Your marketing manager needs to understand why their latest campaign’s MQLs (marketing qualified leads) aren’t converting to sales at the usual rate. They go to check the lead quality data in their marketing automation platform, which shows healthy engagement scores. But when they try to see what’s happening after these leads are handed to sales, they hit their first wall: the sales team’s conversation data lives in a separate tool that marketing can’t access. So they message the sales team for help. While waiting, they try to check if these leads are at least using the product differently than previous cohorts. That’s when they hit the second wall: the product analytics platform doesn’t connect to the marketing automation system, so there’s no way to match marketing campaign data with product usage data. Several hours (or days) and Slack messages later, they finally get spreadsheet exports from all the tools, to try and get the insights they need. But they still need to normalize the data to make it comparable. Something that should’ve taken 30 minutes, ended up taking a full day (or more), multiple conversations, and multiple stakeholders to hunt down. Meanwhile, if something is actually wrong with the campaign targeting or messaging, it’s still running, potentially generating more poorly qualified leads. This is a great example of a data silo: a set of data that’s isolated from the rest of the company – it’s accessible only to a small group, doesn’t integrate with other systems, and can’t be used strategically by the broader organization.I recently sat down with Samantha Riel, founder of Balsam&Cedar, to break down data silos and instead unify your team around data, so they can leverage it to make better decisions and ultimately, improve your company’s performance. After 20 years of helping B2B companies scale from startups to IPO, she’s seen how data silos can silently grow in organizations, and more importantly, how to fix them. As the example above shows, these silos cost your company time and money. How much? Samantha said it’s impossible to put an exact dollar figure on it, but it’s a lot. Disconnected data impacts everything from strategic planning to daily operations. When your team can’t efficiently share and access data, you’re: Unable to scale quickly and efficiently Wasting hours on manual processes Risking misinterpretation of critical metrics Missing opportunities to catch problems early Unable to forecast accurately Listen to the interview Listen to the full interview on Spotify or Apple Podcasts. How data silos form Data silos tend to build slowly over time, and are caused by a number of factors: Tooling Decisions: A marketing leader finds the perfect tool that meets 100% of their requirements. It doesn’t integrate with your other tools, but they decide to use it anyway. Multiply this decision across departments over 5 years, and you end up with a fragmented tech stack. Leadership Changes: New leaders often bring their preferred tech stack from previous companies, implementing these tools without fully evaluating existing systems or integration needs. Status Quo & Egos: Sometimes people get stuck in the status quo or struggle to admit their processes need improvement. It takes emotional intelligence to say, “Maybe this process I built myself isn’t working the way it needs to be.” Crisis Decisions: When teams are under pressure to solve an immediate problem, they might “throw their credit card across the internet” to get a quick solution, bypassing proper evaluation. Companies with 100-500 employees typically use 172 apps actively. That number jumps to 644 apps for companies with 10,000 employees. But here’s the kicker – those numbers only include tools the company knows about. There are even more “ghost tools” haunting the company. These “ghost tools” are ones employees need to complete their work but aren’t officially sanctioned by the company. When Samantha surveyed employees at a previous company, many reported using 10+ additional unauthorized apps just to get their jobs done. Managing access to data Before we get into breaking down data silos, you might be wondering if the result is giving everyone access to everything. Not necessarily. While we definitely believe that transparency and access to data tends to help your team make better decisions, it doesn’t necessarily mean that you’ll share everything. So when the silos are torn down, how do you decide who gets access to what data? Samantha recommends asking this simple question: “Will someone having access to this data enable them to be more effective in their work?” The answer will depend on your company’s unique structure, but here are some helpful guidelines she shared: Senior leadership needs access to everything for strategic decision-making Directors and VPs need all information that flows up to them Individual contributors need access to data relevant to their role (e.g., SDRs might not need full pipeline visibility) Basic systems like CRM and marketing automation should be widely accessible Industries and regulations may impact what can be shared Making your data work together The way you integrate your data should evolve with your growth, according to Samantha. While every company’s needs are different, she shared a clear framework for thinking about integration at different stages. For companies under $100M, start with the basics: make sure your core systems can talk to each other. This means prioritizing connections between your CRM, marketing automation, and customer success platforms. Look for practical opportunities to remove friction – places where teams are doing manual work that could be automated through better system integration. Once you cross $100M in revenue, your needs typically become more sophisticated. This is when you might consider implementing data lakes or data warehouses, building more robust data infrastructure, and creating more complex relationships across your business data. But what does good integration actually look like in practice? Samantha shared several examples: When a customer gives you a low NPS score, it automatically triggers workflows for immediate follow-up Customer success teams can see the full sales context when onboarding new customers Product teams get real-time feedback from customer support interactions Marketing can automatically send targeted emails based on how customers are using your product She also shared a cautionary tale. One company’s customer success platform, which contained all their post-sale customer data, couldn’t integrate with their email system. As a result, every customer communication had to be managed manually, making it impossible to send timely, targeted messages based on product usage or engagement patterns.This example perfectly illustrates why integration matters – without it, teams waste hours on manual work, miss opportunities to engage customers at the right moment, and can’t scale their operations effectively. Breaking down silos, step by step Before you dive into technical solutions or start ripping out old systems, Samantha recommends a methodical approach to breaking down data silos. Here’s how to get started: Step 1: Start With Culture Create an integration-first mindset across your organization. This means: Establishing that new tools must integrate with existing core systems to be considered Getting leadership alignment that disconnected tools create long-term costs Making it clear that data accessibility and sharing is a company priority Setting expectations that all major system decisions will be evaluated based on how well they connect to your existing tech stack Step 2: Define Your Metrics This step is foundational. You can’t unify around data if everyone defines it differently. The goal here is to get all strategic leaders to create clear, documented definitions for key metrics (MQLs, SQLs, etc.), agree on how metrics are calculated, and define what data points feed into each metric. If you’re not sure if you need this stage, Samantha recommends asking different leaders how they define key metrics. If you get different answers, stop everything else and fix this first. Step 3: Audit your tech stack In this stage, you want to map your current tech stack. List all official tools and systems you use, and document how data flows between systems. Then identify critical gaps in integrations: look for redundant systems or overlapping functionality, and note manual processes that could be automated. You might also create a simple survey to learn about the “ghost tools” each team is using in their workflow. Step 4: Make data part of daily life Once you make sure your data can work together, don’t let it live in isolation. Build it into regular workflows: Connect OKRs directly to metrics you’re tracking Make data review part of regular team meetings Create shared dashboards for cross-functional metrics Build regular reporting cadences Encourage teams to use data in their decision-making Getting buy-in for change This is where many initiatives fail. Understandably, people get defensive about processes they’ve created or tools they’ve chosen. To navigate this, Samantha recommends starting with a safe space for honest discussion. Ask your team: “Can anyone tell me a process today that is functioning with 100% accuracy in a very smooth way?” You’ll typically get very few responses. Then ask about processes that aren’t working well. Key principles for managing the change: Frame it around enabling growth, not fixing mistakes Acknowledge that past decisions made sense at the time Nothing should be off the table for change, everything should be up for discussion Focus on reducing friction, not finding fault Celebrate quick wins and improvements as they happen Share examples of how better data access has helped other teams Implementation tips: Start with one department or process to prove the concept Document and share early successes Address concerns about job security early and directly Keep the focus on making everyone’s job easier Create clear timelines and milestones Assign clear ownership for each part of the initiative Samantha says that breaking down data silos isn’t a one-time project, it’s an ongoing commitment to making data more accessible and useful across your organization. The goal isn’t perfect integration of everything, but rather ensuring that teams have access to the data they need to work effectively. The result: faster speed to insight When you break down data silos, you remove friction throughout the customer lifecycle: Marketing and sales teams work from the same data Sales can smoothly hand off to customer success Customer success has all the context they need Product teams get real-time customer feedback Everyone can make faster, better decisions When everyone is using the same “data set” and leveraging it to make more informed decisions, the result is faster speed to insight.