Culling for a cohesive portfolio: fewer, and louder

Selecting a client gallery and selecting a portfolio are two different jobs, and doing the second with the habits of the first is why so many photography sites look like a talented person with no point of view. This is how to select for one voice: what to compare, what to leave out, and how often to do it again.

A portfolio is not a smaller gallery

A client gallery is an act of completeness. You owe that couple the whole day, so the delivered set runs into the hundreds. Typical wedding coverage produces 2,000 to 4,000 RAW frames and 400 to 800 delivered images, and nobody looking at that gallery is forming an opinion about your work. They already booked you. They are looking for their aunt.

A portfolio is an act of argument. Its entire job is to convince a stranger, in about the time it takes to scroll one screen, that you are the right photographer for one specific kind of work. It has a single viewer, a single purpose, and no obligation to be complete. That changes the criteria from top to bottom, and it is why photographers who are fast and decisive on a gallery freeze completely on their own website.

The other difference is the weight each frame carries. A soft image inside a gallery of six hundred is invisible. The same image inside a portfolio of twenty five is four percent of everything a stranger knows about you.

The question that gets you stuck

When you look at one image and ask "is this good?", you are making an absolute judgment, and people are poor at those. We are far better at comparison. It is the same reason a print looks fine on its own and then looks ordinary the moment you set a better one beside it.

Portfolio selection asks a different question, and it is a relative one: does this image belong with the others? A photograph can be the best thing you shot all year and still be wrong for the set, because it argues for a kind of work you are trying to stop taking on.

Volume makes all of this worse. Research on choice, including Iyengar and Lepper's much cited jam study, found that more options can lower satisfaction with the choice you eventually make, and Barry Schwartz's work on maximizers describes the regret that tends to follow. Sitting down with four thousand frames and an empty website is a recipe for choosing nothing at all. Start from a pool that is already small, and start from work you have already picked once.

What cohesive actually means

Cohesive does not mean identical, and it does not mean one preset applied to everything. It means a viewer can tell, without being told, that one person made all of these pictures. Five axes carry most of the effect.

You do not have to score well on all five. You do have to decide which two or three you are holding steady, because that decision is precisely what a stranger reads as style. If you want to see how differently this can be handled and still work, we collected working sites in photography portfolio examples.

A three pass method

Pass one: build a long list, fast

Go through your picked sets from the last two years and flag anything that could plausibly be portfolio work. Do not deliberate, do not compare, do not open anything at full size. You are collecting candidates. The goal is a pool of perhaps sixty or eighty images, and the whole pass should take an hour.

Pass two: compare in pairs

Now stop looking at images one at a time. Put two side by side and ask which one you would keep if you could only have one. This is where repeated ideas get caught, and they are not the near identical frames you already sorted out during gallery selection. They are two completely different photographs that make the same point. Two golden hour portraits in a field are one idea wearing two outfits. Keep the better one.

Pass three: sequence it

Lay the survivors out in the order a visitor will actually see them and read the set left to right. Now you are judging rhythm rather than pictures: two wides in a row, three faces in a row, one color that jumps out of the sequence. Move things around until it reads evenly, then cut whatever you kept having to move. An image that never sits comfortably anywhere is telling you something true.

The comparison habit is the underlying skill here, and it is worth practicing deliberately. We cover it in training your eye, which applies to every kind of selecting you do.

The one per shoot rule

Give each wedding or session at most one or two places in the portfolio. It sounds arbitrary, and it quietly does two useful things.

First, it forces variety of client, venue, season and light without you having to plan for it. That variety is exactly what a prospective couple is unconsciously checking: can this person do this anywhere, or did they get lucky at one very beautiful venue in June.

Second, it protects you from the sentimental run. Every photographer has one shoot where everything worked, and left to our own devices we will put nine frames from it on the front page. To a stranger, nine images of the same two faces in the same location does not read as a great day. It reads as a thin body of work.

The rule also makes the decision easier, because you are choosing within a single shoot, which is comparison, rather than across your whole career, which is not.

Good photographs that do not belong

The hard part of this job is never spotting the weak images. It is setting aside strong ones. Four kinds come up again and again.

• The crowd pleaser. The image everyone compliments, which happens to come from the kind of shoot you are trying to stop booking. A portfolio is a brief. Whatever is in it, you will be asked for again.
• The technical flex. The impossible off camera lighting setup, or the twelve frame composite. It proves you have skill and says nothing at all about what it feels like to be photographed by you.
• The favor. The frame you kept because the client loved it, or because a venue asked nicely. Neither is a reason for it to be doing marketing work on your site.
• The one from your old style. Two years of drift is enough to make an image look like somebody else's work sitting in the middle of yours.

How many images, and how often to revisit

Fewer than you want. A portfolio page gets read in one scroll, and attention runs out long before your images do. Many photographers find that halving a fifty image page makes the work look stronger rather than thinner, because the average goes up and no image has to survive being next to a weaker neighbor.

A practical shape: one tight main portfolio, then a separate page for each kind of work you take on, so a wedding couple never lands in the middle of your corporate headshots. Mixing service lines into one grid is one of the most reliable ways to make a good body of work read as unfocused, and it turns up in our list of photography website mistakes for exactly that reason.

Then review it on a schedule rather than on a feeling. Once a quarter, sit down with the current set and the best of last season and run a straight swap: one in, one out. The schedule stops the page drifting two years behind your actual work. The swap rule stops it quietly growing back to fifty images by Christmas.

When you publish the new ones, spend the extra minute per image on a readable file name, a plain alt sentence and a caption. On a twenty image page that is a twenty minute job, and it is the only part of image SEO that reliably does anything, which we work through in photo metadata and SEO.

Where this sits next to your other selecting

Most photographers are doing three different selection jobs and only have vocabulary for one of them.

1. Gallery selection. Volume, completeness, technical standards. Hundreds of images, decided in hours, and the question of how many is covered in how many photos to deliver.
2. Album selection. A narrative for one specific client, sequenced across spreads. A different craft again, which we separate out in album versus gallery selection.
3. Portfolio selection. Marketing. One voice, a stranger, no obligation to be complete.

Run them as three jobs and each one gets easier. Run them at once, at midnight, over the same four thousand frames, and you will do all three badly. It helps enormously to know what you are looking for before you open anything, which is what writing your own selection criteria is for.

The first job is the one that eats the calendar. A UK survey of more than 300 wedding photographers, reported by PetaPixel, put culling at 11 percent of working time against 4 percent spent actually shooting. That survey ran in 2020, before AI selection tools were common, but the shape of the problem has not changed much.

That volume pass is what Kepla is being built to take off your evening. It picks a shortlist from the shoot, on your iPhone or iPad first and on your Mac when a frame needs a closer look, and it never deletes, moves or renames a single file, so everything it did not pick is still sitting there a year later when you go hunting for portfolio candidates. The picking app is still in development. The booking page is live and free today.

FAQ