The light you actually have: reading it before adding any

Work through these in order and the exposure decision stops being a guess.

Count the stops the lens has already taken

A zoom quoted f/3.5-6.3 gives its second figure at full extension. That is two and a half stops gone before the room is considered. On a prime printing f/1.8, nothing has been taken at all — the difference between the two is more than three stops, which is the difference between 1/30 s and 1/250 s in the same room.

Read what the meter proposed, then decide

The meter averages. Decide whether the subject should be brighter or darker than average, and apply compensation deliberately rather than changing metering mode and hoping the aim was right.

Choose the shutter speed from the subject, not the lens

The old guide of one over the focal length describes the lens and your hands. It says nothing about a moving subject, which is answered by shutter speed alone. Stabilisation compensates for camera movement and not for a walking child — and no title in this catalogue names a measurement protocol for its stop count.

When adding light, match the figures already in the room

Eight titles here publish a colour temperature — 5600 K on studio units, 6500 K on a small flash, one bi-colour panel stating 3200 K to 5600 K. Mixed temperatures put two casts in one frame and no single white balance corrects both. Four titles publish a rendering index, which is what predicts whether reds survive the light.

Respect the sync speed if a flash is involved

It is published per body and belongs to the shutter. Above it, a single burst lights only the strip the travelling slit has reached. One product here publishes a high-speed ceiling of 1/8000 s; that mode spreads the burst and costs output. Otherwise, cut ambient light with a neutral density filter rather than shortening the exposure.

Photograph a neutral reference in the first frame

One frame of a grey card or a colour chart under each lighting setup, and every later frame in that setup has the same starting point. It costs seconds and it removes the largest variable in colour correction.