Good morning, everyone, and welcome to another NachoTuesday! Today, I'm here with Rohan, the co-founder of Datazip. Without further ado, Rohan, could you introduce your company and tell us what you do over there? Hey Andy, thanks for having me here. I'm Rohan. At Datazip, we make data infrastructure and analytics on autopilot. Essentially, any company today that wants to set up and maintain a data infrastructure to answer basic questions like "What's the acquisition cost?" or "What's the total week-on-week cohort?" typically needs two to three different data engineers or a combination of four to five tools. To solve this, we created a single plug-and-play solution that provides the entire infrastructure out of the box and requires no engineers to start or maintain. That's what we do. Sounds pretty straightforward! What led you down this path, and what inspired you to start Datazip? Could you tell us more about your background? It actually started during my university days. I began a startup in the AI NLP space, working on large language models before they were widely known. That experience gave me a deep understanding of how data plays a crucial role in AI. We got into the infrastructure, analytics, and data science aspects even before fully diving into AI. We worked on this for two years and were fortunate enough to be acquired shortly after my university days. Post-acquisition, I joined Rapido, a two-wheeler ride-hailing company, as one of the initial data guys. Later, I joined another startup as the second data person. I've always been involved in data, being one of the early team members at several great startups. These companies started at the seed or Series A stage and have now grown into substantial organizations with hundreds of employees and large data teams. This journey gave me valuable insights into setting up initial data infrastructure, building teams, and establishing workflows. Your background is quite unique. Do you have any insights for companies looking to get started? What lessons have you learned from being part of early-stage startups? One significant lesson I've learned, especially on the data side, is that creating dashboards and insights is just half the job. The real challenge lies in making those dashboards impactful in decision-making. It's crucial to prioritize decisions; if a dashboard doesn't help a user make a decision, it's just a pretty newsletter. The key takeaway is to start with the decisions you want to impact and then determine how data can support those decisions. Data can indeed be overwhelming. Even if you collect a lot of data, if it's not relevant or utilized effectively, it doesn't help in decision-making. Exactly. Collecting unnecessary data is not only unhelpful but also costly in terms of both engineering and infrastructure. While data platforms are essential, it's vital to focus on the decisions that need to be made from the data. When we engage with a client, we don't start by asking if they need a data platform. Instead, we ask what decisions they need to make and how our platform can assist in that process. That's a valuable life lesson: start with the end in mind and work backward to fill in the gaps. It's always easier when you know the answer and can work toward it, rather than trying to create the entire equation from scratch. Exactly. It's much simpler to make the left-hand side equal to the right-hand side than to create the entire equation from scratch. What other lessons have you learned about go-to-market strategies for early-stage startups? What founding principles have been crucial to the success of the companies you've been involved with? A critical insight is the importance of truly understanding your user. By this, I don't mean just knowing them as acquaintances but understanding them deeply, as if you were them. You need to know their journey, where they drop off in the funnel, and why. This knowledge is crucial in correlating data with backend and secondary data. For example, at a ride-hailing company I worked for, we realized that most drivers treated the job as a part-time gig rather than a full-time livelihood. This insight came from closely examining user behavior and data. It's important for startups to look at data early, not just when they have a large user base. Understanding your users and their motivations can significantly increase your chances of success. It's crucial to blend quantitative data with qualitative insights, such as customer surveys or direct interactions, to fully understand your users' needs. The more empathy we have for our customers, the better we can solve their problems and build successful companies. Absolutely. Knowing your customers increases your chances of success. For instance, in my career, I've seen innovative go-to-market strategies where companies focused on a single feature that mattered most to the user. This approach often leads to onboarding customers effectively and then cross-selling and upselling additional features later. Especially for complex platforms, onboarding is a critical aspect, and it can be challenging for SaaS startups. Unlike simpler products, SaaS involves getting an entire organization onboard, not just individual users. Indeed. It's particularly challenging in B2B scenarios, where you're dealing with entire organizations. It's not just about acquiring customers but also about retaining them and reducing churn. This involves mapping user cohorts, understanding retention rates, and ensuring that your platform continues to meet the users' needs over time. Retention is key in any business, especially in SaaS, where recurring revenue and low churn rates are crucial. Understanding customer behavior and adapting to it is essential for long-term success. Exactly. The analytics for B2C and B2B startups can be quite similar in terms of tracking user behavior and retention, although the metrics and timeframes might differ. For instance, B2C startups might have different retention benchmarks compared to B2B SaaS companies, but the principle of understanding and tracking these metrics remains the same. At the end of the day, we're in the business of people, whether we're selling a product or a service. Understanding our customers' needs and aligning our offerings with those needs is the key to success. I couldn't agree more. My past experiences have taught me invaluable lessons, and I've tried to incorporate them into our current go-to-market strategies. Learning from different industries and applying those insights to your own can lead to innovation and success. (13:00) "We have basically suggested the same GTM strategy to all our current clients as well. When we go in, we look at their data—not from our perspective, if they don't want us to, but most of the time, they ask for help. They show us their numbers and ask us to figure out what to do next. All the learnings from past experiences come into play. We understand, for example, that if a cohort looks a certain way, they should take certain actions, or if their marketing spend is allocated in a particular manner, maybe it should be adjusted. All those different facets that we've learned over the years come into play." (13:35) "That's a great point. It's great that you guys do that onboarding too because if you can ensure the success of your customers, they are much more likely to stick around with you. You're the ones who got them there, so to speak." (14:03) "We love our customers. We respect them to the core and understand where they are coming from. As I said, if I'm advocating for understanding your customers like they are you, believe me, I'm trying to do that every day." (14:34) "That empathy point is key. If you can empathize with people, you can better serve their needs, and they will pay you for it. A basic question is, why should startups care about data? It seems obvious, but it needs to be stated. I once met a CEO of a startup who said, 'I don't need data; I know social.' I told him, 'You run a tech company, and your onboarding is 99% of people leaving and never coming back. You need data.' So, aside from that anecdote, why should startups care about data, and where can they get data from to make better decisions?" (15:05) "Let me answer this in a more comic way: startups should care about data if they don't want to be a startup five years down the line. If they want to be a company, they need to care about data and know what's happening. My mentor back in college gave me a great analogy about data and companies: if I'm a founder today and I don't look at data, it's like being given a Rolls-Royce or a Ferrari and driving it blind. My startup is the Ferrari, but without data, I don't know where it's going. Would you treat a Ferrari like that?" (16:11) "Now, onto your second question about sources of data. A few sources we've found effective for our users include event streams. Understanding what your user is doing on your platform, how they move from one page to another, or one click to another, is a gold mine of product usage information. Another source is the CRM, which captures information like lead stages and user demographics. If you're using a CRM like HubSpot or any other, most have capabilities to note down information such as the organization or job role of the person. This information can be crucial." (17:32) "What do you recommend for startups? What are some initial sources where they should get their data from, and what key metrics should they start to track from day one, especially if they're a B2B SaaS company? Can you hear me, Rohan?" (18:12) "Hey, sorry about that. For early-stage startups, CRMs are definitely a good source of data. Other sources include event tracking systems like Mixpanel, Amplitude, or Google Analytics. Your backend data, such as where you're storing orders or customer information, is crucial as it forms the cohort of your customer base. Additionally, your payment gateways and marketing platforms are key sources. If you're doing marketing and not collecting your marketing platform data, you're just throwing money away. You need to look at that data to understand what happened with your campaigns." (19:14) "These are common sources for a typical SaaS startup. The metrics they should focus on include acquisition cost, which requires data from multiple sources. Acquisition doesn't just come from a direct cost like a paid advertisement; it involves salaries for sales folks and other channels. It's essential to calculate the total acquisition cost and average revenue per user over a specific period. Retention metrics are also vital. You need to know how long users stick around because, without this information, you might be growing but still losing money." (20:23) "You should also track marketing and sales metrics, like the conversion rate from a marketing qualified lead (MQL) to a sales qualified lead (SQL) and then to an actual user. These are fundamental metrics. If you don't know them, you're just burning money." (21:00) "It's more than just asking ChatGPT or reading a blog post; it's about understanding why this data is important. For example, churn rate is crucial. You might spend a lot on acquiring a user, and initially, you might not make money, but if they stay with you for a year or two, that's where you make your profit. Tracking metrics like Net Promoter Score (NPS) can help you understand customer satisfaction and likelihood to refer others, which is invaluable." (22:13) "These are the decisions that any startup should list first and then determine what data they need to support those decisions. It's essential to move from decision to data, not the other way around. This shift in approach is fundamental." (22:45) "Don't view bad metrics as a negative. Look at them as opportunities to find solutions. The fun of startups is in solving these problems. When you identify the problem areas, you can make the right decisions to improve." (23:14) "True. If you're in trouble, it's better to recognize it early by looking at the signs around you, rather than ignoring them. So, do you have a great example of a company with an excellent data strategy? Who do you think executes their data strategy particularly well, aside from your own company?" (23:45) "A couple of great examples include Uber, which I've followed a lot. Uber has been a significant inspiration in how they use data effectively. If you look at their research (24:18) "Google has open-sourced a lot of their data strategies. However, their approach is quite methodical; it's essentially a numbers game for them. When they enter a new market, they follow a specific Standard Operating Procedure (SOP) for onboarding a new city. This SOP is not flexible, even if the data suggests otherwise. They have categorized each city they want to enter into five to six buckets, with a distinct SOP for each bucket. This level of understanding of user demographics and operational procedures exemplifies a great data strategy." (24:52) "Google's approach is impressive because they have spent years—almost a decade—refining their strategy. They recently posted their first profitable quarter because they understood what needed to be done to reach that point. Each locality has its unique characteristics; for example, in New York, trips are often short and frequent, whereas in LA, trips are longer due to traffic. Surge pricing and dynamic pricing decisions must account for these differences, which is a complex data problem." (25:26) "To handle dynamic pricing, surge pricing, and driver incentives effectively, Google needs to gather comprehensive data. They have captured this essence well, making them a strong example of a company with a great data strategy." (26:30) "Another example is a cool insurance tech company we’re working with. They have an excellent data strategy; they know exactly how much insurance they want to underwrite and how to manage the incoming data. Since we provided them with our platform four months ago, they have grown twice as much. For a Series B startup, that's impressive growth." (27:09) "AI is often touted as a game-changer, but it only works well with a solid data strategy in place. Many people treat AI as just another feature rather than integrating it into their business model. Upstart, a publicly traded company, is a great example. They use around a thousand data points to disrupt the traditional credit system. Their approach to evaluating creditworthiness with multiple variables showcases how AI can be effectively utilized when paired with the right data." (27:59) "Consistency of payments and various other factors contribute to assessing creditworthiness. Upstart's use of AI in this context highlights how it should be applied appropriately within a business model." (28:31) "To provide some insight into our own product, we recently launched an AI Copilot. It functions as a unified application connecting various data sources in a no-code manner. It manages storage on clients’ own AWS or Azure clouds, enables data cleaning and transformation, and generates dashboards and reports. What previously required multiple tools and data engineers is now offered as an out-of-the-box solution." (29:07) "We realized that AI can streamline this process further. For example, if you're a business user needing data from Facebook Marketing, our AI assistant can connect to Facebook, clean the data, and provide insights. Traditionally, this process involved an analyst, a data engineer, and several weeks of waiting. Our AI assistant shortens this cycle significantly, making data-driven decisions faster and more efficient." (30:18) "In today’s fast-paced business environment, working efficiently is crucial. Our AI assistant helps by managing the entire data workflow, from connection to cleaning to insights, all while providing necessary context. Context is vital for AI to deliver accurate results. It must understand where the data comes from, who is querying it, and how recent the data is." (31:23) "AI tools, like text-to-SQL, are becoming more prevalent, but their effectiveness depends on context. Understanding the data's source and relevance is essential for providing accurate and actionable insights." (31:56) "I appreciate you coming on today, Rohan. I could probably talk with you for hours about these topics. Thank you for sharing your insights and experiences." (32:34) "Thank you all for joining us today. Rohan definitely knows his stuff when it comes to data. If you're serious about your data strategy and want to accelerate the growth of your startup, leveraging the right data sources is crucial. DataZip is one of the best platforms out there to help you achieve this. If you're looking to get your data strategy in gear without spending a fortune on data scientists—something many startups struggle with—be sure to check out Rohan's platform." (33:00) "We already have a fan in the comment section supporting your company, Rohan, so there are a few recommendations already in place. Once again, thank you for joining us today. For anyone interested in DataZip, you can find them on NachoNacho's B2B SaaS Marketplace, the premier place to buy SaaS, or visit their website directly to support this great business." (33:27) "Thanks again, Rohan, for coming on today. And thank you all for watching. See you next time!"