Making Data Make Sense
How I Work
Good evaluation work starts with strong design, well-structured systems, rigorous methods, and plans built to answer the right questions from the outset. Callum has a wealth of experience working closely with clients and stakeholders to get things right for their specific context.
But we all know things don't always work that easily in reality. The principles below are about what happens when solid design meets the real world, messy data, shifting field conditions, inherited systems, and how that gets handled without losing rigour.
People-First Approach
Good evaluation starts with genuinely understanding what a client needs, not just what they've asked for, and who else is affected by the work. That means listening closely to stakeholders throughout a project, then communicating results in a way that respects their time, context, and ability to act on what's found.
Prevention beats recovery
Catching a data problem before it reaches a report is more effective than correcting it afterwards, in cost, in credibility, and in the decisions that get made on bad information. I bring tried and tested training and experience to every project, built from years of designing tools that prevent errors at the point of entry, validating and cleaning complex datasets, and communicating openly about issues so they don't reoccur.
Utility focus
Accuracy matters, but data only creates value once someone acts on it. Part of the job is knowing which findings matter more, and having the judgement to say so clearly. I focus on identifying the insights that actually matter for a client's decisions, cutting through the noise and complexity.
Rigour that survives contact with the field
Survey plans are built on assumptions about the world that don’t always hold by the time you get to endline. Populations move, partners drop out, contexts shift. I design survey and evaluation methods that can flex without losing the ability to conduct valid analysis.
The right tool for the task
Callum has built up a broad toolkit over years, Stata, ODK, R, Power BI, Excel, drawn on flexibly depending on what a project and client actually need. Callum has also been developing AI-assisted approaches to use alongside his own expertise, helping ensure work is delivered efficiently without compromising on quality. The focus remains on getting the most from his skills, judgement, and rigour behind every output.
Data has to survive the trip to a non-technical audience
A rigorous analysis that no one outside the M&E team understands doesn’t change decisions. I build every deliverable, report, dashboard, or presentation, around the audience that has to act on it.
None of this replaces good methodology. It’s what makes good methodology hold up in the real world.
“Callum helped me build skills that genuinely changed how I work with data, and he has a real talent for making complex methods feel accessible without dumbing them down. Beyond the professional side, he’s also a genuinely good person to know and work with.”