Drill to Detail - Drill to Detail Ep.75 'Bootstrapping an Early-Stage SaaS Analytics Consultancy' with Special Guest Olivier Dupuis

Episode Date: November 26, 2019

Mark Rittman is joined in this specially-extended episode of Drill to Detail by Olivier Dupuis, founder of Lantrn Analytics to share his experiences growing a product analytics business using today’s modern, modular SaaS analytics toolsLantrns AnalyticsRittman Analytics"How it Works" - Lantrns Analytics"How Rittman Analytics does Analytics: Modern BI Stack Operational Analytics using Looker, Stitch, dbt and Google BigQuery" - Rittman Analytics Blog

Hello and welcome to Jewel to Detail and I'm your host, Mark Ripman. So I'm joined in this episode by Olivier Dupuy, founder of Lanterns Analytics, who like my own company, Rip Rippman Analytics, are an early-stage, bootstrapped specialist analytics consulting company, and who I've got to know quite well over the past few months as we operate and often collaborate on similar types of projects. So I've asked Olivier to come on the show to talk about what it's like to actually start and bootstrap an analytics consulting business today, and to compare notes and experiences with me on what works, where the opportunities are, and ways in which each of us are making it work. So Olivier, welcome to the show. And it's great to have you with us.

Hi, Mark. Thanks for having me. Yeah. Thank you. And luckily, I got your name pronounced it right in the end there. So I think after about two tries there, I've got that right. So you got it. Yeah.

So just tell us, Olivier, tell us about, I suppose, who you are and a little bit about Lanterns and also the route you had into this kind of business, really. Right. So it wasn't really a straightforward path. I did start out studying computer science. That was like at the end of the 90s. And I was doing web development afterwards. For a few years I got bored with it and then went back to school and studied in political science and figured out that there was no money in there. So at the same time in parallel I was still doing some stuff and doing programming on the side. So then when I graduated from university,

I worked at the university where I was studying for a few years there, mostly as a programmer analyst at first. And then I kind of moved forward in that. I ended up managing a team of, of programmers. And then, uh, I, there was like a big phase where we hired a bunch of business analysts and that was something really intriguing to me. I really got into it. That was something fascinating.

And I decided to, uh, to concentrate on that. And, and for the next few years afterwards, I was doing a bunch of special projects, just doing some reorgs around the university and running some special programs, such as we had one where it was is that we were studying the first year of study of our new students to make sure that they persisted in their studies. And that's where I really got into more the analytical side of business analysis. So we were tracking a bunch of data around their journeys throughout their first year of study. And at that time, I was already using a lot of Tableau stuff. And it was painful to structure the data at that time. But that was part of the challenge that I was facing at that time that I didn't know really how to deal with.

And that kind of ended up like being like almost 90% of what I do today. So anyway, so I did that for a few, few years and I, I really enjoyed that. And I decided that that was what I wanted to focus on. So really doing a bunch of analytics on user journeys and stuff like that. So at that time, it was only students, but I wanted to look at other types of journeys. So I went to work on my own. I was doing a bunch of freelancing, and I wanted to have a really different experience.

So I worked with all kinds of companies for the first maybe two, three years. But they always ended up having the same problem was that the data was always a mess. So people wanted to have Tableau right away and do some analytics on top of it. So I was doing Tableau and R and stuff like that. So R was fine because I

could do some cleanup beforehand, but it gets heavy whenever you have to start a project and do a bunch of cleanup data. So that's where I got into ETL and really trying to clean up the data beforehand. And then afterwards you could do more analysis. But ETL was a challenge because I was doing it in Python, and it was really complicated to handle that. And that's exactly at that same time that I ran into dbt, so data build tool. And that's really something that opened my eyes a lot because here I was like really like struggling with that messy data. And I didn't know exactly what was the best approach to that because all I wanted to do was doing the analysis, but there was always that problem with the quality of data. So by default, I had to deal with data architecture stuff.

And that wasn't really what I wanted to do or that wasn't necessarily my strength neither. But dbt made it so easy and the community around it is just really, really amazing. So I learned a bunch of stuff around that. And now there is an infrastructure that I really like to use for all my clients. And that's really how that packaging of service came about. So what we do, like Lanterns Analytics, what it does exactly is that we mostly work with

really early stage startups. So startups that have maybe five to 10 employees that have a product online, a digital product, and they've gone through, like they tried SaaS-based services such as Amplitude, Mixpanel and stuff like that. They want to go further than that. They want to improve their analytics. They already believe in analytics,

how it can drive their decisions and how it can help them improve their product. So this is really the kind of projects that we work on and again i mean it the stacks are always the same it's just that the service is on top of it this is where it differs a little bit from from one another that was a long introduction that was a long introduction wasn't it yeah that was good though i mean it's um so i came across you really because i i you know found your website

and in fact actually i came I'd noticed you before. I noticed your website before. And I think like a lot of us, you know, you certainly appeared to be a bigger organization than kind of like, you know, you are very early stage like myself. A very professional website as well. And you were talking I think you were talking particularly about about product analytics there as well. Yes. So so, so you know and so just for anybody that is new to this kind of area so um so you mentioned tools like amplitude and mix panel and so on there i mean what do they do that is maybe sort of different to what you'd get with a standard analytics tool and what do you do beyond what those products do right so mix panel Right. So Mixpanel, Amplitude, and well, I don't want to throw Google Analytics in there, but like kind of. So those tools, what they do is that they track user events on your product. So you configure your product to track specific events. So there are popular frameworks out there, for example, the pirate metrics.

So that just maps out the usual journey of a user throughout using your product. So their acquisition, activation, revenue, retention, and referral, something like that. And it's not in the right order. So what you end up doing is that you end up tracking specific events that are important in that journey. And mixed panel amplitude and others like that allows you to analyze

those journeys individually but also in aggregates. Now, that's fine. And I actually really enjoyed those tools. It really serves their purpose well. But there are limits to it. Whenever you have multiple sources of data, that's where you kind of start reaching a limit. And for example, if you want to go beyond only your user journey metrics

and you want to start measuring your product market fit, so there are frameworks for that also. So for example, Dan Olson's pyramid of the product market fit. So there are ways of measuring that, but it's not only based on the user journey. There are other stuff that goes into the mix. And so you might be interested in grabbing data that comes from your acquisition channels also.

You might be interested in grabbing data from your help desk data or stuff like that. And your KPIs may be a bit more complex to calculate than just a specific event, the aggregate of a specific event. It might be a ratio or something like that. And that's where you start ending up doing analytics that goes a little bit beyond those tools. And it does kind of limit you in terms of what you can do with within

those those tools the reason why is that you don't own the data the data is within their own platform so grabbing an extract of that data and doing your own analysis can be painful and can be expensive. So this is kind of what I end up preaching is that, all right, it's fine if you want to use that at first and continue using it afterwards, but maybe you should consider owning your data by using data collectors, such as segment that gives you a lot of flexibility afterwards to do your analysis in multiple SaaS based services or or even in your own BI tool like Tableau, Looker, Mode, whatever. So I mean so originally we were gonna we were gonna um

we were gonna talk about um I suppose product analytics and um and what they are and so on in this episode but as we got talking we were and so on in this episode. But as we got talking, we were talking, planning out this episode, it became clear to both of us that actually what would be an even more interesting topic is to talk about starting a business in this area. And I suppose consulting, the business of consulting in the market today. And as context, really, I mean, I know you came from, you work at university and this is your first kind of business in this area for me this is actually the second one i've done so i i you know co-founded a um a a i suppose a more traditional kind of in that case oracle focused consultancy um about 10 years ago now actually more than 10 years ago ran that for so i was cto

for sort of seven years and so on and certainly certainly that was a different type of, you know, the business environment was different then. The type of consulting, the type of customers we have are different now. But there was also opportunities now that weren't there then at the time. And I thought it'd be interesting, you know, I suppose to talk through with you, you know, what I suppose your motivations about starting a consulting business, you know, how are you making it work? And I suppose some of the challenges really, you know, particularly bootstrapping it where, you know, I think both you and I are similar in that we started these companies on our own and that brings opportunities and challenges really. So let me just start with a straight question to you.

You know, why did you go and start a consulting business? Why didn't you do what a lot of people do and go and work for maybe one of the kind of product companies with VC money and everyone's going to with beanbags and all that kind of stuff? Why did you start a consultancy business? That's a good question. I mean, for me, it wasn't really the other option. Option B wasn't really going to work for one of those companies that you mentioned. For me, like in the region where I live, public sector is really big. Everybody has like a really nice job with great conditions. And that would have been like plan A for me. So that should have been like what I've been what i what i do like for for

for the rest of my career sorry my verb tenses are are wrong sometimes so you've obviously got you got a comedy french accent as i mentioned to you and um and so tell us just in case anyone's clear whereabouts do you work where are you based based, actually? Yeah, yeah, yeah. All right. So I am in the province of Quebec. I am two hours away from Montreal. But specifically, I live in the capital region. So it's close to Ottawa. And now people don't know Ottawa. They know Toronto.

They know Montreal. They know Vancouver. But Ottawa is still the capital of Canada. And that's the region where I live in. And of course, like public sector, the federal government is the biggest employer here. And there's a large ecosystem of other employers that kind of lives around that. And there are a bunch of universities. One of them was the one where I was working at.

And so I think that part of the answer to your question is exactly there. There was a lot of opportunities. But the fact that like at the university, it was such a fun environment because I had the opportunity to work on so many different projects. And I wanted to continue with that. So I did want to have fresh projects, fresh challenges. And for me, like going and start my own consultancy, like it wasn't something really I would have thought of, except that I had a bunch of friends that were really into that also. And they kind of guided me towards that. And so I decided, why not? And then I started as freelancing just for the fun of working on different challenges.

And that's kind of how I got around to it. Now, there is a difference between working as a freelancer and having its own consultancy. And for me, freelancing was really about, so just tell me what you want and then I'll make it happen. Now, consultancy is different because it's more enjoyable for me where I tend to work longer term with my clients. And this is really where they end up trusting me with understanding their business model, where they want to go with their business. And they do understand the power of analytics, but they want someone they can rely on to really help them set up the infrastructure and guide them towards really leveraging data to make their business better. So there is a lot of that in working as a consultancy is that you really have to have knowledge and be invested in your client's business and really trying to provide solutions for them that drives their business further.

How about you, though? I really enjoy consulting. You know, I enjoy building a team. I enjoy building a company. You know, I enjoy building a team. I enjoy building a company. You know, I did it before and I sold my share at a certain point. But I hadn't really given up on consulting. So I went to work in product for a while. And product actually, and particularly working for a company I worked for at the time, Qubit, you know, introduced me to a whole new world of customer types and business problems and use cases and so on that I'd not encountered before.

So all the stuff you've been talking about around product analytics and customer analytics and so on, you know, I think the kind of dirty secret really of a lot of the consulting we used to do, I used to do before when it was in the Oracle world was most of it was actually trying to get things to work in the first place. was trying to install it was trying to install software it was trying to work out how to get this bit of middleware to connect to that bit of middleware and actually the amount of analytics you did at the time was pretty minimal um you know it was mainly installing software and understanding understanding how it all fitted together so you know the thought of going out and actually doing analytics consulting for customers, and particularly this market we work in where it's typically

it's startups, it's digital businesses, businesses that are very much driven by data, it really interested me. But to answer your question about why I started the other consultancy, it's kind of like, it's what you do. It's in your blood. And I really enjoy doing it. But, you know, certainly there were options out there, you know, working for product companies. I mean, at the moment, particularly, you know, with market ease, with VC funding and so on, you know, there's a lot of money going around those companies. There's a lot of potentially there's very high salaries that are there,

whether it will last, I don't know, but there's high salaries there. You know, or there's actually the option and there's the option of actually, you know, working as a contractor, as we call it over here, you know the or there's actually the option and there's the option of actually you know working as a as a contractor as we call it over here you know and i think that there is a lot of kind of um there's a lot of side benefits in that in doing that but for me i i enjoy working with lots of customers but i particularly enjoy building a team and and that for me is something that you know that's the thing that's the big driver for starting another business um but you know i mean what i mean is the team thing something for you or where do you where do you sit in that area i i do have an ideal about it but i feel that i'm still far from it so in my past life when i was working at the university uh i had like the opportunity of

working in multiple teams and multiple sizes but there was one team size that had like the opportunity of working in multiple teams and multiple sizes but there was one team size that was like the ideal size for me and and this is really the one that i want to reach again uh like you you do have to understand that for me like that that whole business it it's end goal is really to um like my end goal is really to, um, like my, my end goal is not to sell the business at some point or anything like that. It's really to have like, uh, fun projects to work on and work with,

uh, with multiple people that I enjoy working with and really just have something that is enjoyable to do. So, and that, that includes having the ideal, uh, uh, team size. And for me, it was like three to five people and that's it. So you used to still be, end up being really agile. You end up, uh, having like enough people to work with that, that, um, like for example, we did talk about that in a previous discussion, but, uh,

like in three weeks time, I'm, I'm going on vacation for two weeks. So I'm gone. I'm, I'm going to Vietnam for two weeks and I'm not reachable at all. So this is complicated to manage. Uh, I do have clients. They are understanding. They do know about it. But I would really prefer to have other people working with me to be able to, for me to leave without worrying about those projects and really have those projects continuing moving forward. So, yeah, there is an ideal team size for me.

I would like to build to to build that team uh but for now it's still too early for me uh yeah sorry well i mean well so i mean i mean yes i mean just to say on that i mean i think that numbers are interesting and i think that um i mean one of the things i think that you and i um talked about at the start was actually the challenge of going from one to many, you know, one to more than one person. And so I think we discussed this before when we were talking about, you know, that, or certainly I found that with just yourself doing things, you know, it's hard to get from the point where you can actually kind of rise above that and start to go to, you know, start to go to two people because you're always working all the time and you're always kind of delivering stuff. I mean, I think I've got past that point

now, but just, just maybe I'll just tell us about why that's a challenge really, when you're starting a business to go beyond just yourself. Well, I think you're living it right now. No, the fact that, uh, well, you do have to have like that critical, critical, I don't know how to say that in English, but the critical mass of clients and revenues that comes in where you feel confident enough to have someone come in and invest in that person and really trust that person to be like that other that, that, that other face of your business that you, you're building. And, and that is, that is really hard. I, I, like, I did try that at one point, it didn't work out. Uh, and it was due to mistakes I've done on my side most for most of it uh but so i for now i i'm kind of uh um uh not scared but uh there is like that part where i think i could bring in someone else but i i don't feel confident enough yet to do it uh because it's still too early and and as you say, it means that you need to really pick up the pace in terms of your work output for the time that you train that person towards being effective, not effective, but being able to kind of replace you on some of the projects. And the other aspect of it is that that means you need to be more on the selling side on the managing side of

things and like although i do enjoy part of it like what i do enjoy most is of course working on the projects themselves i guess everybody does but um so like it's a fine balance but yeah how about you, though? Because I know, I think that you, there is one or two other people working with you right now. So how's that going? Yeah, so I mean, okay, so going back in time. So the original, with my original business, we took the approach of hiring the best people. So we hired, what I mean by that is very experienced consultants who typically, actually, I met on the kind of the speaking people. So we, we, we hired, what I mean by that is very experienced consultants who typically actually I

met on the, on the, on the, on the consult, on the kind of the speaking circuit. So I would speak at a lot of Oracle events, for example, about analytics and we get to know other speakers there and they would come and join us. And we were, we are, we were the, we were the pinnacle of the industry where if you, if you were, you know, the peak of your career was to work as a consultant in my old company um and and that's good but the problem with that is is there's there's to be blunt it's not a very good business it's not a very kind of um profitable business and the reason

that's an issue obviously is because like any you know like any business it goes in cycles and there are times when things go quiet there are times when you need to have some reserves you need to have a bit of kind of margin there and a business that that is always paying its team only fractionally less actually in your earning and you've got a lot of people that aren't actually out there billing um then then that can get quite interesting you know so with the new business you know we i took the approach of thinking well we want to develop a team we want to develop um our own staff we want to go and hire people who aren't necessarily working with tools like looker now but have that talent and ability and interest such that we can train them up you know um and and and that to me is is the kind of yeah that's the strategy i think we need to do to have the quality you want um and and also

to have a business that makes sense really you. Now, what I was curious about was that are the people working with you all locally in the same region as you? Or are you working with them remote based? Yeah, so a combination, really. So a couple of people work with us remotely. People I've known actually from other companies I've worked at. So product people and engineers and so on work with me um and um so they're remote um but actually you know we're trying to recruit um locally if we can so I live down in Brighton I live down in Brighton which is about an hour from London but it's sufficiently far away that there are people who would rather you know not

be traveling to London every day and so are talented and know their stuff and are very kind of um you know keen to learn but the fact that we're based in Bright their stuff and are very kind of, you know, keen to learn. But the fact that we're based in Brighton and we go to London occasionally, but not every day, that's a kind of a benefit really. So, you know, there are people out there we're doing that with. But what it means, though, is you've got to kind of you've got to think quite carefully about training, you know, because what you're not going to find is someone who even in the first you know few weeks or even months is going to be able to replace what you do so you've got to think about it quite carefully and and that creates a cost and it means it needs thinking about and and i think as people like you and i who probably in our if we had a fair spare hour we'd rather go and learn a new product than write down a process. And so it takes quite a lot

of discipline, I think, to think about that and actually maybe kind of, you know, a couple of kind of goes in it before you get it right. Right. So what are you aiming in terms of team size? Well, that's interesting as well. So I think like you, I think there's a, you know, I would like to be in the position of having, say, you know five people working with us full-time within, say, by, I don't know, sort of like spring next year. So at the moment, we've got three. I want to get to kind of five. And there's always a balance there between, you know, going too fast and therefore maybe kind of having an issue with with revenue or just kind of quality is another thing because you're trying to do too many things at once

but but it's interesting you talk about three to five because that that's good and i get that but you know there is a level beyond that really when i think yeah once you start you mentioned about sales and i think it's really important that you and i do sales now because you've got to get in there you've got to do you've got to think about the the what is the financial side of any kind of project you do and you've got to think about what is the actual benefit to the person that you're working with yeah and that's why i think it's important that we know we do sales but also we're probably not very good sales people and at a certain point at a certain point you want to hire someone in but that then means that if you hire a

salesperson in they're going to bring in revenue over here in pounds or maybe a million a year they're going to support six consultants so you need to get to that right you need you need to be about 10 then but really that is the level at which that's the kind of minimum size really you can be at to work with some of the bigger customers the enterprises and so on and that is you know that is in terms of revenue and longevity of projects and so on that knocks the kind of customers that you and i are working with now out of the out of the field you know in terms of business potential so there's a lot in that conversation what do you think about that yeah i mean the so what i'm hearing most like or

not only what i'm hearing but i guess what I'm hearing is mostly my concern about it. And my biggest concern about it is really the type of role that it leads you to. Like I mentioned, I do really enjoy working directly with clients on the projects and doing the technical work and all of that stuff. I do enjoy building the business, of course, that is really fun. And as you say, of course, I am not a great salesperson. But at the same time, I do enjoy it because I think it's important, right? So I'm working with someone that is really good with that and

is is teaching me a lot of stuff that is really interesting like really understanding the struggle that you're trying to to to solve and how you want to um to package your services to to kind of um help those people move forward with you. Not selling the Cadillac right from the start, but how to gradually get them on board with you. If I'm going to interrupt you there, actually, that is kind of something that I think you've done really well. So just to kind of focus on this, actually.

So something that I think you've done well on your company website and the way you do things is you've sort of like productized your services and you've made it you make it I think you know easier for someone to buy what you're doing and the reason I say that is because one of the things that really kind of was a bit of a sort of a shocker for me was I was a I was exhibiting at a looker event looker join event in in London and one thing that struck me was most of the people that came to my stand to talk to us didn't actually know what a consultancy was and didn't know or just didn't even kind of think about the idea of hiring in services to help with their looker implementation and I think part of that is because you know where the products go these days it's you know it's it's self-service it's sass it's whatever

and I think you've done a pretty good job of doing of actually making your services in into a product so tell us about that what what do you what do you what are the sort of packages you do and how have you done that right so i mean there's two things to that so the first one i should say is that there is a company out there that i i it's not that i know them and i've not used them specifically but i really like their business model. And they're called Pilot. Pilot.com. I don't know if you know about them, but it's a pretty simple service.

What they do is that they do your bookkeeping. It's only for QuickBooks. And so they automate a bunch of stuff on it. But it's also a personalized service. So you start off working with them and they automate most, like 90% of all of the work that they do for you. And then the remaining 10% is for really custom needs that you might have. that's kind of how I figured that I wanted to approach that is that, so there is like a specific need that I think that most of the clients that I work with have, which is that they need to organize their data first. And they end up needing,

like the reason why they want to organize their data is because they want to have insights into how their product is doing, so the health of the product, their users, how they're behaving, and all of that stuff. But it's always similar. The way that you build that was that it's based on the concept of a value ladder. So it's really all right. So you're starting out here. You're not really familiar with product analytics and you end up on our website. But you don't want to commit yourself to any kind of like monthly retainer right away. So how do you, how can we help you figure what product analytics and how we can help you and slowly build that relationship that we do it's a weekly thing that we just send like four stories really simple stuff that that just gives you like an idea of what product analytics is and it's my take on what happens throughout the week and that's it so that's like a first level there is like uh there

is a second level which is a a guide that uh we wrote on on subject, and it's always evolving. So it's based on the projects that I work on and kind of my thoughts on how to do product analytics. And the idea behind that is if someone wants to do it themselves, then go ahead. This is our approach to it. Like, take whatever you want in there and and move forward with that now if they want to start working with us then this is where we start introducing the paid packages

uh so the first one which is not on the website now but it's we're working on it it's just like a simple phone call uh just to have like a bit of insights on what their challenges are and how we think that they might be what they should do to to kind of improve their infrastructure and how to better use analytics beyond that then there is a core package the core package is really all right so we have that bunch of data we want to organize it so this is where really what we end up doing is that we grab data from different sources for example using segment we throw that into a data warehouse we build a dbt or an etl on top of it and we transform that towards a data warehouse and then you can start consuming that data in whichever service you want.

So again, Tableau, Looker, R, Mode, whatever. If you want to go further than that, then we have more like custom services. So this is really more like a retainer service. So we start working more closely with the clients. So there is that concept of value ladder where we want people to feel like there is a way for them to move forward with us without having to spend the thousands of dollars that they might not have at this at this moment and they they don't really understand the value yet of what what can be accomplished with with product analytics so so that was kind of the idea behind all of that okay okay and okay and you mentioned you may okay so you mentioned retainers as well i mean one of the i suppose one of the problems

that bedevils i think all early consultanciesancies really is this thing where we're all doing a project here, a project there. And, you know, and you often, it's often feast and famine because when you're working on a project, you can't go out and win new ones. And there's also this, I suppose, we all kind of aim for this holy grail thing of kind of retainers or things where a customer commits to a long-term relationship and lets you kind of like build a relationship, you can maybe do some recruiting based off of that. Right. How have you found the challenge of going from spot projects to this retainer thing? And has it worked for you?

Or what's the kind of thinking behind that? Well, I don't know why. I guess I've been really lucky, but almost from the start, it has always been retainer clients that I've worked with. So, of course, there has been smaller projects here and there, but I've always been fortunate to always have two to three clients that I work monthly with, and that gives me like really a lot of stability in my revenues, of course,

to experiment with stuff and try other things. And yeah, so I don't know. I mean, it's not as if I had to move from one model to another. It's been, I've been really lucky, said so yeah i don't know i don't know exactly how that happened but yeah how about you though is it mostly retainers or is it mostly well it's been um so again one of the things i i learned from the work i did at qubit and particularly from e-commerce and and the analysis we were doing was the value of retaining customers

you know and the the cost in getting new customers and the analysis we were doing was the value of retaining customers, you know, and the cost in getting new customers and the value in retaining customers, you know, understanding their needs and, you know, building that relationship and, you know, turning, I suppose, one-time customers into repeat buyers and so on. And that, you know, that's a classic e-commerce strategy, you know, but it's also, I think, a great strategy for building a business. And one of the things I think I've been very kind of, you know, focused on and pleased about is that every customer that we have has been a reference, which has been great. And also we make a real, real effort to retain those customers. And so that by definition kind of makes, you know, makes customers repeat customers. Although it's not always month on month every month.

It can be maybe a gap of a few months and then come back. But certainly, you know, I was the CTO in my previous consultancy and I could focus really almost entirely on the technical work we're doing. And I think one thing I've tried to do very much in this new business is be a lot more of understanding of what the business problems are that we're trying to solve. Because I think that is the key to it and you know maybe to you again question to you is you know how do you how do you try and make sure that you understand what the problem is you're trying to solve at the end and how do you i suppose differentiate what you're doing from every other service out there that does the same sort of thing how do you particularly win customers and work out what the problem is and so on?

Yeah, that is definitely the biggest challenge, I would say. It's because I don't work in the product team. I don't work within the business. of being directly involved in product teams and working closely with a bunch of multiple product owners and understanding how they go about making their decisions. And I think that the service that we're offering is really about that. So we want to help them make better decisions by leveraging data. So understanding the struggle themselves is difficult because, again, like I'm not doing direct observations of those product owners.

Now that is kind of really different than when I was working at the university where we were observing closely the students and then we were doing a bunch of weekly interviews with them. And so we add like access to the raw situation and the raw struggles. And then we could understand better with the data and then come up with solutions. Now, that's not exactly how it works here. So for me, that is the biggest challenge. So how do I get the raw information from product owners? How do I get the conversations going with them

and really understanding their struggles? It is not easy at all. So I don't know. Okay, interesting, interesting. I mean, for me, I think I was lucky in that having worked in a product company for a couple of years before starting the second business.

You know, I was a product manager. You know, I was by no means the best product manager ever, but I certainly observed what was involved in being a product company, building a platform, and that was useful. And I think the other thing is that I think it never does any harm to go on site and meet people. And I think that one of the things, I'm not saying you don't do this,

but certainly for me, it's very easy. You can do this business remotely. You can deliver consulting. You can do Hangouts. You can do Zoom. You can do whatever. But there's nothing beats going on site and spending time with people, really. And that for me, and also, you know, the meetups we run and just generally getting out of, you know, your, you know, your kind of comfort zone, really going to see and meet customers is kind of important.

But as you say, I know it's, it is a challenge, really, you know, in there. Another topic I want to talk to you about, actually, just moving on, you know know it's it's actually running your own business okay so there's a phrase in in england um we say you know dentists have the worst teeth and you know hairdressers up hairdressers have the worst hair and so on because oh yeah we have one in french also yeah yes because part of part of it is you can't you can't be a dentist on your own teeth but also you tend to neglect your own business or certainly i mean so so how have you gone about, I suppose, running your business in an efficient way, given that you are the main delivery person, the main salesperson, the main ops person? How have you done that? Wow, I don't even know if it's efficient right now.

But I mean, I guess that it is. At the end of the year, everything is working out. I was able to pay myself and the collaborators, and then I still have enough money in the bank to start the next year with enough leverage to be comfortable experimenting with some projects and still have those clients that are retainers and that will be continuing kind of working with us. So, I mean, it is a small business. I've been running it for the last five years in terms that it was like freelancing business at first. And I started out by really,

I did invest in working with professionals that are really good. For example, an accountant. And we have that concept of not tired in here, which is a lawyer, financial lawyer, or something like that. But in Quebec, it's a little bit different. So those people are working closely with me just so that the business is is is well structured is

is healthy and but in terms of daily operations it's really a balance for me the the number one priority is always the client projects. So I do make my rounds speaking with clients almost every day. And as you say, I mean, I do work remotely with them. So they're based pretty much everywhere. I have clients in California and I have ones in London, actually. So they are pretty spread. So that means it's a bit complicated to touch base with them every day. But that is key for me to do that.

But yeah, I mean, besides that, I wouldn't say that there is any. Okay. How about you, though? So I took quite a different approach really so so you know having having i suppose been co-owner of a business that had 150 staff at one point you know i certainly was aware of how what the systems look like at that kind of scale and i'm also was aware of what it what you know i was aware of what happened when you didn't have those things in place on day one because we were new to it at the time.

So I think, for example, we didn't do management accounts for many years. We didn't have kind of an integrated platform for many years because you learn as you go along. But what I did this time around was I had the luxury of a little bit of time before I started this new business where I could get things set up. So I kind of built out the infrastructure and the company structure to scale, you know, to be of the scale that can easily get to sort of 20 people, really. So, you know, we've got, we've got, you know, back end CRM systems, accounting systems, delivery systems, all that kind of stuff in place. So such, you know, things like a service desk and so on, they're all integrated in together. you know and the data as you know all gets landed into say big query or do analytics on it and i'm also hired in someone as well on up for ops early

on so i just generally i want you to get to the point where watch words were first of all you know have have the room to scale up you know don't have to kind of throw it all out and start again in a bit of time um watch watch the cost of it because you could obviously spend a fortune on this, but with SaaS software these days, it's not so much. Next bit was to make it as efficient as possible. So it's little paperwork and nothing is printed out. Nothing is like that. It's all kind of electronic and it's also all linked together with things like Zapier

and workflows and pipelines and so on. But I kind of thought to myself, let's get that scale right at the start and let and pipelines and so on. But I kind of thought to myself, let's get it, let's get that scale right at the start and let's make it so that the, the business operations are as efficient to run as possible such that we can then, you know, so I can then concentrate on delivery and sales really in this first part, because if I spend all my, I remember back then we used to spend hours just

printing out things like, you know, invoices and photocopying things and whatever. And, and, and, you know, for me it was important to get that right at the start, really. So, yeah. I wish I had like a better planning story than I have. I mean, it really was always hacking something together as things evolved. And I did have a look at your infrastructure based on the blog post that you published. And I was impressed about it. But for example, for me, there are a bunch of stuff that I do that is so inefficient that I know I could do better.

But it all started out working with Gsheets, right? And then afterwards, you start, all right, you get a little bit more fancier, like Trello and now Asana and stuff like that. And then there's QuickBook in there. But for example, just sending my invoices, it's such a complicated and inefficient process that it's laughable. And I know there must be a better way of doing that. So I totally agree with you that the end goal should be for me to figure that out. Because like the objective is to scale, right? Yeah. And also the data, the data is interesting as well. And for me, you know, one of the first things I did was to land all the data for all these systems into BigQuery, then use dbt as we talk about, and then actually understand what

the data is telling you. Because for us, data is what we do. And there are so many insights in our own data to understand about patterns. One thing I did actually that was very interesting at the start, what I thought was interesting was there's kind of projects that I worked done for customers, one of which, one of the projects I had, you know, in terms of amount of days that I'd spent on the project was kind of maybe a medium-sized project, but the customer was incredibly, you know, this is fantastic. It was just everything was fantastic on the project the other project i did was one where um where i put in a huge amount of effort

uh and you know you apps i absolutely knocked it out of the park in terms of kind of effort and so on but the customer was always a little bit kind of like you know um ambivalent and it's because it was because the um and looking at the data behind it, you know, we export all the data from Jira and so on into kind of like into BigQuery and look at it. And it was because of the expectations we set at the start. And actually it was because in this case, I let the customer drive what we were doing and the rate at which we were doing it. And it was clearly unrealistic. And it meant that things, you know, a lot of things got done maybe two-thirds all the way through and and but so i suppose having the data on that and being able to look back at it and think well what does that tell you about your own business i found that very

useful really um so but yeah it is a little bit of kind of um i suppose i had time at the start to do it but also awareness of of what you need to get in place later on and wishing you'd sort those things out at the start. Right. And this is kind of the lessons that you learn as you move forward. For me, it was, all right, so how do I acquire clients? How do I structure my website? How do I get insights into how people are using my website? And now the operations is something that I've been trying to focus on a little bit more in the

last few months, like standard operational procedures and stuff like that. I'm trying to document all of that stuff because I'm thinking that someone else might be coming in at one point. But it is a challenge because it's not something sexy to do. It's not something that is really interesting that you just want to do it first thing in the morning. But it is something that is a big undertaking. And so it would have been nice to have the luxury, as you say, to start things off and have a big plan on how you want to run your operations. It wasn't the case. But I think that it's true also, especially in my case, your expression of, in England, it's dentists who have bad teeth

or something like that. In French, it is cordonnier malchaussé, which means the shoemaker who doesn't have nice shoes or something like that. So it is a bit true. There is a bit of truth in there. Everything is not all well structured on my side, but it does make sense. In the end, I do figure my stuff out i do rely on data i do have like the means of going and and and digging into that data but it's not always as best

structured as i wish it was so changing topics now so partners okay so so one of the i suppose a difference between this business i have now and the one i had before was that back in the days um before it was very much you would typically kind of partner with one company in our case it was oracle and oracle did so many things that you know you could really do you know virtually any aspect of it would be done by oracle and and and so you could do that and and but actually you know certainly something happened in the and the analytics world you know in the in the intervening years since that business a new one is this kind of i suppose this this almost like um uh what's the word disintegration of you know disintermediation of analytics and we now have a lot of companies in this market now question to you is really you know what is but and obviously there's a benefit potentially in

partnering with these companies um and so partnering you know it potentially means you collaborate on some sales opportunities. It means maybe you've got a bit of a closer relationship with the development team, for example, a partner team. But, you know, but there are lots of companies out there and some of them are, you know, I suppose there's a limit to what you can do. I mean, what's your strategy around partnering? Who do you choose to partner with?

And on what basis do you do that? Right. And I think that was kind of similar for me. For a while, I did consider joining the partnership program for Tableau. It would have been like a nice move in terms of getting projects that revolved around Tableau because I've been using that product for such a long while and I know the ins and outs of Tableau. But at the same time, it was an exact type of project that I wanted to run on to work on exclusively.

So I was always kind of feeling, I'm not sure about that one. And that's kind of where at that point that I stumbled upon DBT. And then there was a lot of blog posts from the Fishtown guys about the modern BI stack and everything like that. And that really caught my interest. And so as you say, there is like now an ecosystem of companies around that, not that they're coordinating themselves together, but there is like a bunch of services that makes up that modern BI stack, for example. So, of course, there's DBT, but there is Segment, there's Stitch Data, and now there's Snowflake. And all of those services are really interesting,

and you can use them interchangeably. But there are ones that I prefer working with. So the idea of joining partnership programs with them is, what is the benefit behind that and for me the question wasn't really about uh can i get more clients through those partnership programs currently like the one that i'm um um going through is the segment partnership program and and this is really an interesting one because there is now a bunch of products that, not a bunch of products,

there is the Personas product and there is the Protocols product that they've built on top of their core product. And to have access to it, you do have to pay for it. And it's really interesting additions extensions that they've they've created uh to their core offering so going through their their um their partnership program does allow us uh to have access to it and there is a lot also of thoughts that they've put into how to help your clients really structure your analytics. So there is a lot of stuff in there that I'm really learning from. And so there is value to it.

And it's the same thing with Fishtown Analytics. I did have that conversation with them two weeks ago or something like that. And I think that they're interested in leveraging that partnership to really help the consulting businesses that are in their ecosystem really learn more about DBT and how to really leverage DBT for their own projects. So for me, it's more about learning at this point i don't know if there is anything more to it that i'm thinking of right now but really having access to those products and really learning more about them how about you okay so i think it's i think partnering with vendors is is an important um is an important kind of, I suppose, sales channel.

And it's an important way to, I suppose, give a better service to customers because I think by partnering and going through their certifications, you have a certain level, there's a certain level of competence there, which I think is good. And I think it's also something that is very, if you're very mindful of, is that probably the time at which a customer most needs services is at the time when they first implement something so and so it's a bit like if you're a builder or you're a decorator or something you know when you want to when when you you want to

partner with kind of you know in our case estate agents or builders or whatever because that because it's when the house is being bought and people are moving in that's when there's opportunity and so I think that you know in in our world when there's opportunity. And so I think that, you know, in our world, when someone's buying a new Looker system or they're working with Snowflake and they're putting in Snowflake, for example, that is when there's an opportunity to do, you know, that's when they most need services. You know, so that's certainly the time when they're most receptive to that.

So the strategy I've always had is to partner, you know, initially the main partnership was with Looker, and I think they've been fantastic. And certainly, you know, there wasn't an established, a dominant Looker partner in the UK, and so there's an opportunity there for that. You know, the work I did with Qubit in the past meant that there was a kind of a critical mass of people, customers I knew who used it.

But I think also the thing to bear in mind is that, you know, a Looker project, an analytics project, is only going to be potentially a small part of a bigger thing going on. And that's why I think it's also, we also partner with Snowflake, for example, we still partner with Oracle, you know, and we're hopefully pretty soon getting a partnership with Google Cloud sorted out where, you out where in a typical kind of project that has Looker in there, there's usually a database migration

to the cloud or there's a new implementation of that. There's a lot more going on than just the analytics. And I think customers want you to be able to do more than just analytics, but also on projects we work on typically, maybe 20% of the revenue is analytics the rest of it is what people will call now data engineering so you know having those partnerships with with with vendors like snowflake for example is important but i think the other part of it is realizing why do they want you right and they don't want you these partners these vendors just

go to just go in there and pick up business they They want you to generate business for them. And that's the other part to it, really. But isn't it true, though, that that kind of defines your acquisition strategy is that people are working with your company first because they have like a specific technical challenge that they're encountering with whatever, for example, Looker. And they know you're the Looker expert in London, so they'll start working with you on that specific project. And then based on that afterwards,

you start building that relationship towards more longer-term contract or whatever type of work. Yeah, there's that. yes that and that certainly is the the route that we've you know that that is one way in there and and so what you go in as is as the you're the technical experts you go in there they've got a problem with they've they've maybe kind of implemented looker they've got maybe a problem in a certain area and i had one customer recently that was um they were using power by looker so they're using it embedded in their application and they just need as a bunch of issues around that and the performance of the

snowflake database they had that they wanted us to help with you know but actually you know really you know that that is you know the problem is with that once that's done that's done and you can build it out but it's you know really the thing to be doing is trying to be talking to the people you know the owner the founders the owners and on, getting involved at a work for example is you know look as motto or look as sales pitches you know we democratize analytics and we help you know you can scale analytics within your business and you can you can become data-driven okay and i think that the challenge though is what does that mean and how do we do that how do you actually make a business become data-driven and and if you can get involved in that kind of work

and if you can truly make a difference for customers, you know, in helping them actually become more data-driven and truly democratize analytics, that's where, you know, on both sides it becomes really valuable, really. And if you can tell a story around that and do that repeatedly, that is kind of like, you know, that is really good. But, you know, it's hard to sell that as a thing. Right, right, right.

And that's where, you know, that's where our strategy for the first year was just build a reputation. You know, build a reputation, you know, get people to know about you and focus on this. What is the end business goal for them in hiring you in? Why did they buy Looker? Why did they buy, you why did they buy you know snowflake it must have been to do some change in their business so it always assumes though that they are pretty far along in their their analytics journey right because they've already invested in looker or in snowflake and they're encountering like challenges and that's where they kind of start working with

you uh because they've got to the point that potentially i mean so so just interrupt you i mean what i mean one of our customers is you know it was pre-series a we got caught we were brought in so we were brought in really by the founder um and and you know for the first year or so really we were mainly just working with that founder. And they've got their Series A funding. That's great. And now we're working with them to really kind of drive the business with this. So I think one lesson I learned was, you know, tools like Looker, they are a better fit for companies that are post-Series A because of the cost of it and so on. But actually, you know, you can get involved in earlier companies and you can help them on this journey. But, you know, I think that it's, you know, it's harder to do that. But certainly,

you know, if you can be a part of the team that got them to that series A, series B, you know, maybe a kind of like an IPO thing. That's really interesting, really. Yeah. And I think that's part of the idea that is behind the weekly newsletter that we're running and the guide that we've published and all of that stuff is that I think the goal here is that we would like to start working with those product owners before they end up running into those, like having invested in Looker and then run into problems and then we fix it and then we start building a relationship with them. Probably it's a bit idealistic to try to grab those product owners

before they've reached the wall uh but but yeah i think in the end that we're trying to reach for the same kind of uh of not mandates but relationship with those with those clients so so to round up then i mean so yeah obviously you know you mentioned that you're you're not building this to sell it you're you're looking to kind of build a company where it's you do fun projects and so on but but you know where i mean where do you i mean i feel like i'm interviewing you for a job here but um where do you see yourself where do you see yourself in five years time i mean is it is that you know is it could you you know is this something that is you know what we call a lifestyle business where yeah you know or is it something where

you're thinking it would be nice to maybe kind of cash out and do something different and whatever and buy a van selling those chips with gravy and cheese on it that you have over there or something? Exactly. But what's your five-year plan? What's your kind of thoughts on this? First off, I have to say that it's been a while I've heard that question. What's your five-year plan? But yeah, I mean, to be honest, it is like a lifestyle business. I have three kids. My girlfriend has three kids. We're pretty busy. We like to do a bunch of stuff like we practice sports and all of that. So there is a bunch of other stuff that's going on in my life. So for me, that business is really about really finding clients that I like their business model and they want to invest in analytics and that they're ready to work multi-years with us. Not multi-years, but for a long run with us,

that they really want to partner with us and build their analytical capabilities with us. So for me, that is really like the holy grail of Lantern Analytics. So in five years, hopefully everything, like the business model will be more clear. The acquisition channels will be clearer also. Maybe there will be like three to five people in that team of mine. situation in five years. How about you though? Like, I'm curious though, because it's not exactly the same angle that we have both of us, right? Well, I mean, I think, you know, I suppose you can't really, I mean, I think, I suppose it's more about kind of guidelines for how we do things. So

first of all, I think if you've got a business where you look after staff, you look after customers and you watch the margins, then by definition, you've got a business where you look after staff, you look after the customers and you watch the margins, then, then you, by definition, you've got a good business. And I think if you put your, if you put your time into it, if you don't kind of, you know, if you don't, if you don't kind of, um, wing it, or as you say over here, phone in with customers, you put the effort in, then there's a business and that will be good. You know, I think that, um, it's, um, so, so I think beyond that, anything, I guess who knows really,

but I think it's, for me, consulting is what I enjoy doing and building a team is what I enjoy doing. And one of the things that I am most proud of, in fact, probably the thing I'm most proud of with my whole business is the fact that, you know, I suppose probably half a dozen or whatever more so of the people that work there, you know, as consultants are now working in pretty kind of serious roles in, you know, in various kinds of companies. They're out there now. And one of them, you know, one of them,

very good luck to him sold his business the other day and, you know, building that team up, you know, being a parent myself, my team, my kids are sort of teenagers now for me, building that team is important really. I think beyond that, who knows really, but certainly for now it's about, I suppose, you know, watch the margins, you know, live a good service, all that kind of stuff really. And the rest of it comes from suppose you know watch the margins um you know deliver good service all that kind of stuff really and the rest of it comes from there but at the moment you know going back to the thing about building a business I would like to get the hours down a little bit I

mean I think I logged 78 hours last week you know on harvest which was which was crazy but yeah it but it is a it is a it is a kind of like a seasonal business in some respects but but yeah for me it's about building a business that you're proud of, that you've got your name above the door, that where people who work for you, you develop them as well, and your customers are successful. That's all you want, really. That's the thing that's most important.

Hey, I'm curious about that. You said it's a seasonal business. Is that what you said? Well, I think it's not like we're farmers over here or something. No, no, no. Do you think there is a cycle to it? Oh, yeah, definitely, definitely. I mean, yeah, first of all, we've got summer over here.

I mean, I know in Canada you're just sort of nine foot in snow and all that sort of thing. Already. And polar bears and whatever. Yeah, yeah, yeah. But, yeah, I mean, certainly there's a period, say, from, say, May through to september when people you know you're not going to get many new projects kicking off at that point so whereas in in now i get to august september everybody is kicking off projects

so the busy see this busy season is typically september through to kind of april may you know that time there um so you know and then you know there's time there. So, you know, and then, you know, there's also, I think as owners and technologists, there's also what we have, the conference season. So you get to kind of around sort of like October, November, December, you've got things like Joy and you've got things like going on. And so we might want to do things there as well. But generally, I mean, certainly the summer is quieter. You know, obviously a business objective is to make it

so the summer isn't quieter. But certainly there is an element of seasonality to it. But, yeah, I think it's about building a team. It's making sure that, you know, your expenses don't exceed your income all the time. It's very easy to always be investing and always be chasing your tail. And I think the doom doom i think the worst situation to be in is where you are um you know you've got you're just filling seats you know

where you're you have to always be searching chasing work and getting people on projects regardless of the work just to kind of make you know just to make payroll each month that's an awful situation to be in and i think the the other one is where you're in a situation where I think consultancies have a sort of bit of a lifespan really. You know, if you do build one where you're kind of, you know, you're starting out, you recruit some people, you know, give it sort of six, seven years. And those people typically have got families.

They don't want to work away so much really. And so it's very hard then to contain your costs. But you end up with a situation where your numbers are going down and it's quite a hard one there so i think there is naturally a lifespan to a consultancy um and you know keeping it going is a challenge and i think well done people who can do that but you know but it's fun and it's what i enjoy doing and and you know and the market is very interesting at the moment and it would be interesting to have that same conversation in five years yeah we'll both be old and bitter and uh and we'll be putting people at any project comes along just to kind of make the numbers and stuff and uh and all that kind of

business so anyway look it's been great speaking to you olivier i'm conscious of time um yeah same here mark so it's been great um and um look good luck with everything you're doing and i'll probably see you presumably at one of the events later in the year and uh. So it's been great. And look, good luck with everything you're doing. And I'll probably see you presumably at one of the events later in the year. But yeah, it's been great speaking to you. And take care and look after yourself. All right. Thanks. Bye. Thank you.