How I work
Great research isn't about method.
It's about staying useful to the people making the decision.
A few principles shape how I approach research, whatever the method or industry. They run through all the work in this portfolio.
Research should serve a decision.
The most useful research is tied to a decision someone needs to make. When I built the intake process at Metacore, the first thing it asks is what decision the research will inform, and that one question changes the quality of everything that follows. Research anchored to a decision tends to get used; research that isn't tends to sit in a folder.
Connect the dots across different kinds of evidence.
Game data tells you what players do, a survey tells you what they think, and community feedback tells you how they feel. No single source gives you the whole picture. I'm at my best bringing these together: behavioural and attitudinal, quant and qual, and weaving them into something a team can act on. Often the most telling finding is where the sources agree or where they don't.
Pair quant rigour with qual depth.
I lean on numbers for scale and measurement and on conversations for meaning and motivation. A survey grounded in interviews tends to ask better questions. A model means more when you understand the people behind the data points. I've found the methods are usually stronger together than either is alone.
Be honest about the evidence, and about who did what.
I'd rather flag a limitation than oversell a finding. When a model only explained part of a retention gap, I'd say so; when a sample skewed one way, I'd name it. The same goes for credit: research is almost always a team effort with decisions usually sitting with stakeholders. So I'm clear about what was mine, what was shared, and where my work informed a call someone else made. Being honest about the edges, I think, is what makes the rest worth trusting.
The shape of a project
How a study usually runs.
Every study is a little different, but the shape is fairly consistent.
Align with stakeholders first.
I start by getting clear with the team on the decision at stake, what's already known, what we're trying to find out, and the higher-level questions behind it. This step matters to me more than any other. Research goes wrong most often when everyone assumes they're aligned and they're not, so getting it right up front is what makes the findings land later.
Scope and design.
From that alignment, I shape the questions and choose the most appropriate methods to fit the goal and the decisions we've set, planning for the segments and comparisons that will actually matter.
Run and analyse.
I collect the data and bring in collaborators where their expertise sharpens the work: a data scientist for advanced modelling, analysts for behavioural and commercial context, Community and Player Experience for an emotional pulse check. Then I dig into what it's really saying.
Synthesise and share.
I turn findings into a clear, decision-oriented readout (what we found, why it matters, and what to do next) and bring it to the people who'll act on it.
Keep it usable.
Insights go into a shared repository so they're findable later and the knowledge compounds, rather than evaporating once the deck is closed.