Why a matrix beats a list of studies
The hardest step in a review is moving from summarising each study to saying something about the evidence as a whole. A list of study summaries keeps you trapped in study-by-study description; a synthesis matrix rotates the problem so that themes, not studies, become the organising spine. Once the recurring concepts sit in columns, reading down a column shows you where studies converge, where they disagree, and where the literature is simply silent.
How to build your matrix
Put one study per row, named by author and year, and one theme per column. Work through the studies one at a time, writing a short note in each cell for what that study contributes to that theme. Keep the notes brief and comparable, add a page number where it will save you a hunt later, and leave a cell empty when a study does not touch a theme. Those blanks are not failures; a column that is mostly empty is telling you about a gap in the evidence worth naming in your discussion.
From the grid to a written synthesis
When the grid is full, write theme by theme rather than study by study. For each column, state what the body of evidence agrees on, explain where and why studies diverge, and cite the studies that back each point. That is the difference between a synthesis and an annotated bibliography: the narrative is organised around ideas, with individual studies cited in support. The same studies you compare here are the ones that survived your screening and eligibility decisions, and a defensible review documents how they got there.
Where the matrix fits in the review
The synthesis matrix lives between data extraction and writing, drawing on a well-documented literature search and feeding the narrative you ultimately report. When you assemble the figures for that report, the PRISMA flow diagram tool documents how the studies in your matrix were identified and selected in the first place.