How to Structure Your Sales and Team to Scale in Enterprise

All right, let's, let's get started. So, hi, so my name is Philippe Botteri. I'm with Accel. So we're, we're a global venture firm with offices in, in Silicon Valley where we have our origin, and I'm based in the London office, and we also have an office in India. So we're one of the very few early-stage venture firms with a truly global coverage, and it is that network that we give to our portfolio. So we're here because we are, you know, we're pretty big on SaaS. We've invested in companies like Dropbox, Slack, Qualtrics, Atlassian. In France, there is a pretty vibrant ecosystem, so we're in Algolia, Doctolib, PeopleDoc, and Shift technology. So I'm very pleased to introduce Jeremy, who is the founder of and CEO of Shift. So Shift is a SaaS leader in fraud detection and claim automation for insurance. So as you all know, insurance are large, quite large, and the deals when you sell to insurance are big, quite big, and they're not very easy to land, especially when you're small. So in this session, we'll discuss how to structure your sales and marketing team to scale fast when you sell to enterprise. So Jeremy, it's great to be with you today. First, as a start, can you give a sense to the audience about what is your typical deal size? You know, how many customers do you have? So we get an idea of what you're dealing with every day.

Hi everyone, so I'm Jeremy. Thanks for attending. So at Shift, we sell software service solution for insurers. So it's a recurrent business. So insurers pay a yearly usage. The average deal we have is around €500,000 recurrent. So it's not small deals. Of course, you have small deals that go to like €100,000 a year, but larger deals can be multimillions. So our current largest deal is €200 million recurrent, which are—

Can you repeat that?

Sorry, €2 million recurrent.

Okay.

No, I think the average would, wouldn't be €500,000.

OK, good. So maybe— well, yeah, I wish it was $200 million. I would be a very happy investor. But I am still very happy with $2 million. OK, so let's talk about your first customer. Because I mean, it's always hard when you start. You were the smallest you've ever been, and you have to sell to a big insurance company. So tell us about— how did you land your first customer?

So yeah, for Shift, it wasn't an easy sell. So to give you the context for those who are in the same situation, Shift, we are 3 founders. We were at the time 26, 27 years old, and we created Shift saying we want to sell software as a service AI algorithm for insurers. So we were targeting insurance companies in France and we went to see them claiming like, we will automate your fraud detection, you just need to send the data and we'll do it. And everybody was looking at us like, you're 3 27-year-old guys, you want to have all our data processed and send it to us? Are you mad? And so we were like struggling and having our first deal. So The way it went is we got introduced to one of— it's funny, the first one is quite funny. We got introduced as all incubators and startups to an insurance company, and we got introduced to the head of innovation. And the head of innovation was very excited to meet us because his job is meeting startups and introducing it. And we spent 6 months having dinners and lunches and discussions and business case. With the head of innovation. And we were not meeting anybody else. So we were like, whoa, I don't know how this works, how we can land a deal of several— it was like a million euro recurring deal and we're not meeting everyone. And we were like, it's— we were first-time entrepreneurs. So we thought maybe this is the way you sell stuff. And then we had our investors did our seed round. They were like, you're never going to land a deal like this. You need to find who's the key decision maker. And this is where we realized that the key decision maker wasn't this guy. And the funny story is that we met, we get the intro, met this guy, signed the deal in the office, which is exactly next to the office of the head of innovation. And a month after, he just called and say like, where are you guys? We need to start working on the deal. And we already signed in the same, in the office just next to him. So the first deal took us a year to land.

OK, so the first learning is don't talk to the head of innovation at large companies. All right, so nobody can tweet that because my wife is actually head of innovation for a large company. So the key learning is just make sure you're speaking— you can identify in the organization the right decision maker. OK, so you've landed your first customer. Very sizable deal. So that obviously gives you a lot of credibility when you speak, reach out to other customers. So now from there, you start thinking, you know, how do I scale? And the first question for you is like, how do you organize your sales team to scale first in France and then across different countries?

Yeah, so then once we had our first customer, it was about scaling sales. The first thing I learned is that I'm not a good salesperson. So the first thing you do when you're terrible at something—

We did not know that at the time of the investment.

Is that you hire somebody to do it. So the way we scaled it, we spent 6 months looking for head of sales. So we hired, she's now our chief revenue officer, somebody that was head of sales for SaaS to large organizations. And we took somebody that already scaled the business. So she was the head of sales of IBM Cloud worldwide. And 6 months because you need to convince somebody to quit a job with 450 sales reps to 1 sales rep in Paris. And then this is where you learn how to scale. So it's about having the right proposition value, having the right references, organizing packages for sales, and then hiring sales. And all the work we've done is packaging how we can identify who's the key decision maker, what is he interested in, and how we sell it to these people, and what's the right business model. And this is what we've done for the first 6 months. And then it's about having the right team under the right team.

So that was for France. So basically the structure of the organization, you hired a head of sales, and then under that head of sales you had different reps, though, did you have a specific sales ops team to prepare, I mean, to do all the account management preparation, identification of the target decision maker, or was this done by the reps?

So, um, we are organized in a way where, um, we have the reps that do all of this work, but all the technical sales is done by data scientists in the company because, um, it's very important that the people that gonna do implementations and work on production are the same people that talk to the customers. And this is also what forced us to have data scientists from different countries because you want them to talk to— in German to German, to talk in Spanish to Spanish. And so the organization is you have head of sales with sales rep that really literally takes the whole sales process. And then within the sales process, you have data scientists that are experts that work with these guys and are in front of customers.

So you send your data scientists in front of the customers?

Yeah, can you imagine sometimes situations how it can go?

I'm not going to comment on that. So how do you hire your data scientists? Because usually data scientists, I mean, they love data, they love science, so they love to work in front of their computer. They are not the people— I mean, if you're a data scientist, you don't do that because you want to go and sell to people. Otherwise, you go into sales. So how do you screen and hire your data scientists to make sure that they can also do this sales role?

So we did an internal data scientist study, and we realized that the first feature was being a fan of Star Wars. So this is the first thing you want to look at the data scientist. No, I'm just kidding. It's true. We're big fans of Star Wars in the company, so everybody makes jokes about that. So yeah, what's important in our interview process is, of course, there's like mathematical questions and coding questions, and then there's a last interview where what we try to find out is, is this data scientist would be able to work with users that are today claim handlers working insurance company looking at papers and dealing with claims on a daily basis, or they're gonna be like Crazy geniuses in front of their computer don't want to go to meet customers. So this is the first thing, is hiring the right people. So once they're hired, the first thing we ask them to do is work with salespeople, and they spent 4 months with a user in an insurance company. So before hiring data scientists, we tell them the first 4 months you'll be sitting with claim handlers and working as a claim handler. So You come out of university doing a lot of mathematics and then the company puts you with a claim handler taking calls and settling claims. This is a good way of selecting the ones that really want to do it. So this is the complete process we have.

Okay, it's very interesting because as we see, I mean, a lot of the new products we see right now, I mean, there is machine learning everywhere. And when there is machine learning, that means you have some technical sales to do. And so it's not the first time I hear that, you know, data scientists have to be in front of the customer. And so it's very important what you describe in terms of how do you onboard them, making sure they really understand what's happening at the customer before they're actually placed in front of new customers. So in your organization, so how did you handle the geographic coverage? Like, was everything based in France to cover all of Europe, or did you start to separate the country organization? And then how did you address the organization for the US market?

So the way we're structured is you have sales reps that are in countries where you want to land deals. We have a rule internally where these sales reps need to be at least once a month, a week in Paris. It's very important. And then for the data scientists, either it's in Asia, so you put them on site. So what we do is we send French people that go on site and create teams. Easier for the rest of Europe, it's in Paris. And we are quite an international team. So if you visit our office, you'll see we have a lot of people that don't speak French. We have a German team, a Spanish team, a UK team. So every country has its own team. And they all work on the same solution. So it's quite interesting because this is where you see that, for instance, German people really focus on identity aggregation in terms of pushing into the solution. Spanish people focus a lot of networks detection. So it's a quite international culture where people are per country working with the sales team but work on the same platform.

Okay, and so do you have a country head managing all the different people in the country, or do they report to, you know, whatever head of function based in France?

Yeah, so this is the toughest part of Structuring the company. So we don't have local managers. When I say like, we don't have a manager of Singapore, manager of Japan. We have head of sales reporting to the chief revenue officers, and we have data scientists reporting to the chief data scientists. And this is where the biggest part of my job is being sure that there's a harmony between these people and they work together, and it goes with the right processes that people need to execute. Like we're saying, like in sales, you need to have sales rep taking data scientists. So he can't go to a meeting without a data scientist. And on the other side, you need to have a data scientist team being able to execute what the sales team sell. So it's all about making sure that there's a real team spirit, the right process, and the right harmony. And this is the tough part of what we do. But when you looked at it today, we have We have a really amazing conversion rate. We don't have any problem in getting our customers. And it wasn't easy at the beginning. Can you imagine, like, you hired a bunch of IBM sales guys and you tell them, you know what, we don't have technical sales, you take these data scientists, 25, in your sales meeting pitching the product. It was a small time of adaptation.

Wow. So that's— so basically, if I try to summarize your sales model, what the The key thing is making sure that you have a dual team, like a joint team, sales and data science, making sure that the data science people are trained at the customer so they really understand the business. And then you develop your country teams, but they're managed kind of centrally, right? That's, that's exactly, that's a playbook. And did this scale into the US as well, or did you need to do something different from, for the US market?

So for the US, it's like we've done in Asia, like we've done in Europe. So it's hiring local head of sales, local data scientists, training them from Paris, and repeating exactly the same, the same modus operandi. And it's funny because one of the funniest things is sometimes you have this ranking about startups. And once you've done this opening an office with local data scientists, local head of sales, since we're quite international. There's a lot of startup ecosystem tweeting like Shift, a Singaporean startup in the top 100, I don't know what. Shift, a Japanese startup in the top 20-something. And everybody thinks we're local because like the guys that are on-site local and we have this really harmony where the head chief revenue officer goes to these meetings and help the local sales. Travel everywhere. So what's important is make these insurance people feel they're local while everything technically is some kind of centralized in Paris.

And how much time do you spend? I mean, you have customers both in Asia, Europe, and the US. So like, how much time do you spend on the road?

A lot.

Okay, well, you don't have to answer that question. So let's talk about hiring the sales reps. So what are the traits that you're looking for when you want to hire a salesperson who can sell and like a $200 million a year deal? No, maybe a $2 million a year deal.

We're working on that. Promise, 2 years, $200 million. So in terms of hiring, this is also, yeah, so I spend most of my time meeting customers and hiring. This is my job. So about hiring, the first rule is Trust the person you hired. So this is the first rule. So our chief revenue officer is, she source candidates, her team source candidate, gets candidate. And what they try to catch is people that know sales. So can stick to a sales process, understands the key sales processes in terms of who are the key decision makers, what I'm selling, understand what I'm selling. And then the last meeting is with me, and what I try to understand is if they can fit into the culture of the company. So it's very important, and there's a lot of work, and we have a whole bunch of questions to really— we worked a lot on it to, depending on how the person answers, you can feel if they are going to fit in the company or not. There's a lot of—

So can you give an example of these kind of questions?

An example is you ask, like, you have some basic questions, what he thinks about something. And my job is saying, not approving what he's saying for 10 minutes and taking the data scientist role. Like, I don't agree, it's not like this. And like, I'm annoying for 10 minutes. So in 10% of the cases, they leave the interview, which happens sometimes. And then this is where I'll see, like, this guy's going to be in front of the data scientist that needs to go with him to meet a customer. And the data scientist won't agree because He won't agree selling this because he won't be able to do it. How do you react? So these are kind of things that we have in our process. And the last thing is, does the person understand, will the person understand our business and a part of what we're doing? Because like every startup, we have failure when you hire somebody that may be a great salesperson that you hired from a great software company, but then you know that this guy would never understand what he's selling. And even with a data scientist, he will never understand it. And we had some failures where you have insurers calling you like, you're sure you're doing this? It's funny. It doesn't look achievable. And you end up signing deals where you think it's not achievable.

Interesting. And at the early phase of the selection process, I mean, do you say, well, you know, I hired like 2 salespeople from, you know, with this kind of background, you know, very specific on companies and years of experience and you're like, okay, well, that worked, so let's go and look for this, or do you actually find more diversity in the background of your salespeople?

Yeah, so in terms of selection, exactly, we're looking for two— there's two components, and depending on countries. So if you want to hire somebody in a country where the biggest roadblock you have is software as a service, in some countries people don't go for software as a service. They don't want to send data outside of the company. So you're going to look for somebody that already sold SaaS or cloud in this company. So IBM Cloud is a good way of hiring. And in some countries, SaaS is much more common. What's important is having the insurance stamp because insurance is very local. So for these countries, you're going to look for somebody that's already working in an insurance environment. Insurance knows them, and this is what you're going to look for. So we, per country, try to find what are the biggest roadblock, and this is how we define people we want to hire.

Okay, and how do you define the quota? So what's your typical quota, and, you know, what's the comp, you know, fixed bonus?

So this is where I trust my Chief Revenue Officer. She's like, comp, it's 50 fixed, 50— so OT is 50/50. And in terms of quota, it's, we define the strategy, this is what we need to hit, and then she does her stuff. And at the end, I can see how many, like, which each salesperson, how much quota they have. And when they start, the quota is quite low and then goes up. And I ramped a salesperson from our side that is, somebody's like, bit experienced, their quota is between $2 to $4 million. Like 2 million average.

A year?

Yeah.

Well, that's a big quota. And when you start, so how long does it take to get to the full ramp and where do you start?

So get full ramped, it's quite long. So we observe that people in order to be ramped first and start signing deals, there's 9 months because they need to get to know the market, understand the product. It's not an easy sell. And then so for the first 9 months, you're going to hit a quota that's around like $500K, $1 million. And then you have 9 more months in order to hit the full ramp. So we have a selection where if the first 6 months, you know he's not going to hit his 9-month plan, you just don't keep him. And once he hit the 9-month plan, how you can ramp him to be sure he's going to achieve the high quota.

So when you have— a long ramp like this for a sales rep, what are the milestones that you give the reps in between just to make sure that he knows that he's on track and that you know he's on track?

So this is where we have internally a good process is you need to be able to track every quarter the KPIs you give it to your salespeople because you're close to them, they can tell you like, I'm not hitting it because of this or that. So you can, actualize your KPIs, but it's important to really track it very precisely and being able to put the emotional part aside and understand like, is he hitting his KPIs or not and why? And if the why is that it's not the right person, we just fired him. So it's— there's a lot of work when we hire somebody in the sales team, which is much more work than when we hire data scientists. It's being able to have the right KPI make the person understands why he has KPI and being able to track it. So it's all about KPIs per quarter that are very precise.

But what kind of KPIs? Because if you're not expecting them to make a sale before 6 to 9 months, then what is the KPI for the first quarter?

So the first thing our Chief Revenue Officer did when she arrived is putting different sales steps. So identification of accounts, qualifications, like she's very, very annoying on qualifications. You put something qualification not qualified, she goes mad in company and start shouting. So it's qualification, proposition, approval, signing. So there's a lot of different stages. We have 7 stages in the company. And then she has her plan where the first quarter you need to hit this number of quota in this column and this number of quota in this column. And she has a nice statistics on how do you go to a stage to another and what's the conversion rates. So it's quite statistics and mathematics. So you look at KPIs and then you look at how it ramped. It's very easy now with all the process we have.

And how do you check that a stage is actually achieved? Because I mean, salespeople, they're pretty good at, you know, being a bit forward-looking, right?

So if you don't tweet that, I'll give you the secret is you ask the data scientist that goes with them.

That's why you want the data scientist in the team. Okay, that's a nice tip. And so how much do you— for a $3 million quota, like, what's the salary that you pay sales rep?

They're gonna poach our salespeople.

Not here.

No, no, we pay them a lot. It's— what we do is we take like other companies' salaries, and we pay 20% more.

Okay. And how do you set your overachievement curve? Like, is the— when you look at the bonus payouts, like, is it a linear payout up to 100 and then continues to be linear, or do you have an ex— you know, first you're kind of an exponential where at the beginning if you do 50% you get nothing, then it accelerates to 100, and then How do you manage the acceleration beyond 100?

Yeah, this is, this is quite interesting. As a first-time entrepreneur, like, I didn't get it. People were asking questions I didn't understand. So it's very important. I didn't know that the way you structure compensation is very important. So the way we have it internally is if you achieve under 70%, you get zero. If you achieve 200%, it's linear, and above 100%, there's an acceleration. Where everything more than 100%, you have a multiple that's quite high. And this is what attracts very good salespeople. And this is where also I learned the 30/30 stuff. So the 30/30, I didn't understand what was it, but it means that salespeople will tell you, I can achieve only this, which is 30% lower than what he thinks he can achieve. So you need to know it so you can put 30% more. And this is where like it's It's now business as usual. You just take 30% more and then put 30% lower because of this acceleration stuff.

Nice. So 30% is the number. Good. So what are— so if you were to, you know, to summarize, I mean, when I hear what you're saying, I mean, it sounds like to sell into enterprise, you basically need to put in place a very organized machine. So it starts with, you know, mapping the accounts to understand really who are the decision makers. It's about making sure you have the right team, which is a good mix of sales and technicals, figuring out how do you have local— I mean, people in the local region where you sell to make sure that your customers locally think you are local, and then put in place a very structured sales process to be able to measure the progress of the salespeople, even though you're— they're not expected to make a first sale for the 6 to 9 months. So then once you've put all this in place, so what are the challenges that you're facing today now as you scale?

The challenge is— so it's not easy to do all of these, but the challenge after that is making sure that people stick to the process and that the people that hire people that hire people are still exactly in the same culture that people you hired. And this is my job today, is being sure that processes are put, really put in place, and being sure that data scientists go to all of these meetings and get the feedback from the data scientists. So after that, it's about looking at things scaling, being sure it's still scaling the right way.

Great. Well, thank you very much, Jeremy. It was a pleasure to be with you today.