For most of my career in digital analytics, I’ve done a bit of everything.
I’ve managed tagging setups and cookie banners, built dashboards for every stakeholder group imaginable, occasionally written BigQuery SQL to dig into raw data, and tried to help the business make smarter decisions every now and then.
In theory, that sounds like a full-stack analytics dream.
In practice, it’s exhausting.
Doing everything demands a lot (maybe too much) from anyone.
Yes, it’s possible. But it comes with trade-offs.
Something always gets less attention than it deserves: data quality, analysis, compliance, or alignment with business goals.
And as analytics becomes more strategic, and AI and automation enter the picture, this generalist setup gets even harder to sustain.
For years, I thought two roles were enough for a digital analytics team:
- a technical analyst and
- an analyst.
That model made sense when tooling was simpler and expectations were lower. But today, that’s not enough.
Here’s how I’d break it down now.
1. Technical analyst / digital analytics engineer
Owns the data collection, integration, and compliance layer.
This role is about more than just tagging. It’s about building a solid foundation: a data pipeline that’s stable, scalable, compliant and AI-ready.
In practice, they:
- Design a robust data layer
- Set up tagging
- Integrate CRMs, CDPs etc with websites
- Ensure conversion and customer data are passed to analytics and advertising tools cleanly
- Are responsible for consent management
- Prepare and model data for machine learning
Without a technical analyst, the team fixes broken tracking, works with unreliable attribution data, or faces legal risk. And this happens all the time in the real world.
2. Analyst
Turns data into insight—and helps stakeholders make better decisions.
This role is about clarity. Not just reporting, but shaping the questions that lead to better outcomes.
An analyst:
- Builds dashboards that people use
- Adds context to data
- Segments users based on behaviour or value
- Spots patterns that drive product, commercial, or marketing decisions
- Designs and analyses experiments
- Starts working with predictive analytics and machine learning models
A great analyst doesn’t flood people with numbers. They show what matters and why. And they partner with data scientists or use low-code ML tools to enrich their insights.
And, of course, they should also be willing to analyse qualitative data.
3. Analytics strategist / product owner
Keeps the analytics effort focused on business outcomes.
This is the missing piece in many teams. Without this role, analytics is reactive: we answer questions, add tags, build dashboards and fix broken setups.
But with a strategist, the team can:
- Prioritise the work
- Connect analytics with business strategy
- Own the roadmap
- Identify where AI, automation, and advanced analytics can create value
- Balance short-term requests with long-term capability building
This person might come from marketing, product, or analytics as long as they can speak business and data.
Real-world overlaps and tensions
Of course, the lines aren’t always clear.
In lean teams, the same person might do everything. Even in larger teams, roles can overlap or even create friction.
Analysts are sometimes pulled into technical tagging work, which limits their time for insight generation.
Technical analysts may be expected to build dashboards when they prefer to focus on infrastructure.
Strategists might clash with stakeholders who expect faster delivery, not better prioritisation.
These tensions are a sign that your team needs clarity.
Clear definitions, shared goals, and mutual respect make all the difference.
But if one person is responsible for everything, these tensions can cause burnout.
Is this three-role model realistic?
In many teams, there’s still just one analyst doing it all.
They set up tracking, build dashboards, answer questions, debug consent banners, and figure out how AI might fit in.
They’re doing everything, everywhere, all at once.
So yes, this three-role model can sound like a luxury.
But whether it’s a realistic or distant dream depends on two things:
1. How important is your digital presence to the business?
If your website or app is central to acquiring customers, making sales, or providing services, analytics isn’t a “nice-to-have.”
It’s part of your core operations. It deserves real support.
2. How much are you spending on advertising?
If you spend hundreds of thousands per month on media, even a 1–2% optimisation can add tens of thousands in added value.
More spend means more complexity, and a greater need for specialists who can protect and grow the investment.
One skilled person might be enough if your spending is in the low thousands, but only if the scope is realistic and the trade-offs are understood.
Career takeaways: Where do you fit?
If you work in digital analytics or want to build your career in this field, this model isn’t just theory. It’s a way to ask the right questions about your path.
Ask yourself:
- Am I more of a generalist, or do I want to specialise?
- Do I prefer working on the technical foundations or the insights that drive decisions?
- Do I want to influence strategy or focus on execution?
- Do I want to work in a small team with a wide scope or in a larger organisation with focused roles?
- Am I ready to work with AI and automation?
And perhaps most importantly:
- Where is the money?
Because where there’s investment, there’s room for depth.
High-impact analytics work usually happens where digital channels drive business, and media budgets are large enough to justify specialists.
Final thoughts
One person can’t do it all. Not well, and not sustainably.
If you only have one person, ensure they’re set up to succeed, not expected to do the impossible.
(Otherwise, they will burn out.)
If you’re growing a team, be intentional about the roles. And if you’re building a career, choose your lane or pick the overlap where you thrive.
Analytics is not just about tools or tags.
It’s about outcomes, strategy, and increasingly, AI-powered action.
Start where you are.
Invest where it matters.
And build a setup where good data leads to better decisions.