Key Moments

Dan Siroker at Startup School 2013

Y CombinatorY Combinator
Science & Technology4 min read22 min video
Oct 25, 2013|11,096 views|90|1
Save to Pod
TL;DR

Entrepreneur Dan Siroker shares lessons on building successful startups, emphasizing feedback loops, continuous improvement, and customer focus.

Key Insights

1

The critical importance of paying customers for validating a business idea early on.

2

The immense value of feedback loops for product development and iterative improvement.

3

Building a product that solves a problem you personally experience significantly increases the chances of success.

4

Distribution is a key challenge; find scalable and leveraged methods.

5

Hiring the right people, focusing on those better than the average current employee, is crucial for growth.

6

The 'Universal Startup Algorithm' centers on continuous improvement through feedback, applicable to product, hiring, and fundraising.

EARLY VENTURES AND THE PAIN OF NO CUSTOMERS

Dan Siroker's entrepreneurial journey began in college with a sentiment classification algorithm applied to the stock market. His first solo startup, "Sentiment Solutions," focused on this, but he faced a critical issue: an infinite period before acquiring its first paying customer. This experience taught him the paramount importance of having paying customers to validate a product and a business idea, a lesson that would shape his future endeavors.

LEARNING AT GOOGLE AND THE POWER OF FEEDBACK LOOPS

Seeking to gain essential entrepreneurial skills, Siroker joined Google, where he was a product manager for Google Chrome. This role emphasized the strategic advantage of Google's distribution to gain immediate users. More importantly, he learned the immense value of feedback loops. Direct user feedback on Chrome allowed for rapid, incremental improvements, highlighting what was significantly missing in his previous startup where he lacked customer input.

INSPIRATION FROM POLITICS AND THE BIRTH OF OPTIMIZELY'S IDEA

A pivotal moment occurred when Siroker heard Barack Obama speak about using data, science, and feedback in government. Inspired, he volunteered for Obama's campaign, joining the New Media team. Here, he experienced firsthand the power of A/B testing to drive decisions. The campaign's success with A/B testing, despite reliance on developer resources which created bottlenecks, planted the seed for a tool to democratize this process.

FAILED ATTEMPTS AND THE LESSON OF BUILDING FOR YOURSELF

After the campaign, Siroker and a friend started "Carrot Sticks," an online math game for kids. Despite building a good product and learning from customer feedback, they faced challenges. They weren't the target audience (parents, teachers, kids), making prioritization difficult. Distribution was also a major hurdle, requiring one-to-one engagement with little leverage. This led to a crucial realization: build products you personally need or wish you had.

SPRADLEY AND THE FUNDAMENTAL FLAW OF SOCIAL DISTRIBUTION INCENTIVES

Their next venture, "Spradley," was an attempt to leverage social sharing for discounts. While it was easier to build because it addressed a perceived need from Carrot Sticks, the fundamental model proved flawed. The social capital required for users to spam their friends with promotions was often worth more than the discounts offered. This venture, though also generating feedback, reinforced the need for a truly viable business model and product-market fit.

OPTIMIZELY: THE PRODUCT THEY WISHED THEY HAD

Applying the lesson of building a product they personally needed, Siroker and his co-founder conceived "Optimizely." This tool aimed to make A/B testing accessible to marketers without technical expertise, replicating the power they harnessed during the Obama campaign. An early conversation with a former campaign colleague revealed a significant demand and willingness to pay for such a solution, leading to their first paying customer before writing any code.

GROWTH AND THE REFINEMENT OF THE UNIVERSAL STARTUP ALGORITHM

Optimizely's journey involved a clear focus on continuous improvement, driven by customer feedback on features like targeting and analytics integration. Their hiring strategy also embodied this principle, aiming to hire individuals 'better than the mean' to keep improving the team's average. This iterative approach, coupled with strategic press leveraging their Obama campaign origins and 'David vs. Goliath' narrative against Adobe, fueled substantial revenue and team growth.

INNOVATIVE FUNDRAISING AND BUILDING A WORLD-CLASS PRODUCT

As Optimizely scaled, they innovated in their Series A fundraising by hosting mock board meetings with investors, allowing them to assess the true value of potential partners. This process led them to Peter Fenton, whose insightful questioning, rather than prescriptive advice, proved invaluable. The company continued to evolve, offering a visual editor to create and test website changes, driven by a vision to enable all businesses to turn data into action.

THE POWER OF EVANGELISTS AND THE UNIVERSAL STARTUP ALGORITHM REAFFIRMED

Siroker concluded by emphasizing that while a strong sales team is important, delighted customers who become evangelists are the biggest secret to success. He distilled his entrepreneurial experience into the 'Universal Startup Algorithm': get feedback, use it to get better, focus on continuous improvement, and write your own algorithm rather than just executing existing ones.

Startup Success Principles

Practical takeaways from this episode

Do This

Focus on learning from feedback loops.
Build products you yourself would need.
Prioritize distribution as much as product development.
Hire people who are better than the average ('better than the mean').
Embrace continuous improvement in all aspects of the business.
Define your own 'algorithm' rather than just executing others'.

Avoid This

Don't neglect paying customers and their feedback.
Don't struggle to prioritize without a clear focus.
Don't underestimate the difficulty of user distribution.
Don't get trapped in the 'activity trap' – focus on writing the algorithm, not just executing it.
Don't assume traditional distribution models are the only way.

Time to First Paying Customer

Data extracted from this episode

VentureTime to First Paying Customer
Sentiment SolutionsInfinity months
Carrot Sticks6 months
Spradley1 month
Optimzely1 day

Common Questions

Dan Siroker shares his entrepreneurial path starting from college research in machine learning, his initial failed startup 'Sentiment Solutions', working at Google, his experience on the Obama campaign, and founding multiple companies including Carrot Sticks, Spradley, and ultimately Optimzely.

Topics

Mentioned in this video

More from Y Combinator

View all 362 summaries

Found this useful? Build your knowledge library

Get AI-powered summaries of any YouTube video, podcast, or article in seconds. Save them to your personal pods and access them anytime.

Try Summify free