How Datadog went to market

PrimarySupportingTried, then dropped

1. 2010 to 2015

   Self-serve monitoring for DevOps teams

   Datadog was founded in New York in 2010 and launched Infrastructure Monitoring in 2012. Engineers installed it themselves, and it passed 1,000 customers in 2015 before any enterprise sales team existed.

   • Developer-led

     Datadog was built to bring development and operations teams together. The S-1 says the product lands with development and IT operations teams first, then spreads organically to broader teams.^[1],[10]

   • Product-led

     A hand-picked closed alpha gave weak signal, so Datadog opened a wide beta and let users self-select. The platform installs self-service within minutes and is offered through free trials and a free tier.^[11],[1]

   • Sales-assist

     The first head of product also ran sales. Pomel says the first salesperson came only after a modest amount of ARR, with the team built bottom-up, and deals were sold month to month for a long time so churn showed up fast.^[10],[11]

   • Owned events

     Datadog exhibited and gave demos at cloud conferences, with engineers running the demos, because companies migrating to the cloud go to events to learn. Its technical leaders also spoke at events such as DevOpsDays and AWS re:Invent.^[11],[1]

   • Community

     The agent, integration plugins, SDKs, and documentation are open source, with developers from over 650 organizations contributing. This is community around the client software, not an open-source product funnel.^[1]

2. 2016 to 2019

   Inside sales, enterprise team, and multi-product expansion

   Datadog set up an enterprise sales team in 2016, launched APM in 2017 and logs in 2018, opened a Dublin sales office, held its first DASH conference, and went public in September 2019.

   • Land and expand

     Customers land on one product and expand on a self-service basis. Net retention was 141% to 151%, and about 60% of new customers in the first half of 2019 landed with more than one product, up from about 15% a year earlier.^[1],[2]

   • Sales-assist

     A self-service tier, a high velocity inside sales team, and an enterprise sales force worked as one model. Sales turns a trial into a first purchase, and customer success drives renewals, upsells, and cross-sells.^[1]

   • Usage pricing

     Subscriptions commit to usage measured mainly by hosts or indexed data volume, and customers pay for usage above the commitment, so revenue grows as their cloud workloads grow.^[1]

   • Sales-led

     An enterprise sales team sold to large businesses from 2016, and the sales and marketing organization had 474 people by June 2019.^[1]

   • Owned events

     DASH, Datadog's own user conference, was first held in New York in 2018, alongside regional, national, and global events.^[1]

   • Partner channel

     A partner team worked with resellers, distributors, and managed service providers.^[1]

   • Inbound content

     Marketing reached the dev and ops community through content, email, events, digital advertising, social media, and community programs. The S-1 lists it as one tactic among many, not the main demand source.^[1]

3. 2020 to 2022

   Platform expansion and peak net retention

   Datadog added security, database, and incident products from 2020. By the end of 2022, 81% of customers used more than one product and net retention stayed above 130%.

   • Land and expand

     Cloud SIEM, Incident Management, Cloud Security Posture Management, and Database Monitoring were among the 2020 and 2021 launches. About 81% of customers used more than one product at December 2022, and trailing net retention for that month was later restated as mid-140s.^[3],[4],[5]

   • Usage pricing

     The usage commitment model held, and revenue concentrated in larger accounts: about 2,780 customers above $100,000 in ARR made up 85% of ARR at December 2022.^[3]

   • Partner channel

     A Marketplace let customers buy products partners built on the Datadog platform, alongside resellers, system integrators, referral partners, and managed service providers.^[3]

   • Cloud marketplace

     The filings name cloud-provider marketplaces as a sales route, a pricing factor, and a margin risk. They do not disclose how much revenue flows through them.^[1],[3]

   • Sales-led

     The sales and marketing organization grew to about 2,300 people by December 2022.^[3]

4. 2023 to now

   Usage optimization, enterprise scale, and AI

   Customers optimized usage in 2023 and trailing net retention fell to the mid-110s. Datadog kept adding products, including LLM Observability and the Bits AI SRE agent, and recovered to about 120% in 2025.

   • Usage pricing

     Usage pricing cut both ways. The FY2023 10-K blames the drop to mid-110s on slower usage growth from existing customers, and by 2025 an AI-native cohort that includes the largest customer added about seven points of Q4 revenue growth.^[4],[5]

   • Land and expand

     At December 2025, about 55% of customers used four or more products, 18% eight or more, and 9% ten or more. Launches in 2024 and 2025 included LLM Observability, OnCall, Product Analytics, and Bits AI SRE.^[5]

   • Sales-led

     About 4,310 customers above $100,000 in ARR made up 90% of ARR at December 2025, 603 customers were above $1 million, and the sales and marketing organization had about 3,600 people.^[5]

   • Cloud marketplace

     A December 2025 collaboration agreement with AWS covers AWS Marketplace availability and joint go-to-market programs. The 10-K still lists marketplace pricing as a factor in results.^[8],[5]

   • Owned events

     DASH continues in New York, and Datadog launched more than 100 new capabilities at DASH 2026.^[7]

   • Product-led

     A free trial and a free tier of up to 5 hosts remain the front door, and the 10-K still lists both as a way to drive adoption.^[9],[5]

“Reported” figures come from trade press, not the company, and are shown with their date.

• “Datadog grew without a sales team; the product sold itself.”

  Not supported

  The early years did run without a dedicated salesperson, and the first rep came only after a modest amount of ARR. But Datadog set up an enterprise sales team in 2016, had 474 sales and marketing staff before the IPO, and spent 44% to 45% of revenue on sales and marketing in 2017 and 2018. By 2025 that was about 3,600 people and $956.4M. Pomel rejects the framing himself.^[1],[5],[10]

• “Datadog is just monitoring.”

  Partly true

  It started as infrastructure monitoring in 2012 and observability is still the core, but the filings now list products across security, software delivery, service management, product analytics, and an AI SRE agent, and multi-product metrics show customers buying across them. Datadog does not disclose revenue by product, so the share outside monitoring is unknown.^[1],[5]

• “Datadog's net retention is always above 130%.”

  Not supported

  It held from 2017 to 2022: 141% to 151% before the IPO, and mid-140s on a trailing basis for December 2022. Then customers optimized usage, the trailing rate fell to the mid-110s in 2023, and it had recovered only to about 120% by 2025.^[4],[5]

• “We hear a lot about product-led growth, but I don’t love that term because it implies that sales don’t have a role and that’s very far from true.”

  Olivier Pomel, Co-founder and CEO, Datadog^[10]

  Answer to "Do you think those early constraints shaped Datadog's culture?"

• “it’s a lot easier for users to self-select and to go and start using your product than for you to understand for whom you’re going to be the right thing”

  Olivier Pomel, Co-founder and CEO, Datadog^[11]

  Answer on the closed alpha and open beta, before "You're not a strong believer in the MVP, right?"

• “our products are able to capitalize on the decision-making power of the developer community to initially land with development and IT operation teams”

  Datadog, S-1 registration statement^[1]

  Business, Sales and Marketing section

• Price on the unit that grows with the customer's own business.

  Works when: The product's value scales with infrastructure, data, traffic, or transactions the customer is already growing, and usage is easy to meter.

  Backfires when: Customers can cut usage quickly without losing value, or a few large customers dominate usage, so revenue swings with their cost programs.

  Based on: Inside sales, enterprise team, and multi-product expansion, Usage pricing; Usage optimization, enterprise scale, and AI, Usage pricing

• Make the second product the growth plan, not an upsell of the first.

  Works when: Products share one agent, one data model, and one buyer, so the second product needs no new install or new budget owner.

  Backfires when: Each new product needs a different buyer or a separate deployment, so a cross-sell costs as much as a new logo.

  Based on: Inside sales, enterprise team, and multi-product expansion, Land and expand; Usage optimization, enterprise scale, and AI, Land and expand

• Treat self-serve and sales as one system, not a choice.

  Works when: The entry user can install and see value alone, but the buyer of a large commitment is a senior leader.

  Backfires when: The product needs a long implementation before any value appears, so there is no self-serve stage for sales to build on.

  Based on: Self-serve monitoring for DevOps teams, Product-led; Inside sales, enterprise team, and multi-product expansion, Sales-assist

• Let users self-select early instead of hand-picking pilots.

  Works when: The user who feels the pain can sign up and install without procurement.

  Backfires when: The buyer is a regulated enterprise that cannot trial software without legal review, so design partners are the only route.

  Based on: Self-serve monitoring for DevOps teams, Product-led; Self-serve monitoring for DevOps teams, Sales-assist

Compare with every other company on the GTM matrix.

1. [1]SEC filingDatadog, Inc. Form S-1 Registration Statement (August 23, 2019)
2. [2]SEC filingForm 10-K, fiscal year ended December 31, 2019 (February 25, 2020)
3. [3]SEC filingForm 10-K, fiscal year ended December 31, 2022 (February 24, 2023)
4. [4]SEC filingForm 10-K, fiscal year ended December 31, 2023 (February 23, 2024)
5. [5]SEC filingForm 10-K, fiscal year ended December 31, 2025 (February 18, 2026)
6. [6]EarningsDatadog Announces Fourth Quarter and Full Year 2019 Results (8-K exhibit) (February 13, 2020)
7. [7]EarningsDatadog Announces Second Quarter 2026 Financial Results (8-K exhibit) (August 6, 2026)
8. [8]CompanyDatadog Announces Expanded Collaboration Agreement, Highlights New Capabilities with AWS Across AI, Observability and Security at AWS re:Invent (December 3, 2025)
9. [9]CompanyDatadog Pricing (October 18, 2026)
10. [10]Founder interviewBuilding Datadog against the odds (Q&A from RTP Global's 2023 Founder Summit) (Olivier Pomel, June 30, 2025, at answers to "Do you think those early constraints shaped Datadog's culture?" and "in the early days, you were all engineers and there were no salespeople?")
11. [11]Founder interviewHow to Build and Sell a Product that Customers Love (SaaStr Europa, video transcript) (Olivier Pomel, January 20, 2024, at answers on the closed alpha and open beta, on staying customer-centric, and on event marketing)

Every phase, motion, and myth verdict cites a filing, the company's own pages, or a founder speaking on the record. See the methodology.