Overview
In this episode of The Dashboard Effect, Brick Thompson and Landon Oaks tackle a question that comes up in nearly every data conversation: should we build an internal team or work with an outside partner? The answer is more nuanced than the hourly rate comparison most companies start with, and this episode makes the case that the real cost calculation looks very different once you account for everything involved in building and maintaining a capable in-house data function.
The conversation is direct and practical, drawing on the team’s experience working across a wide range of companies and data environments. For any organization weighing this decision, see how Blue Margin’s Managed Analytics & Insights gives you access to a full bench of specialists without the overhead of building a department from scratch.
What This Episode Covers
The “Batman” Hire (1:04 – 1:40)
Companies often go looking for a single expert who can manage data lakehouses, pipelines, data modeling, and reporting all at once. That person is rare, expensive, and hard to retain. Outsourced partners bring a bench of specialists, giving clients access to the right expertise for each challenge without requiring it all to live in one hire.
The Bench Advantage (1:43 – 2:56)
A team-based approach means problems get solved faster. A managed partner has likely encountered the same technical roadblock dozens of times before, while an internal employee facing it for the first time may feel isolated and spend significantly more time working through it.
Hidden Costs of In-House Teams (3:00 – 4:10)
Covering the full range of necessary skill sets typically requires hiring two to four people, not one. When you add recruiting, onboarding, training, and management overhead on top of base salaries, the cost of building an internal data department climbs quickly and quietly.
Context and Continuity (4:11 – 5:00)
When internal employees leave, their institutional knowledge often walks out with them. Managed partners maintain documentation, wikis, and account history that survive staff turnover, keeping projects on track and reducing the cost of starting over.
Scalability (6:03 – 6:37)
Outsourced partners allow businesses to flex capacity up or down as needs change, whether that means accelerating through a period of rapid growth or pulling back after an acquisition settles. A fixed internal team is much harder to scale in either direction.
When In-House Makes Sense (6:54 – 7:38)
Brick and Landon are clear that there are situations where building internally is the right call. But for companies that do not want to manage the overhead of recruiting and maintaining a full-scale data department, outsourcing typically delivers more value, more efficiently.
Who It’s For
This episode is worth your time if you are a CEO, CFO, or operations leader evaluating how to staff your data function, a hiring manager who has struggled to find or retain strong data talent, a data team leader making the case internally for a different resourcing model, or any organization that has experienced the disruption of losing a key data employee mid-project.
Why It’s Worth a Listen
The build versus buy debate comes up constantly, and it rarely gets examined with this much honesty. Brick and Landon are not simply making the case for outsourcing because it benefits them. They acknowledge the situations where in-house makes sense and focus the conversation on the factors that companies consistently underestimate when running the numbers themselves.
The points on continuity and the hidden costs of turnover are particularly valuable. Most organizations budget for salary and benefits but do not fully account for what it costs when a data engineer leaves six months into a project and takes their context with them. This episode puts a sharper frame around that risk and offers a practical way to think about mitigating it.
If your organization is at an inflection point on data investment, this conversation will help you ask better questions before you make a decision you will spend the next two years living with.
