Start with the observed peaks
The first step is not naming the compound. It is building a reliable list of observed peaks with positions, relative intensities and uncertainty.
Noise threshold, background subtraction and peak width assumptions can change the peak list. That is why a phase identification result should show the settings and not only the final label.
Compare against candidate reference patterns
A candidate phase becomes plausible when multiple measured peaks line up with a reference pattern under the same wavelength assumption.
The opposite evidence matters too. If a reference pattern predicts strong peaks that are absent, the assignment should be treated carefully.
- Use several peaks, not a single dominant peak.
- Check whether peak shifts are systematic or random.
- Inspect unmatched observed peaks for impurities, substrate peaks or artifacts.
Use structure and refinement as additional context
CIF-derived structure information helps explain why peaks should exist and how lattice changes can shift positions.
Refinement-style fitting can support a phase assignment, but it should not be treated as magic. Residuals, Rwp, Rp, GOF and parameter plausibility all matter.
Preserve the decision
The best phase identification workflow leaves a review trail: what was matched, what was ignored, what remains uncertain and what should be measured next.
That review trail becomes especially valuable when a lab compares many samples across dopant ratios, annealing temperatures or synthesis batches.