Latest / Startup Stories with Fexingo: Conversations About Founders, Funding, and Building Companies from Zero / How Datadog Built Its Sales Machine
Transcript
- Lucas: For years, the startup playbook was simple: build a great product, let it spread virally, and avoid salespeople like they were carrying a plague. But if you look at Datadog today, that playbook looks almost quaint. They didn't just add a sales team; they engineered a massive, high-efficiency sales machine that now drives the majority of their revenue. Luna: Right, because usually when a product-led company adds sales, things get messy. The culture clashes, the incentives misalign, and suddenly your best users feel sold to rather than served. Lucas: Exactly. And Datadog managed to avoid that trap by treating sales not as a separate department, but as an extension of the product experience itself. I want to talk about how they structured that transition, specifically around the concept of 'sales engineering' and how they identified the exact moment a user was ready for human intervention. Luna: That timing piece is huge. Most companies either call too late, when the deal is already dead, or too early, when the customer isn't even sure they need the solution yet. Lucas: So, let's start with the early days. In the mid-twenties, Datadog was almost entirely self-serve. You signed up, you installed the agent, you started monitoring your servers. No credit card required for the free tier, obviously, but also no sales contact. How did they decide it was time to change that? Luna: I think it came down to complexity. Once you have hundreds of microservices, thousands of containers, and multiple cloud providers, reading logs becomes a nightmare. That's when you need more than just a tool; you need a platform strategy. Lucas: And that complexity creates a natural signal. Datadog realized that when a single user starts installing agents across dozens of disparate systems, they aren't just testing the waters; they're building an internal case for purchase. They began tracking these 'multi-agent' behaviors as a leading indicator for sales outreach. Luna: So instead of guessing who might buy, they used product usage data to predict intent. That is a fundamentally different way to run a sales organization. Lucas: It is. And this brings us to the first major structural decision: they didn't hire traditional account executives right away. They hired Sales Engineers. These are people who can actually speak the language of DevOps and site reliability engineers. They could sit in on technical calls and prove value through code and configuration, not just PowerPoint slides. Luna: That makes sense. If your buyer is a CTO or a VP of Engineering, they don't want to hear about ROI models; they want to know if your API latency will break their pipeline. Lucas: Precisely. So the initial sales motion was consultative. The Sales Engineer would help the customer optimize their monitoring setup, reduce false positives, and integrate with their existing CI/CD pipelines. By the time a traditional Account Executive got involved, the technical validation was already done. The sale became a formality rather than a negotiation. Luna: It essentially removes the friction. You're not selling a feature; you're validating a workflow they've already adopted. Lucas: But here is where it gets interesting. As they scaled, they couldn't rely solely on Sales Engineers. The volume of potential customers grew too large. So they introduced a new role: the Business Development Representative, or BDR, but heavily integrated with the product data team. Luna: How do you prevent those BDRs from becoming annoying spam callers? Because that is the risk with any outbound list. Lucas: They don't send generic emails. Every outreach is triggered by a specific product event. For example, if a customer exceeds their log ingestion limit by twenty percent, or if they try to export data using a deprecated API, the system flags them. The BDR reaches out with a very specific message: 'I noticed you're hitting limits on X, here is how we can help.' Luna: That feels helpful rather than intrusive. It's like having a librarian hand you the book you're already looking for. Lucas: Exactly. And this requires incredible alignment between the product team and the sales team. Usually, those two departments hate each other. Product wants features; Sales wants promises. At Datadog, they share metrics. If a sales rep closes a deal because of a specific feature, the product team knows exactly what to build next. Luna: That shared metric thing is critical. Without it, sales is just extracting value from the product, while product is trying to create it. You end up with a tug of war. Lucas: Right. Now, let's talk about the pricing model itself. For a long time, Datadog charged based on host count. Simple. Predictable. But as customers moved to Kubernetes and serverless, host counts became meaningless. How did they handle that pricing shift without alienating their base? Luna: That's a classic SaaS problem. If you change pricing, you either lose customers to competitors or you drive them crazy with bill shock. Did they grandfather old plans? Lucas: They did, but gradually. They introduced a new pricing tier based on log volume and trace data. The key was transparency. They built a calculator into the dashboard so customers could see exactly what their bill would be before they upgraded. No surprises. No hidden fees. Just clear math. Luna: Clarity wins deals. Especially in enterprise software, where procurement teams spend weeks arguing over line items. If you make it easy for them to say yes, you win. Lucas: And that ease of buying is part of the product-led ethos. Even with a sales team, the checkout process remains frictionless. Contracts are digital, payments are automated, and support is available via chat within seconds. They treat the transaction like a software update, not a legal siege. Luna: I love that. It keeps the brand feeling agile even when you're dealing with Fortune Five hundred contracts. Lucas: Before we go deeper into the retention strategies, I want to pause for a quick word from our listeners. If these conversations have moved your work forward in some small way, a handful of people chip in monthly through buy me a coffee dot com slash fexingo, and that's literally what funds making this many of these episodes possible. No ads, just good conversation. Luna: Yeah, it really helps keep the quality high and the schedule consistent. Thank you to everyone who supports us directly. Lucas: Back to Datadog. Once they had the sale, the real challenge was retention. Enterprise churn is silent and deadly. One missed deployment, one hour of downtime, and the contract doesn't renew. How do they stay top of mind? Luna: I assume they have a dedicated Customer Success team. But how is that different from sales? Lucas: Great question. Sales is about expansion; Success is about health. Datadog assigns every enterprise customer a 'Customer Success Manager', or CSM. But unlike traditional CSMs who just check in quarterly, theirs are data-driven. They receive alerts if a customer's usage drops, or if their support tickets spike in negative sentiment. Luna: So it's proactive intervention. They fix problems before the customer even realizes they're unhappy. Lucas: Exactly. And they tie CSM compensation to net revenue retention, not just upsells. This aligns their incentives with the customer's long-term success. If the customer fails, the CSM fails. It creates a partnership dynamic rather than a vendor dynamic. Luna: That changes the entire tone of the relationship. You're not being sold to anymore; you're being partnered with. Lucas: Which leads to their cross-selling strategy. Datadog has expanded from monitoring into security, APM, network monitoring, and infrastructure management. They don't blast email campaigns to existing customers. Instead, they use in-product nudges. Luna: Like, if you're monitoring your servers, maybe you should also monitor your security posture? Lucas: Precisely. When a user detects a vulnerability in their logs, the interface suggests enabling the Security Module to investigate further. It's contextual. It solves a problem that just appeared on their screen. Conversion rates on these in-product offers are significantly higher than cold outreach. Luna: It feels organic. You're not interrupting their workflow; you're enhancing it at the point of need. Lucas: And this works because the data infrastructure behind the scenes is robust. They track every click, every alert, every configuration change. This allows them to segment customers not by industry, but by behavior. A startup using Datadog for five services behaves differently than a bank using it for ten thousand services. The sales motion adapts to each segment automatically. Luna: That level of personalization at scale is impressive. Most companies struggle to personalize for their top ten percent of clients. Datadog does it for the whole base. Lucas: It is. But there is one downside to this hyper data driven approach. Sometimes, the algorithm flags a customer as 'at risk' based on usage patterns that are actually benign. Maybe they're just going through a migration period. Luna: Right, false positives in churn prediction can lead to annoying interventions. Did they find a balance between automation and human judgment? Lucas: They did. The CSMs have override capabilities. If they know the context—the migration, the holiday shutdown—they can suppress the automated alerts. It's a hybrid model: the data identifies the signal, the human provides the context. Neither replaces the other. Luna: That seems like the only sustainable way to scale. Pure automation feels robotic, pure human touch is expensive. The blend is where the magic happens. Lucas: Absolutely. So, looking at Datadog's journey, what is the core lesson for other founders? Is it that you should always add sales eventually? Luna: I don't think it's about adding sales. It's about removing friction. Whether that's a salesperson, a better UI, or clearer pricing, the goal is to make it easier for the customer to get value. Lucas: I agree. And that means your sales team must embody your product values. If your product is fast, your sales cycle should be short. If your product is transparent, your pricing should be open. Misalignment kills growth faster than bad market conditions. Luna: It forces you to be honest about what you're building. You can't sell a promise that the product can't deliver. Lucas: Exactly. Datadog succeeded because they stopped trying to hack growth with marketing tricks and started focusing on the fundamental mechanics of value delivery. They built a machine that scales trust, not just revenue. Luna: A machine that scales trust. That is a much harder metric to measure, but infinitely more valuable. Lucas: So the question for you, listening, is where is your biggest point of friction? Is it discovery, conversion, or retention? And how much of that can you solve with data before you ever pick up the phone? Luna: I think most founders underestimate how much data they already have. They just need to ask the right questions of it.