Pod Insights Y Combinator Podcast Enterprise Sales with Pete Koomen | Startup School About this episode Y Combinator Partner Pete Koomen led his startup, Optimizely, to $100M ARR. In this video, Pete breaks down the enterprise sales funnel and shares his top tips on how a technical founder can start closing real deals for their startup.
Transcript 0:09 Speaker 00 My name is Pete Koomen. I'm a group partner at YC and a YC alum. I was co-founder and CTO of Optimizely in the winter 2010 batch. In this talk, I'm going to walk step by step through the process of closing your first enterprise customers. I'm going to do that by focusing on successive steps in the sales funnel. Prospecting, outreach, qualification, pricing, closing, and implementation. I'll do my best to include lots of tactical advice and counterintuitive lessons I picked up while I was learning how to sell at Optimizely. I'm going to focus on enterprise sales for software startups. But this talk should still be broadly useful to any founder getting started with sales, regardless of the size of your customers or what you're selling.
0:53 Speaker 00 Why am I giving this talk? Well, first, I know there's demand for this. Sales is the number one concern during the batch for most of the founders that I work with at YC. Second, I know from experience that sales is a learnable skill. My co-founder Dana and I both had technical backgrounds. We knew how to build a product, but we didn't know how to get people to use it. We figured it out through trial and error. And that's the first big lesson I want to impart today. If you're the founder of an early-stage startup and you're building a product that you're hoping other businesses will buy, you are capable of selling it. That's the good news. The bad news is that you're probably the only person capable of selling your product.
1:37 Speaker 00 That is, if you aren't able to sell your product yourself at first, chances are you're not going to be able to hire somebody else to do it for you. Now, if you're anything like we were, you're probably thinking, "There are lots of talented salespeople out there. Wouldn't it be faster to hire one of them than try to do it ourselves?" After all, that's what you'd probably do with any other role, like designers, or lawyers, or accountants. The problem is that sales before you find product-market fit is very different from sales after you find product-market fit. Sales pre-PMF is fundamentally entrepreneurial.
2:10 Speaker 00 It requires vision and credibility with customers and lots of experimentation, and a tight feedback loop with the people building the product. This is a role for founders. So does that mean if you're a team of technical founders building a product, you should go find a business co-founder to do sales? Well, you probably don't need a business co-founder to sell either. I've worked with many technical founders who turned out to be great at selling, in some cases, very much to their surprise. So why is that? Well, if you're a technical founder building a product, you have several advantages that will give you a big leg up in selling. First, you're an expert, both in the problem you're solving and the product you're building. And second, you have conviction.
2:54 Speaker 00 You sincerely believe that your product will solve your customer's problem. Expertise and conviction are surprisingly important in sales. This is especially surprising to people who mistakenly think that selling is a dark art full of psychological tricks. Sales isn't about tricking people. It's fundamentally about helping people solve their problems, and engineers are great at doing that. Anyway, now that I've hopefully convinced you that you're capable of selling your product, let's talk about how to do that. Like I said, we're going to go through the step in a typical sales funnel together. Let's start with prospecting. Prospecting means finding potential customers. The output of this step is a list of companies you think might need your product,
3:39 Speaker 00 and the specific humans at those companies you think might buy it. There are lots of tools you can use for prospecting, but before you start, you need a hypothesis. A sales hypothesis goes something like this: Customer X has problem Y, and our product will help them solve it. A good hypothesis makes prospecting easy by clarifying who you should be talking to. For example, at Optimizely, our initial hypothesis was something like this: Marketers at small and medium tech, media, and e-commerce companies want to run A/B tests on their websites. But they can't because off-the-shelf experimentation tools require users to write code. Optimizely will enable them to run A/B tests without writing code.
4:24 Speaker 00 And once you have a clear hypothesis like that, you can get to work on prospecting. Start by identifying companies that are likely to suffer from the problem you're solving. One way to do this is to buy industry lists of all of the companies in a given sector, and then use some filtering criteria to qualify those companies and narrow your target list. For example, at Optimizely, we used a tool called BuiltWith to figure out whether prospects were using analytics tools and JavaScript frameworks, because those were signals that a company was relatively sophisticated and cared about their website. Once you have a list of companies, you'll need to find the right humans at those companies and their contact information. There are tools that make this easier.
5:08 Speaker 00 This video was recorded during the Winter 24 batch, and many founders in the current batch are using Apollo and LinkedIn Sales Navigator for that. Now that you have a list of leads, that is, specific humans that are likely to buy at the companies you're selling to, you'll need to get their attention. This step is called outreach. The goal of outreach is usually to schedule a meeting with your prospect. Most founders think of cold outreach as the primary mechanism for doing this, but the easiest way to get a meeting with a prospect is to get them to reach out to you. Even if you're planning on using a sales-led approach, you should still do everything you can to generate inbound demand. Launch early and often. Create technical content like videos and blog posts
5:50 Speaker 00 that prospects can find while searching for a solution to their problem. Build self-serve demos that people can share. Find online forums where your customers hang out and establish yourself as an expert by answering questions. There's no one way to do this, but the better you get at grabbing your customers' attention and getting them to reach out to you, the more efficient your sales process will be. On that note, if your customers all hang out at industry conferences, you should be there too. Find a way to get a list of attendees ahead of time and set up lots of meetings in advance. Once you've identified a specific prospect you want to talk to, start by trying to find a warm introduction if you can. Look on LinkedIn for shared connections and ask for an intro.
6:33 Speaker 00 Sending cold emails is usually the least efficient way of getting prospects' attention. But it can still be effective if you approach it the right way. Start by writing each email by hand. Make your emails short and to the point, and make the ask clear. You should also make it clear why you're reaching out to each recipient specifically. Humans have built-in spam filters, and if your email looks like it was sent to thousands of people, it's going to get deleted. On this last point, there's a handy rule of thumb to keep in mind for cold emails. Only send emails that you yourself would be excited to read. If you wouldn't be excited to get the email you're about to send, your prospect probably won't either. Before we move on, I want to spend some time talking about a particular anti-pattern
7:17 Speaker 00 I see with a lot of YC founders. Many founders start by talking to anyone who will take their call. And the problem with this approach is that it selects for the people who are easiest to talk to, not the people who will be great customers. So if you're not disciplined about it, you'll end up wasting all your time chasing bad customers that are easy to talk to. I see YC founders make this mistake all the time, and I get it. When you're starting a company, it's hard to get people to pay attention to you. Cold emailing is a demoralizing grind. So it's tempting to go after the people who will talk to you, even if they won't ever buy your product. And the reason this mistake is so dangerous is because talking to bad customers gives you the illusion that you're making progress when you're not.
8:01 Speaker 00 You'll get lots of great product feedback from people who think they're doing you a favor, but because you're not actually talking to someone who needs your product, this kind of feedback is useless at best and counterproductive at worst. In practice, I see founders make this mistake in two ways. First, by trying to sell enterprise software to startups. Now, if your product solves a problem that companies only have when they get big, like, for example, an HR information system, then trying to sell it to startups is a waste of time. But founders still do it all the time because other startups are much easier to talk to than busy, big-company executives. And the second is trying to go bottom-up with a product that needs to be adopted top-down.
8:45 Speaker 00 And that's a little jargony, so let me illustrate it with an example. Imagine you're building productivity software like Notion. Your product can be adopted bottom up, meaning that individual employees or teams can start using it independently without having to coordinate with anyone else inside the company. And in this case, talking to individual contributors or their direct managers is totally fine. But what if you're building billing software for large hospitals? In order for a hospital to start using your product, you're going to need a lot of different teams to coordinate with each other. So you might need the CIO to sign off on your security and compliance. And you need their software team to integrate your product with their in-house systems. You need the doctors to enter billing codes after each appointment.
9:30 Speaker 00 And you need their ops team to manage collections and so on and so on. And in this case, talking to an individual doctor won't be useful. You need to talk to a senior leader like a CFO or a CIO to do a deal. Now, there's a meme that YC says you should sell to companies who will buy quickly even if they aren't good customers. This is a misconception. You should try to find companies that will buy quickly, but you shouldn't spend time trying to sell to companies that don't actually need your product or won't be good customers. You need to find people who have the problem you're solving and the budget and decision-making authority to buy your product. We'll spend more time on this point later. Let's get back to our sales funnel. So you've managed to get your prospect on the phone.
10:14 Speaker 00 Your job on the first call is not to sell your product. That comes later. In the first call, we're just trying to do two things. First, we're trying to qualify our prospect by figuring out whether they have the problem we're trying to solve and the budget and decision-making authority to buy the product. And second, we're trying to schedule a follow-up call for a product demo. Now, many founders face plant in the first call by diving straight into their pitch. These founders are making one of the biggest founder sales mistakes: not asking enough questions. And they make this mistake because they misunderstand how sales work. They think of the company they're trying to sell to as a big, monolithic entity, and they think of the sales process as adversarial,
10:59 Speaker 00 where it's their job to come up with a perfect pitch that will break down their target's defenses. But outside of some used car dealerships, that's not how sales works in the real world. In the real world, you're almost always selling to an individual human, not a big monolithic entity. And that's good news, because humans are easier to understand than organizations. And that turns out to be really important. In the real world, sales is not adversarial. It's about deeply understanding a customer's problem and helping them solve it. And great salespeople spend most of their time listening, because that's the best way to understand someone's problem. They ask all sorts of questions. What made you decide to take this call? Tell me about this problem. How long have you had it?
11:44 Speaker 00 How bad is it? Who else does it affect? How do you quantify the impact? Why haven't you solved it already? What's your budget for solving it? How does your organization buy software? Who makes the buying decision? Who else will need to weigh in on this decision? Now, sometimes when you ask questions like these, you discover that your prospect doesn't actually have the problem you're trying to solve. Or they have it, but they don't care enough about it to buy a solution. Or they don't have any budget, or any number of other reasons that they won't actually be a good customer for you. And if you do that, that's great. You just saved yourself and your prospect a lot of time, and you can focus your energy on other prospects that are more likely to buy. Now, if, on the other hand,
12:28 Speaker 00 it turns out that your prospect does have a problem you can solve, you're in luck. You've earned the opportunity to show them how your product works. The next step is a demo. Most founders think of a demo as a chance to finally show off their product. In my experience, thinking about it this way is a surefire way to deliver a bad demo. That's because your job in a demo is not to show off your product. It's to convince your audience that you can help them solve their problem. And one helpful trick I've learned is to think of your demo as the script for a great movie. A great script always starts with a recap of who the main character is. That's your user. And the problem she's trying to solve.
13:11 Speaker 00 This is your chance to demonstrate how well you were listening during your first call. If your audience believes you understand their company and their problems, they're going to take you seriously when you talk about how to solve them. When you're ready to show the product, resist the urge to take your audience on a feature tour where you walk from screen to screen showing them everything your product can do. Instead, tell a story that shows exactly how your main character solves her problem. And this is the point. Great demos actually feel like good stories. They have a flow where each step leads to the next and every feature you show has a clear reason for being there. They usually have one or more magic moments where you surprise your audience with how easy or delightful something is.
13:56 Speaker 00 And great demos are also personalized for the audience. This is where you get to use all of the information you collected during that first call. Tailor the demo to their company. Use their logo, their website, their customers, the names of the people on their team. The more you can do to help them visualize exactly how your product would work in their company, the better. I'll give you an example from the early days of Optimizely. When Dan and I started building, we booked demos with all of our competitors. And every single one of them used a dummy website to show what it was like to use their products to run A/B tests. And we thought this was really lame. So we spent weeks building a feature that made it easy to demo our product
14:40 Speaker 00 right on our customers' websites instead of a dummy website. And I knew it was worth it when I saw marketers' eyes light up when they watched us change things on their landing page that would have taken them months to do on their own. So if you do a good job, your prospects and their team will come out of this meeting convinced that you can solve their problem. And if that's the case, it's time to talk about pricing. I get a lot of questions from founders asking how to price their products. And the truth is, there isn't a simple formula for doing this. So if there isn't a formula, how do you pick a number? Well, fortunately, you can ask questions earlier in the process that will make your job easier here, like, how much is this problem costing your company? How many people are responsible for maintaining your in-house solution?
15:25 Speaker 00 What's your budget for solving this problem? How much are you spending on my competitor? And it's okay to wait to share your pricing until you've had a chance to ask these questions. In fact, if your product requires lots of work or customization to implement, you probably shouldn't quote a price until you understand exactly what your customer needs. In any case, even if you asked all of these questions upfront, the reality is that pricing involves a lot of guessing in the beginning. And the advice that I give to startups is to think of each pricing conversation you have as an opportunity to run an experiment in which you test a price point and then learn from your prospect's reaction to it. In the early days of Optimizely, we had published self-serve pricing for customers who just
16:09 Speaker 00 wanted to swipe a credit card for a basic version of our product, and an enterprise plan that required you to go through sales. We didn't publish pricing for the enterprise plan, which gave us the flexibility to try a different price each time. The pricing mistake that founders make most often is charging too little for their product, or even making it free in exchange for product feedback. Founders do this because they're worried about charging too much. They think they're going to scare customers away. One of the most surprising things I learned was that when a customer really wants your product, it's hard to scare them away by quoting a price that's too high. For example, I remember my co-founder Dan coming out of a sales call and telling me that he'd worked up the nerve to quote the prospect $10,000 a month for our software.
16:54 Speaker 00 And the prospect ended up talking us down to $2,000 a month and then buying. Our initial quote was 5X what they were willing to pay, and they still bought. In fact, higher prices can help you figure out whether customers actually need your product. The Colson Brothers famously charged more for Stripe in the beginning than their competition did. The fact that they were able to sell their product anyway was compelling evidence that they were on to something. And it helped them focus on the customers who were most desperate for a solution. High prices make customers more serious. This brings me to another important point about pricing. Remember that the most important conversations about pricing will happen without you in the room. Your prospect will need to convince others in the organization that your product is worth the price you're asking.
17:39 Speaker 00 You can make their job easier by giving them slides or a PDF one pager that explains how your pricing works. It's usually a good idea to include an overview of your product and the benefits of using it in case your prospect needs to talk to people who aren't familiar with it. In any case, don't spend too much time thinking about pricing in the beginning. Pick a number, ideally one that makes you a little uncomfortable, and pay attention to how your prospect reacts. It's okay to let them negotiate you down. Remember, in your first few sales, you're optimizing for learning, not unit economics. Now that you've agreed on a price, it's time to close the deal. Closing is not a single conversation. It's a bunch of things that need to happen from the moment your customer decides they want your product
18:24 Speaker 00 to the moment they actually buy it. Big companies, especially ones in highly regulated industries, have formal procurement processes that usually include stuff like security and privacy reviews, legal reviews, and sign off from compliance teams. It's less formal in smaller companies, but you should expect at a minimum to go through a redlining process with their legal team. The biggest mistake that I see founders make at this stage is getting surprised and discovering that what they thought was a done deal is, in fact, not done at all and may take weeks or months of additional back and forth, or fall through completely. Now, the way to avoid getting surprised is, once again, to ask a lot of questions. Asking your prospect upfront how they buy software and who needs to sign off
19:08 Speaker 00 will give you a clear picture of the hurdles you're going to have to overcome in order to get a signature. You should do everything you can at this stage to move through the procurement process quickly. Ask explicitly if there are steps like filling out a security questionnaire that you can get started on early and execute in parallel. And keep your legal documents as simple as possible. I recommend starting with the open source templates published by YC company Common Paper. Keep timelines and scope of work out of the legal contract if you can and put them in an order form or a shared project tracking document instead. And most importantly, remember that your prospect, who at this point has become your champion, is your biggest ally. You should be in constant communication with them.
19:51 Speaker 00 And when you need help getting something unstuck, you should ask them first. Remember, they can't solve their problem until you get through procurement. So they're heavily incentivized to help you make it happen. Now that you have a signature, congratulations. It's time for your customer to actually start using your product. This is the implementation, and it's the last step we're going to talk about today. And I'm going to lead with this. The single biggest mistake that founders make is thinking that implementation is the customer's job. We made this mistake more than once at Optimizely. In fact, we closed six-figure deals with customers that were excited about our product. And then we discovered a year later, when it was time for them to renew, that they hadn't run a single A/B test with Optimizely.
20:35 Speaker 00 And this was confusing at first. Why would a customer who was willing to pay so much for a product fail to use it at all? In this case, the proximate cause is that the marketing team who bought our software couldn't convince the software engineering team to help them install it on their website. The real cause, though, is we didn't do our jobs. We thought our customer was buying a product, so we sold them one and left the rest up to them. In reality, our customers were buying a solution to a problem. And all of the work required to get from product to solution was our responsibility. We learned to start asking marketing leaders about the work required to implement Optimizely very early in the sales process.
21:19 Speaker 00 We started building detailed implementation plans with marketing and engineering leaders well before a contract got signed. And in fact, if we weren't able to do that, we wouldn't sign a contract at all. And the trick we learned was to treat the customer implementation the same way that we would a high priority project inside of our own company, by project managing it. So we put together a shared roadmap. We made sure that every task had its owner. We set up regular check-in meetings to hold everyone on our side and theirs accountable for getting it done. Your sales funnel only really ends when your customer is using your product habitually. And when you get to that point, congratulations. Hopefully, you have a customer for life.
22:01 Speaker 00 All right, we've covered a lot today, from prospecting to doing your first call, delivering a great demo, pricing, closing, and finally, managing the implementation. Of course, there's so much more to learn about sales, and the best founders devour everything they can on the topic. If you want to go deeper, I recommend Peter Kazanji's book, Founding Sales. It's a fantastic resource, and it's free online. But like most of the hard things about building a company, the best way to learn is by going out and doing the thing. So if you only remember one thing, it should be this: just get started. You'll make mistakes, but with enough attempts, you'll figure it out, and selling will start to feel natural. Pretty soon, you'll discover that you've acquired a new superpower.
22:46 Speaker 00 You'll find it's useful not only in getting customers and revenue, but in fundraising and hiring, too. And soon enough, you'll be the one giving advice like this to new founders. Thanks for watching.
Previous Startup Experts Discuss Doing Things That Don't Scale | Office Hours Next Why Startup Founders Should Launch Companies Sooner Than They Think | Office Hours Page updated October 26, 2026 · transcript and summary are machine-generated
