Oct 12, 2023 · 1h 14m · in-depth
How to leverage intuition, customer support, and raw effort | Colin Zima (Omni & Looker)
0:00 / 0:00
▶ Watch on YouTube →
gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions
In this episode of In Depth, Omni CEO and former Looker executive Colin Zima shares tactical playbooks on scaling enterprise software companies through white-glove customer support, conviction-led product prioritization, and seizing architectural shifts in modern data infrastructure.
How this conversation actually went
Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. Brett holds 14.5% of the talking time here. How this is scored →
Brett as informed peer 4.0 Guest teaching 4.9 Guest disagreement 1.8 Brett pushing back 1.9
speaking balance: gold is Brett, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 33:51 Direct rejection of customer-dictated product roadmaps
Colin forcefully rejects conventional prioritization wisdom, stating he has no deeper disagreement than letting customer requests dictate product strategy.
Hardest push from Brett ▶ 46:50 Pushback on internal misalignment from intuition-led leadershipBrett directly challenges Colin's reliance on taste and judgment, pressing him on whether it creates severe internal alignment and clarity issues for the team.
Biggest teaching moment ▶ 51:17 Deconstructing Looker's 24-quarter pipeline predictabilityColin breaks down the exact mathematical mechanics of Looker's sales predictability, showing how 30% intra-quarter and 70% pre-quarter lead generation powered their streak.
Brett holds their own ▶ 48:35 Citing First Round benchmark across 450 portfolio companiesBrett demonstrates his institutional vantage point by citing Looker as the singular exception across First Round's 450-company history that never missed a quarter.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | Brett as informed peer | Guest teaching | Guest disagreement | Brett pushing back | Why |
|---|---|---|---|---|---|---|
| Colin's Entry to Looker and Customer Discovery | 3 | 4 | 1 | 1 | Brett asks open-ended questions about Colin's entry to Looker. Colin explains his transition from Hotel Tonight to visiting 75 of Looker's first 100 customers for field discovery. | |
| Looker's Philosophy on Empathy and High-Touch Support | 4 | 5 | 1 | 1 | Brett probes whether high-touch support was an intuitive choice or strategic thesis. Colin outlines Lloyd Tabb's philosophy of software as an act of empathy and explains why support spending generated outsized financial returns. | |
| Executing Unscalable Support and Managing Scale | 4 | 5 | 2 | 2 | Brett presses on whether this extreme support model could survive scale. Colin admits absolute quality drifts but argues top-percentile standards were maintained through deliberate resourcing and unscalable interventions. | |
| Support Economics, Hiring Profiles, and Internal Pipelines | 5 | 5 | 2 | 2 | Brett asks for the specific business parameters where high-touch support is optimal. Colin breaks down unit economics (25k ACV, 90% margin) and explains using college STEM hires as an internal talent pipeline for engineering and sales. | |
| Looker's Kitchen Table Culture and Training Model | 3 | 5 | 1 | 1 | Brett asks how Looker's cultural practices carry over into Omni. Colin describes Looker's kitchen table peer-learning environment and demonstrates how sub-second response times in Slack transform support into product discovery. | |
| Product Prioritization and the Benevolent Dictator Model | 4 | 6 | 4 | 2 | Brett inquires about feature prioritization rubrics. Colin rejects formal scoring frameworks and customer-voted roadmaps, arguing strongly for benevolent dictatorship where product leaders set strong opinions and defend unpopular trade-offs. | |
| Strategic Roadmaps, Looker 500, and Embedded Analytics | 4 | 5 | 2 | 2 | Brett pushes to understand how Colin chose when to go upmarket. Colin explains managing resources at the margin, forcing initiatives like Looker 500, and learning from experimental bets like embedded analytics. | |
| Downsides of Intuitive Decision-Making and Strategic Focus | 5 | 4 | 2 | 4 | Brett challenges Colin on the explicit organizational downsides of relying on intuition over structured frameworks. Colin concedes that dictatorial prioritization creates alignment friction and internal team churn. | |
| Looker's Predictable Sales Pipeline and 24-Quarter Streak | 5 | 5 | 1 | 2 | Brett highlights Looker's record across First Round's portfolio of never missing a quarterly plan over 24 quarters. Colin breaks down pipeline mechanics and reflects on how zero-miss targets may constrain ambition. | |
| Market Shifts and the Architectural Foundations of Looker | 4 | 6 | 2 | 2 | Brett asks Colin to define what made Looker's product fundamentally great. Colin details the architectural timing of Redshift and Snowflake alongside cloud data models, contrasting it with the market opportunity prompting Omni's creation. | |
| Customer Demos and Scoping Omni's Minimum Viable Product | 4 | 4 | 1 | 1 | Brett asks how Omni scoped its MVP given extensive product surface area. Colin explains doing over 500 live customer demos in a year to pinpoint the minimum viable bundle combining SQL, BI, and modeling. | |
| Lightning Round: Startup Selection and Stellar Employee Traits | 3 | 5 | 1 | 1 | Brett begins a lightning round on startup selection and employee excellence. Colin advises viewing job selection as a single concentrated venture bet and highlights extreme proactive effort. | |
| Founder-Market Fit, Product Management, and Career Mentors | 4 | 5 | 3 | 3 | Brett explores PM hiring and founder-market fit. Colin offers the contrarian view that product management is not a craft but rather deep domain obsession, prompting Brett to playfully summarize Colin's philosophy as 'just be good.' |
Statements from this episode (25)
Zima: Emergent user convergence on flexible software signals special product value
“I think that is when you know that you're doing something really special in terms of like company building is when you hand them something that can be used in a lot of different ways. And a lot of people are sort of circling around the same types of ways to us…”
Looker required all employees to do customer support up to 100 headcount
“So we had everyone in the company doing customer support all the way up to, you know, like a hundred employees. And it wasn't every single day, all employees, but everyone got a taste of support.”
Looker scaled support staff linearly with customer count through acquisition
“We were hiring linearly with support with customers all the way through our acquisition. And like support was the fastest growing org inside Looker.”
Zima: Looker's 6% spend on support drove 15-20% lift across metrics
“Even post Google, I wrote up a case study that said Google should retain Looker support because this six percent that we spend on support as a percentage of revenue is worth 15 or 20% in terms of, you know, like growth, retention, every kind of metric that we …”
Zima: Omni supports third-party software in Slack to build trust
“If someone wants me to help them write SQL, or like, I don't know, debug something in Excel, like I'll go do something I do support for other people's products in Slack, because I just think it shows like a positive association with how much we care about thei…”
Zima: Looker's support quality inevitably declined as team reached 80 reps
“When you're training your 70th through 80th support reps with like support reps that you've hired in the last two years, it's really, really hard to maintain a bar where just everyone knows everything about everything. So the universal answer is like, yes, cer…”
Zima: $25k ACV, 90% margin software startups can afford high-touch support
“I think at like a 25 K ACV, 90% margin software business. So like a classic sort of like high end SMB, low end, mid market sales. Everyone can afford this.”
Zima: Looker seated up to 50 support reps together at one table
“So one of Looker's cultural values was called the kitchen table. And it was this idea that the support team literally used to physically sit next to each other. And this was up to, you know, like it'd be a 50 person long table and everyone be sitting next to e…”
Zima: Side-by-side support training cannot be replicated any other way
“And so it was this like active learning by doing next to each other that I don't think could be replicated in any other way.”
Zima: Sub-minute support response times drive massive customer dialogue
“You will be surprised at the difference in the conversation when you respond in one second versus one minute versus 10 minutes. Because when someone asks a question and you respond in 1:02, it immediately turns into a conversation. And It is very likely that y…”
Zima: Rigorous feature prioritization frameworks are a waste of time
“I just, I really think it's hard to come up with a great, like scoring framework for how to pick out the next 10 features, because I found that every single time that we tried to apply some sort of rigorous, like scoring framework to difficulty and reach and Y…”
Zima: Looker deprioritized app crashes for two years to build features
“Whereas like I had to go to the support team over and over again for probably a period of two years and say the app crashing is not a priority right now because I want to do more feature work and we can suffer through this and I know it's bad, but I don't thin…”
Zima: Product leadership at scale means driving a few unpopular priorities
“And I feel like being able to really crisply state a small handful of unpopular priorities
Almost is, I think, a good description of product leadership at scale.
It's sort of like, where am I reallocating resources from steady state?”
Zima: Embedded reporting grew to 30% to 40% of Looker's business
“And like, lo and behold, four years later, it's 30 or 40% of Looker's business.”
Zima: Looker could have taken bigger risks aligning with dbt or acquiring
“And I think that we actually made some of those mistakes, even like, I think that we could have taken some big risks to align more closely with DBT or to acquire possibly, or to like move into different areas of the stack that we didn't do. Because we were com…”
Zima: Looker calculated it could hit $1B revenue in core product alone
“Like we calculated how much revenue we could get in our area of the product. And it's kind of nice when you set up a company and you're like, we can get to a billion dollars in revenue doing exactly what we do today, just more of it.”
Berson: Looker never missed a quarterly target across First Round's history
“Is I think in, in the history of first round across 450 businesses, it's the only company that didn't miss a number all the way through sort of massive exit or sort of somewhat directionally, that's the case.”
Zima: Looker hit its 80th-percentile target for 24 consecutive quarters
“We sort of just pick like the 80th percentile of a great business. And we said, that's our goal. And then we drew that line out for 24 quarters and hit it, which is, you know, no, no startup founder is going to be like, I am not happy with that.”
Zima: Looker closed 30% of deals intra-quarter and 70% from prior quarters
“Like, we closed 30% of our business intracorder. Which means that they were leads generated inter quarter. The 70% was generated pre quarter. And like, I think 40% of it came from the previous quarter and 30% of it came from the three quarters previous.”
Zima: Looker's original product was built via the command line
“The original product was literally a model that you had to build in the command line. It just was technically elegant for the exact problem that it was solving. And it really aligned well with the user that we were selling it to.”
Zima: Looker and Tableau Acquisitions Left a Strategic Void in BI
“Looker and Tableau got acquired within a week of each other. And so again, now this gigantic market that is BI that has existed for the last 40 years has a pretty large sort of strategic hole in terms of there's not a new generational company that exists in BI…”
Zima: Looker Failed to Adapt to Modern Data Ecosystem Shifts
“Looker was not able to adapt to the changes that its sort of founding and popularization had resulted in.”
Zima: 20 of Omni's 28 employees previously worked at Looker
“I mean, I think one interesting aspect is our company as is 28. And I think 20 of us spent time at Looker and five of us spent time at our co-founders company.”