majority of the modeling for the company where it's required, as well as, like we do a lot of deep growth analysis on, you know, why, why certain things are happening and and how we can improve them. And that's largely what this individual does. We have a separate central we call infrastructure team that is largely built up of senior engineers, and they also support horizontal data engineering requirements for the broader company. So there are, of course, one specific to individual products and use cases, but there are a lot of them in terms of managing our different DBS. But. Managing the quality and accuracy of the pipelines, the reverse, ETLs, all those things generally fall to that infrastructure engineering team. So when I say data reports to growth, marketing and growth more broadly, it's really sort of the data science and data analyst type components of that team that work deeply with us, while the data engineering side of the coin is really being propped up by the engineering work at the moment, very interesting.

Philippe Gamache  40:25  
I love that you guys are essentially not forcing, but, like highly encouraging engineering managers and senior engineers to expand into that data engineering world, if, if it's not something that that they've done yet, I find that really cool, because there is like, so much overlap in that world. Like I'm kind of like in growth operations on my current startup, and often play in this, like data product manager role, where I'm, like, translating marketing and growth requirements to a data person, but then I also need to do that for the engineer who is building out the custom registration flow. It would be really cool if those two were one person, because when we're making changes to the reg flow or in the product, that person is also thinking of the pipeline. And if we're gonna migrate the like calendar tool, we're doing it in a way that, like, the piping stays in place, and the tables stay in place, and all we're doing is changing the front end. So I think there's a lot of, like, long term also, like, personal future proofing advantages there for the engineers on staff. So I think that's really cool, yeah, for sure. And I think that's one of the advantages of a smaller organization like buffer as well as,

Simon Heaton  41:38  
you know, it's more of that, you know, startup type atmosphere. You know, everyone hands on deck. Everyone wears multiple hats, myself included. And big part of our culture is, you know, encouraging that personal and professional growth while you're at buffer. So, you know, we're always saying, how do we get, you know, 40, 50% better every year, individually, so that, you know, we can improve buffer in the same way.

Philippe Gamache  42:02  
Yeah, love that mindset. Maybe you didn't put too much thought into the order of the tech stack list when, when you shared this out. But I thought it was really interesting how you started off with pm tools, project management tools. You got notion in JIRA kind of listed on there. And in a separate post, he also shared insights on how you at Buffer run sprint cycles with your team. And I have, like, I've done this successfully at past companies when we were all bought in and we had a lot of collaborative nature, and all these other teams were using sprint models as well. We're just like, hey, look, let's, let's give this a shot for marketing too. But it's also failed spectacularly at other companies, where we were a bit more on the ops side, like a bit more supporting function, and it was just impossible to plan for, like, all the ad hoc shit that just like, came up out of nowhere, and it's like, where are we spending a full day every two weeks, sprint planning for like, shit we want to do next week when, like, we have no idea what's gonna, like, fall on our plate, and we're a startup, and we need to, like, react and move fast. So I'm really curious to ask you, in your opinion, like, the times that this has worked well for you, like, what do you think is the key to running successful agile or sprint, sprint models for marketing teams.

Simon Heaton  43:23  
Yeah, I'd say, you know, if you ask anyone who works with me, I am definitely a big process and documentation kind of guy. I always encourage our teams to stay super organized, keep our workflows, Chris, because I have a belief that, you know, if we're clear on how we do the things, then we're able to do them with a greater degree of focus and clarity and more impactful with, you know, tuning out our distractions. And you know, maybe, you know, that probably influenced me putting it first. But honestly, it was, it was really just a random selection. I think what I was putting together the post, honestly, I love the agile approach. I haven't always worked in this either, but I've become a big fan over time, you know, I've worked in, you know, the traditional annual planning cycle. You're working on monthly roadmaps, that sort of thing. That's pretty traditional in most marketing groups, I would say. But in my last few years at Shopify, we sort of pivoted the growth org to sprints, and it was really interesting to see the benefits that came out of that. And you know, there's a lot of parallels to what we're doing at Buffer, that happened there as well. One of the biggest ones, and I think you kind of touched on this, is alignment to the other teams. So especially in a product led growth company like buffer, there's no sales team that marketing supporting, there's no account management team. We're really that for the product team in many ways, and so being able to align our working and planning cadence with the different product groups is been extremely beneficial to make sure that, like our roadmaps are aligned our go to market motion and programs related to that can align to what product is doing outside of peer. Lead just, you know, launching new release features and things like that. And so that's been incredibly valuable for us. You know, we do support that entire go to market motion as a marketing team, and aligning those roadmaps make a ton of sense. You know, I do think outside of alignment for teams to be successful in adopting more of an agile workflow. So we do two week sprints, and our planning cadence is about every quarter. So we actually don't have annual plans at all. On marketing at Buffer, we have themes and goals and things we're trying to accomplish and targets and forecasts that span throughout the year, but the actual work we're planning to do gets planned on a quarterly basis, and our execution cycle is about two weeks. This allows us to have a couple things, you know, focus and clarity on our work. It allows us to be able to pivot as things come up, and so we're not necessarily tied to things we put on a roadmap that are coming up six months down the line, new information comes in, and we can easily reprioritize and do something different or go after something more impactful, but to be able to do that successfully, yeah, the team needs a couple things. One is clarity, so clarity on the goals and targets that we're doing. You know, you mentioned, there's a lot of distractions that can come up, especially for a marketing team, we have people pinging us probably every day saying, Oh, this word is misspelled, or button isn't working right anymore. A lot of that kind of stuff comes up. And, you know, we do have to get to it, but it allows us to give a sense of, you know, prioritization to those items. Some of those are urgent. Yeah, you know, you can't sign up because the button's not working. We want to make sure we address that immediately copy on one of the commercial pages on the website can get addressed. Eventually, it might not be the most pressing issue, at least in our mind. And so having clarity on goals of sort of what we're trying to achieve this cycle, what we're trying to achieve this quarter, can really help focus the team and keep that focus during that time, I would think, clarity on what we're trying to accomplish too. So it's really nice about two weeks or three weeks sprints, let's call them. Is that, you know, being time bound in such a clear way gives us a finite amount of time to reach a milestone with working, whether that's actually shipping an experiment, which we try to do once a cycle, or, you know, making progress on a more long term project, you know, being able to actually have, okay, these, these every two weeks, naturally can be marked towards milestone in our work, and having that defined, you know, time horizon to get the work done really helps focus the team towards progress versus, you know, letting some things drag on. Because we want to release it this quarter, you know, we want to release it this cycle, which is only two weeks as an example, to get that alignment and clarity. Though, you know, rituals are really important as well. So we we do grooming meetings just like a product team. We meet every Monday to align on the work in that week as well as the progress on tasks within the cycle. That's our full team. So it includes not just the marketers, but also our marketing counterparts. I call them that work on our team. So our engineers, UX designers, our data scientists, those individuals are able to join that meeting and input and share related to the work. I think team buy in is super important, too. So where I've seen this not work out is when some teams within marketing, or members of the team just don't want to lean into the process. They don't want to use JIRA or the task management tool. They want to just, you know, they don't want to be bound to two weeks maybe their work. No, we have, this case, the buffer some of our teams work don't fit neatly into two week cycles. Growth marketing, we're a little bit more project based, time experiment based, whereas, you know, our content marketers, they're producing, you know, X amount of content every every week. And that's sort of an always on initiative. It doesn't fit within the cycle. But it's not as clean as, let's say it could be for like, a project based, based initiative, so that buy in is really important. The team needs to adopt the process, believe in it, the tools that you're using, and I think, to me, the most important thing is, is actually on the name, it's being willing to be agile. I think the way it sometimes, I've seen people use agile as just a shorter form of the regular approach and planning, it's just Okay, let's just chunk this into smaller buckets. But we're still thinking really long term with our planning. We're still, you know, sticking to whatever we committed to at the start of the quarter. What I really like and what we try to, you know, encourage a buffer is a willingness to actually be agile with your work. What that means is be willing to pivot as things change, as information is discovered, towards things that will actually be impactful to your goals. So in many ways, we have this concept of like test and learn towards a big. Initiative within the growth marketing team, we might have this hypothesis that there's this big kind of program we can do that is going to drive tons of leads for the business, or whatever, and we try to test into that many times, early tests will start to disprove that hypothesis, and we're able to use that information to actually pivot away from what might not be, you know, the most impactful project, but could have resulted in a ton of sunk cost fallacy where we're saying, Hey, we've invested, you know, two months in here, we have to get this project at the gate, because we've done all this work, and we don't want to be for naught. We actually, you know, the Agile process actually allows us to say, Okay, we're looking at these iterations that we have released along the way, we're getting new information, and we can use that to, you know, pivot entirely or or make new decisions that allow us to to evolve what we're working on into something that's going to be more impactful for the business. I think that's the key to the to the value unlock for agile marketing structure is, is that actual willingness to say, Okay, let's be agile, let's pivot, and let's use this, like, you know, two week threshold, as that opportunity to do so, such

Philippe Gamache  51:09  
a great answer. Simon, there's, there's so many jumping off points there. I feel like this could be a second episode in of itself. A Yeah, I think there's a lot of key learnings there, especially, like, the point about just like, we're doing Agile, because agility is essentially, like, one of our key values here, and like, that's maybe one of the best ways to get people to buy into this. Because, yeah, I think you're right. Like, where this hasn't been successful for me at past companies was when we just had a process that wasn't agile, and a lot of people were just, like, using the tools they were familiar with, and like, we were creating a lot of change in their structure. So there was, like, a change management process that made this really tricky. But yeah, I think there's, there's a lot of wisdom in your answer there. I got two more questions for you. I'd be remiss if I didn't ask you a question of like, going from an enterprise, VC backed company to a bootstrap startup, like 18,000 person company at Shopify. I don't know how big it was when you left, but like 75 ish people, Bootstrap, profitable startup at Buffer. Now, complete opposite ends of the tech spectrum, right? Like, you've got two completely different sides of the coin. Walk us through the decision of, like, why buffer, obviously, like a big name in tech and cool company, but like, why not? Like, another big, you know, 1000 plus person company, since you, like, spent seven years in that mindset, like, building teams there. And what do you love more about the current setup at Buffer?

Simon Heaton  52:45  
Yeah, I've been, I've been asked this question a lot, actually, because it is, it is quite the shift. And personally, I love, I loved my time at Shopify. You know, I was there at a very exciting time for the business. Is a lot of scale. During that, that era, I joined back in 2015 and I think there were probably about 900 800 employees at that time, so still much larger than buffer. And then once I left, you know, seven years later, so, yeah, this company had scaled up to 14,000 I believe so, a massive difference over those seven years. And you know, you know, being along to that ride. In itself, was an experience, I'd say, to really, you know, be able to work at a business that's really on a hyper growth path, and be part of that, you know, journey of Shopify, growing up into, you know, the enterprise company we know today, and being along that, you know, in my own career as well, was really, was really awesome because, you know, it gave me a lot of opportunities to to work in a variety of different departments and teams and problem spaces within Shopify, which I'm very grateful for and and was really rewarding. But I think eventually, you know, at the end of a seven year stint, sometimes we just want to work on new problems, I guess. And I think it was just for me, it was my time to go try to solve similar problems in a different context. And I think for myself, that that was the personal growth opportunity like I was looking for at that time in my career of just, you know, let's go do this somewhere else and see, you know, if I can be successful in another business, I guess, and buffer. You know, I've known buffer for a long time now. You know, they're pretty well known in the social media space. You know, going back to my days in the agency, it was like the tool that we used to schedule and manage social media. So it's always been something I was aware of, and it's really one of those, you know, darling companies, I think, in tech, you know, it has a really well known reputation for its employee first culture. They're very innovative in a lot of, you know, cultural and operational ways, and that. You know, they were probably one of the first tech companies to be fully, fully remote, well before the pandemic and anything like that, which is, you know, really, really exciting. And I think, you know, one of the reasons that buffer had, you know, always been on a radar for me as a company that, like, you know, maybe I would love to go work for one day. And, you know, I joined at a time at Buffer was going through, you know, buffer itself was going through a bit of like a transition. And, you know, coming out of the pandemic, a lot of companies were, were feeling that the great resignation had people kind of leaving and changing roles and teams. And we saw that as Shopify as well. And I'm sure every company did. And because of that, when I joined in early 2022 at Buffer, the marketing team is actually going through a little bit of a reboot. A lot of you know, senior people had had taken the opportunity to go somewhere else and go on their next chapter. And for me, there was a really clear and unique opportunity to join an awesome team doing really cool cultural stuff that was very innovative, a team that has, you know, helped establish social media management as an area and a product offering in the industry, and to join at a time where they were working through their next era of business growth and marketing for buffer. And to me, you know, to do this at a company that is well established, that has product market fit, has a well known brand, but was still not a 14,000 person company, right? Was really appealing to me as a growth practitioner, you know, there was a lot of interesting opportunities from the problem space. Then I felt, hey, I can really help buffer, you know, evolve and grow in terms of the maturity of the growth and growth marketing areas. But then also, just part of me was looking for an opportunity to work with a smaller size team, if I'm being honest. You know, it's, it's one thing to work in growth and marketing and do so impactfully in a VC backed organization that has endless pockets. It's another thing to go and try to flex those muscles and do the same thing in a bootstrapped company that doesn't have the access to the same sort of capital as Shopify did. And that was just a really interesting problem and opportunity for me to work on, to grow professionally, to work with a brand that I had a lot of respect for, and to work within a new set of rules, I guess, and try to grow myself as a marketer. And so all of those things came together. It's been an incredible almost two and a half years now, because it feels like it was just yesterday that I joined. And honestly, all these reasons that I wanted to join are all the reasons that I'm loving what we do every day. I love being closer to the work again. I love the size of the team, the type of people and organization and our mission that we're pursuing, and just the people in the work has just been really great to get involved with.

Philippe Gamache  58:02  
So cool, so cool to hear the the journey, the the big change there. And you mentioned like being lucky that that you landed at Buffer. But I think it's definitely a two sided look there. I think they're, they're lucky to have landed a leader who has spent time at a company as prestigious as Shopify in like seven years, like seven years is like almost like two full careers these days in tech. So kudos to you. And yeah, I think buffers is lucky to have a hands on leader like you, and definitely need it in a bootstrap environment. Simon, really appreciate your time. I got one last question for you. We asked this to all of our guests. You're a director of growth, a mentor, professor, a traveler, hiker, golfer, gamer, weightlifter, gardener. You got a ton of stuff going on in your life. How do you remain happy and successful in your career? And how do you find balance between all the things you're working on while staying happy,

Simon Heaton  58:55  
I would say, working for an organization whose product you love and values are aligned with yours, and like the way they operate aligns with your personal convictions is absolutely fundamental to being happy in what you do. You spend, you know, eight plus hours every day, five days a week, you know, at this work and at this job, and so finding a place that really aligns with you, you know, makes all the difference. I'd also say for myself, so personally, I found there's, you know, someone once mentioned this to me in terms of, you know, I was looking for my next thoughts of career and growth opportunities before I even launched Shopify. And one thing I was told was always to avoid chasing titles. Avoid chasing, you know, brand names, and chase interesting problems, problems that you enjoy working on, that give you a degree of fulfillment, that challenge you intellectually can make going to work. You know the difference between hating your job and loving your job, right? If you. If you like, to work on tough problems. I'd also say the people you work with can be just as important as the work you're doing or what you work on. You know, I've worked in highly performant, high trust teams, and I've worked in low trust teams as well that have a little bit more chaos and a little bit more, you know, tumultuousness to them, and I would take a high trust team any day, over over that, regardless of what business and role I was in. And then, you know, I think balance is something that personally, I'm, I'm on a personal journey to have more balance in my life. In general, it can be hard when you're, you know, you're so invested and interested in, like, the professional space that you operate in, but for me, it's just really making space for that balance between work and play, knowing you know when you need to turn off and turn into the other parts of your life. You know your partners, your families, your hobbies, those types of things become super important to achieve that level of balance. And I'm lucky where I work at Buffer, which is the people first organization, who really, you know, values having strong balance for the employees. So we're really enabled to be able to pursue that in our daily day, like day today, but also, you know, outside of work. And so that's what I would say, if anyone listening, struggling with balance, just got to make the time. Find ways to split up your day. You know, I end work every day. I go straight to the gym, and so when I come back, my my work brain is turned off and my personal brain is turned on. Yeah.

Philippe Gamache  1:01:35  
Such great advice. Love, love the answer. There your first point about like picking an organization, the product you love, but also, like, the values are aligned with you. My wife and I were watching the Ashley Madison documentary on Netflix right now, and I was just like, how the hell do people who worked at that company, like, were able to, like, find any type of joy or happiness and selling that product? But yeah, Simon, really appreciate your time, man, this has been super fun. Time has flown by. I think this is going to be super valuable for folks that are really curious what what the heck is in buffer stack, stack, I think you did a great job walking us through that, and a ton of value sitting with you today. So really appreciate your time. Yeah, thanks

Simon Heaton  1:02:15  
for having me. Philippe, you,

Philippe Gamache  1:02:25  
folks, thank you so much for listening this far. We really appreciate you being here. I just wanted to call out two things before we go. Number one, the best way to support the show is by signing up for our newsletter on humans of martech.com we send you a quick email every Tuesday morning, letting you know what episode just dropped, we include our favorite takeaways, so if you don't have time to listen to that one, no pressure. We have you covered with some learnings anyway. And number two, proceeds from sponsors this year have allowed us to venture into video. We recently launched a YouTube channel where we publish full length episode. So if you want to see our radio faces, check that out. That's it for now. Really appreciate you listening again. Thank you so much. You.

Transcribed by https://otter.ai