Sep 8, 2021 · 18m · top-founders
DevOps SaaS Data Tool Breaks $4.9m ARR Bootstrapped Using Its Marketplace To Grow
0:00 / 0:00
gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions
In this interview with Nathan Latka, Keboola co-founder and CEO Pavel Dolezal explains how the data engineering platform scaled to $4.9M ARR through a bootstrapped business model, consumption-based pricing, and an open developer marketplace.
How this conversation actually went
Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. Nathan holds 23.3% of the talking time here. How this is scored →
Nathan as informed peer 6.0 Guest teaching 2.3 Guest disagreement 1.2 Nathan pushing back 3.2
speaking balance: gold is Nathan, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 7:49 Pavel defends rapid sales hiring strategy
Pavel directly pushes back on Nathan's premise that hiring 10 reps was premature, clarifying that early experimentation had already justified scaling immediately.
Hardest push from Nathan ▶ 7:45 Nathan challenges hiring 10 reps without quotasNathan refuses to accept scaling unproven sales reps and directly questions why Pavel didn't stick to a conservative 2-3 person test group.
Biggest teaching moment ▶ 11:18 Pavel educates on open-source ecosystem incentivesPavel explains the enterprise motivation behind 400 external engineers writing custom integrations without direct compensation, reframing Nathan's assumptions about partner economics.
Nathan holds their own ▶ 13:56 Nathan instantly derives expansion rate from churn metricsNathan demonstrates sharp SaaS fluency by computing on the fly that 20% logo churn alongside 111% NDR implies an underlying 31% expansion rate.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | Nathan as informed peer | Guest teaching | Guest disagreement | Nathan pushing back | Why |
|---|---|---|---|---|---|---|
| The Origin Story of Keboola | 5 | 2 | 1 | 2 | Nathan probes into the transition from a services agency to a SaaS product and drills down into the cap table split. Pavel explains the 3-year earn-in agreement openly without resistance. | |
| Customer Segmentation and Consumption-Based Pricing | 6 | 3 | 1 | 4 | Nathan presses Pavel to clarify blended MRR figures when the initial tier breakdown gets muddy. Pavel breaks down the consumption-based compute minute pricing model. | |
| Scaling the Sales Organization and Outbound Playbook | 6 | 2 | 2 | 5 | Nathan challenges Pavel's decision to hire 10 sales reps simultaneously without pre-established quotas instead of testing with a small cohort. Pavel explains the initial validation done with their inside sales lead. | |
| Bootstrapping Philosophy and Historical Revenue Milestones | 7 | 4 | 1 | 3 | Nathan synthesizes the company's historical revenue progression and questions the incentive model behind third-party developers building 1,400 apps. Pavel educates him on how agency partners and internal teams productize data workflows. | |
| Retention Dynamics and Word-of-Mouth Expansion | 7 | 2 | 1 | 3 | Nathan prompts a distinction between gross logo churn and net revenue retention, immediately calculating the implied expansion rate. Pavel confirms word-of-mouth referral dynamics from serial tech leaders. | |
| Reinvestment Strategy and Future Capital Allocation | 5 | 1 | 1 | 2 | Nathan inquires about bottom-line margins and future capital plans before transitioning into the Famous Five rapid-fire questions. Pavel clarifies that their current negative margin is an intentional reinvestment of accrued cash reserves. |
Statements from this episode (15)
Dolezal worked three years to earn his equity stake in Keboola
“I had to make, I had to actually, I had to work for my share for three years.”
Keboola's three co-founders split equity evenly
“Oh, we, all three of us have a split three percent, one third.”
Keboola added over 50 new paying customers in 2021
“Last year it was it was like around 115, 124 actually last year. And this year we actually already added over 50 new customers.”
Keboola reaches $4.9 million in annual recurring revenue
“We are now actually doing 4.9 ARR, you know, on annualized basis.”
Keboola saw 900 projects in its first six months of self-serve
“In the first six months we got 900 projects started, and in July, we got 260 projects started.”
Keboola scaled its sales team from one to ten in eight months
“Last December we had one person in sales, then we hired three people into sales in first half of this year. And as of today, we actually have 10 people as of August.”
Keboola revenue: $2.4M in 2018, $4.1M in 2019, $4.7M in 2020
“In 2020, we actually ended four seven, 2019, we actually ended 4.1, and 2018, we actually ended 2.4.”
Keboola bootstrapped to $4.9 million ARR without outside capital
“No, it's all bootstrap until now.”
Small companies lack the setup to run multiple experiments simultaneously
“When you are small, you cannot do five different experiments because your organization is not really set up for it. If you do five, you cannot really devote attention to it. So you have to be very, you know, you have to be very precise and very diligent in wha…”
Keboola leverages 400 marketplace developers alongside 28 internal engineers
“We have 28 28 engineers and plus 400 actually engineers in our marketplace.”
Third-party developers built 1,400 applications on Keboola's marketplace
“So there was, there's actually 1400 apps in our marketplace, which was created by 400 different developers.”
Keboola's logo churn jumped to 21 percent in 2020 due to COVID
“Until last year we had eight, 10%, eight to 10%. And last year it was like 21% in the base. Of customers who either, you know, like went bankrupt or, you know, like totally scaled down their businesses or stopped for a bit of the year.”
Keboola net dollar retention was 105 percent in 2020, rebounding in 2021
“Net retention rate, you know, in dollars was actually, 2019 was 119 and 2020 was 105%. Already now we are in 100 111%. So like 11% on top this year.”
Bootstrapped Keboola generated $600,000 in profit in 2020
“We actually, until last year, end of last year, we were profitable, like and we had like 600,000 profit”
Keboola intentionally shifted to unprofitability in 2021 to fund aggressive hiring
“When we started hiring all those new people in June, July, we actually made the decision that we will go within the next six months, we will not be profitable and reinvest all of that because we were, you know, hoarding the cash from last year.”