The first time a client told me my retouching looked “plastic,” I had no idea what she meant. I’d smoothed her skin, brightened her eyes, removed a few stray hairs. It looked clean to me. Professional, even. She sent back a side-by-side comparison of her face before and after, and I saw it immediately: I had obliterated every pore, every shadow, every tiny variation that made her face look like a face. She looked like a rendering of a person, not a person.

That was the edit that made me spend the next year actually learning what skin is.

What Skin Texture Really Is, and Why It Disappears

Skin has two distinct layers of information sitting on top of each other: texture and tone. Texture is the physical surface, the pores, fine lines, tiny bumps. Tone is the color and luminosity underneath, the blush in the cheeks, the slight darkness under the jaw, the warmth across the nose.

Most beginner retouchers attack both at once. They paint over uneven tone with a healing brush, and in doing so they flatten the texture on top of it. Or they blur the whole face to smooth texture, and they blow out the subtle tonal variation that makes skin look three-dimensional. Either way, the result is that polished, waxy look that clients hate and casting directors reject.

Frequency separation fixes this by splitting those two layers apart so you can work on each one independently. In Photoshop, the high-frequency layer holds texture. The low-frequency layer holds tone. Once they’re separated, you can smooth a red patch without touching the pores above it.

The Frequency Separation Setup That Actually Works for Headshots

I keep a Photoshop action for this (mine is called “The Godfather,” because of course it is) that runs in about three seconds, but here’s what it’s doing:

Duplicate your background layer twice. Name the bottom copy “Low Frequency” and the top copy “High Frequency.” On the Low Frequency layer, run a Gaussian Blur of 3-4 pixels for most headshots shot at 100 ISO, or bump it to 5-6 pixels for noisier files. On the High Frequency layer, go to Image > Apply Image. Set the layer to Low Frequency, blending mode to Subtract, scale to 2, offset to 128. Change the High Frequency layer’s blend mode to Linear Light.

Now you have your separation. For headshots, I do almost all my tonal work on the Low Frequency layer with a soft brush at 10-15% opacity, painting in a color sampled from nearby skin. I leave the High Frequency layer almost entirely alone except for cloning out individual hairs or blemishes that have a distinct textural edge.

A good headshot retouch for a corporate client takes me 20 to 30 minutes this way. Before I locked in this workflow, the same job took 45 minutes and looked worse.

Dodge and Burn: The Step Most People Skip

After frequency separation, the work that actually makes a headshot look expensive is targeted dodging and burning on a 50% gray layer set to Overlay. This is where you restore or enhance the three-dimensional quality of the face by slightly brightening high planes and slightly darkening recessed ones.

Create a new layer, fill it with 50% gray (Edit > Fill > 50% Gray), and set it to Overlay. It will disappear. Now paint on it with a soft black brush at 5-8% opacity to darken, or a soft white brush at the same opacity to brighten.

For headshots specifically, I focus on three areas: the slight shadow under the lower lip (which defines the chin), the bridge of the nose (which anchors the face), and the catchlights in the eyes. A tiny amount of brightening in the iris, just a few gentle strokes at 6% opacity, can make someone look alert without making them look retouched.

Eye and Hair Detail Without Going Overboard

Headshots live and die on the eyes. My rule is to use the Sharpen tool at 20% strength on the iris only, never on the whites, and never on skin. I’ll also occasionally run a very subtle Camera Raw filter on a stamped layer, pulling Texture up to about 15-20 on the hair only, masked off from the skin. This recovers the sense that hair has individual strands without needing to manually paint them.

The biggest mistake I see in headshots is over-whitening the sclera. The whites of the eye are not actually white. They’re a warm, slightly bluish-gray that reflects ambient light. If you pull them too bright or too cool, they look like contact lenses and the whole portrait feels off. I bring sclera brightness up a maximum of 10-15 points in the Luminance panel, and I do it with a tight mask.

Exporting Headshots So They Survive the Real World

A retouched headshot that looks beautiful at 100% on your calibrated monitor can look flat or noisy on a LinkedIn page or a theater bio. I export all headshots as JPEG at quality 90 in sRGB color space, sized to 2000 pixels on the long edge. That’s large enough for print submissions, small enough for web without needing a separate web version, and sRGB ensures the color doesn’t shift when someone opens it on an uncalibrated screen.

I also add a very slight output sharpening in Lightroom before export: Amount 40, Radius 0.8, for screen output. It’s subtle but it compensates for the softening that JPEG compression introduces, and it’s the difference between a headshot that looks sharp in an actor’s online portfolio and one that looks just slightly soft in a way nobody can quite articulate.

The thing I keep coming back to is this: great headshot retouching is not about making someone look different. It’s about removing the things that were never really there, the harsh catch-light on an oily forehead, the shadow from bad studio lighting, the blemish that appeared that morning, so the person in the photo looks like themselves on their best day.