Suppose you've got a 'wise woman' treating people with herbs. How does she decide which herbs to use? If she (or the generations before her who've passed on their knowledge to her) tried different ones, noticed what worked, and passed on that knowledge, then however different from modern medicine it may be, that's still science at work. Experiment, observation, refinement of a theory (e.g. the theory that this willow bark may help headaches, or whatever).
So that wise woman in her cottage, becoming skilled in knowledge of the herbs and rare plants that can heal by decades of experience treating people with them - she's a scientist too. She's also a scientist if she notices that looking at the whole person, listening to them, looking at more than just the ill bit, leads to them getting more better, more quickly.
Anyone looking at something and observing what works and what doesn't work, and acting on that basis, is a scientist. Not everyone who does that may be a good scientist - some may not realise when what they're observing is an experimental artefact, or understand what counts as statistical significance, or all sorts of things. But anyone just basically trying to base their behaviour on observations of what works and what doesn't work, and repeated trying out of new things to see what works best, is essentially just as much a scientist as the classic 'man in a white coat at a lab bench'.