clientgallery.io

Proofing

How to Help Clients Narrow Down Photos for Their Album

Album paralysis is not indecision. It is one decision asked too many times. Split it into two passes, give each pass its own mark, and the shortlist arrives ready to lay out.

Updated 5 September 2026

The client opens the gallery, loves everything, and picks nothing. Three weeks later the album is still unstarted and you are writing the fourth polite reminder. This is the most common failure in album work, and it is almost never a client who does not care.

The cause is structural. You asked a person to look at 600 photographs of the most emotional day of their life and answer one impossibly loaded question about every single one of them: is this good enough for the book. Nobody holds that question for 600 frames, so they stop.

The fix is to split the question in two and give each half its own mark: a fast, low-stakes pass first, then a slow, ruthless pass over a much smaller set. Below is the method, the album craft that makes it produce a book instead of a pile, the exact mechanics, what the export file does not carry, what the other platforms do instead, and the case where this does not work.

Why one pass always stalls

A single-mark gallery asks the client to do two incompatible jobs at once. The first job is emotional recognition: which photographs mean something. The second job is editorial: which of those earns a page in a finite book.

The first job is fast. People know within a second whether a frame moves them. The second is slow, comparative, and needs the whole shape of the album held in your head. Forcing both onto one heart icon makes every click a small argument, and it produces the wrong output: one undifferentiated list of 180 photographs, which is not a shortlist. It is the same problem handed back to you with the client's name on it.

The two-pass method

Two passes, two marks, two different questions. Say both questions out loud when you send the link, because the client will not infer them.

Pass one, the star. Go through the whole gallery once and star anything you would be sad to lose. Do not think about the album, do not compare, do not go back. If you hesitate, star it and move on. This pass is deliberately generous, and its only job is to cut the field to something a human can deliberate over.

Pass two, the love. Now look only at the starred set and ask a different question: does this photograph earn a page. Love only those. This pass is slow, and it is supposed to be. It is the first time the client is comparing photographs against each other rather than against nothing.

On clientgallery.io the two marks live in the same grid, on the same photo, and the client never leaves the gallery to manage a list. A star is level one and a love is level two. A single tap stars, a double tap loves. The client-facing labels are literally "Star" and "Love", and the gallery's own instruction reads: open a photo, star what you like, love your favorites, then send your picks.

Picking is on by default and can be turned off per gallery. There is no client account to create, because picking is authenticated by the share link. Sending is the one place a client types anything: the dialog requires a name and a valid email, and the server rejects a submission without both.

Cull the duplicates before you send, not after

A second pass only works if the first produced distinct choices. If the gallery holds five nearly identical frames of the first look, the client stars all five and then has to rank the same moment against itself. That is not editorial judgement, it is a spot-the-difference puzzle, and it is exhausting in a way that feels like the client's fault when it is yours.

Cull near-identical frames in your catalog before the gallery goes out. One frame per moment, unless a second genuinely says something different. Every duplicate you leave in is a decision outsourced to someone less equipped to make it than you are, who will make it slower.

The one exception is the vertical and horizontal pair of a frame you know you want big, because an album designer needs both orientations for layout freedom. Keep the pair, and tell the client a pair counts as one choice.

Break the gallery into chapters

Nobody sorts 800 thumbnails in one sitting. They sort a chapter in one sitting, four times, and finish. Sub-galleries on clientgallery.io are folders exactly one level deep inside a gallery, so a wedding becomes preparation, ceremony, portraits and reception.

A photo belongs to at most one sub-gallery, and photos in none sit at the top level. Sub-galleries cannot nest, and a deeper folder drop is flattened on import, so plan one flat layer. Each sub-gallery has its own share link and can carry its own access code, which is how the parents get the ceremony chapter and not the whole day.

Chapters help you too. An album is chronological in almost every case, so a client who sorts by chapter hands back a shortlist already in rough sequence. Galleries hold up to 1,000 photos, a fair-use ceiling and not a target.

Give a number, and say what happens to the rest

"Pick your favourites" has no stopping condition, so the client never stops. Give an arithmetic target instead, derived from the book you actually sold. Count the spreads in their package and multiply by the number of photographs you place on a spread in your own design style. Say that product as a single number, not a range: "love about 70 photographs, and star everything else you want kept."

Then answer the question the client is actually afraid of, before they ask it. They are not agonising because they want a perfect book. They are agonising because they believe an unloved photograph is a deleted photograph. Tell them plainly that it is not: starred photographs stay in the gallery, stay downloadable, and can go into a second volume or a print box later. Hosted galleries never expire, so "you are not throwing anything away, you are only deciding what goes in the book" is literally true, and it removes most of the paralysis on its own.

Album craft: sequence beats quantity

Two passes produce a better book, not just a shorter list, because the second question is an album question. An album is spreads, not a set of favourites. A spread has to carry a moment, hold a pair of facing pages together, and sit next to the spread before it without repeating it.

So phrase pass two as "does this earn a page", never as "do I like this". Those two questions produce different lists. The second one produces eleven portraits of the bride and no photograph of the room.

This is why the star tier matters as much as the love tier. Coverage matters more than peaks in an album. The connective frames, the room before anyone walks in, the hands on the door, the wide of the tables, are usually nobody's favourite photograph and they are what makes a book read as a day rather than a highlight reel. Tell the client explicitly to keep those as stars even when they are not loves.

When the picks come back you have exactly that structure. The loves are your spread candidates. The stars are your connective tissue and your substitution pool when a spread needs a horizontal and the loved frame is vertical.

What comes back, and in what form

The point of two marks is that both tiers survive the trip back into your catalog. The Lightroom Classic plugin matches picks by filename against the open catalog, and the mappings are exact.

Client markedArrives in Lightroom Classic asWhat you do with it
Love5 stars plus a Pick flagYour spread candidates. Filter on the flag and start laying out.
Star3 stars, no flagConnective frames and substitutions. Kept, not queued.
Colour labelThe matching Lightroom colour labelWhatever meaning you agreed with the client up front.
Comment on a photoIPTC caption, opt in, off by defaultTurn it on when the client leaves layout notes.

Everything lands in a collection set named ClientGallery, holding a collection named after the gallery. The menu item is Import ClientGallery Picks. Nothing nests beyond that set, and matching is by filename, so your files need the names they were published under.

That is the whole point of the method: your album edit queue exists in the catalog before you open a single file. Five stars and a flag is a filter, not a checklist you transcribe. Email fires on picks as well, with the body stating how many were loved and how many were starred.

The .txt export flattens the two tiers

If you do not work in Lightroom Classic, read this carefully, because it is the one place the two-pass structure does not survive.

After the client sends their picks, the completed-gallery banner offers a button labelled "Export picks (.txt)". It is not a button sitting on every gallery, and it does not exist before the client has sent. The file is filenames only, one per line, with no header and no level column. Loves are listed first and stars follow, but the file does not label which is which. The ordering carries the tiers by convention, and nothing in the file carries them as data. Paste that list into Capture One or a spreadsheet and you get one flat list of picked filenames.

The counts do arrive, in the pick email, which states how many were loved and how many were starred. So the workaround outside Lightroom is to read the two counts from the email and split the file at that line. It works, it is manual, and you should know that before building a process on it.

What the other platforms give you instead

Two marks in the same grid is unusual, and the honest way to say so is narrow: none of Pixieset, CloudSpot, Pic-Time or ShootProof offers two levels of mark on the same photo in the same grid. Three of them offer multiple favourite lists instead.

PlatformSecond layer, as documented
PixiesetOne heart, plus multiple favourite lists. Clients click the heart icon on any photo to add it to a new or existing favorite list, and can create multiple lists to organize different selections.
CloudSpotOne heart, plus additional lists created with a Create New List button.
Pic-TimeOne heart and multiple lists, plus a photographer-initiated Selection Request with a count cap.
ShootProofPhotographer-defined labels for a single Linked Contact, entered behind a 6-digit access code, with a hide option.

Multiple lists are a real alternative, not a consolation prize. If your client needs separate baskets rather than ranks, one for the album, one for prints, one for social, a list per basket is exactly right and two levels would be the wrong shape. Lists ask the client to manage folders. Levels ask the client to rank a photo. Pick the one that matches the question you are asking.

Pic-Time's Selection Request solves album overshoot from the other end. The photographer sends a request with an Amount, and the help centre is explicit that the Amount is the maximum number of images the client can select, though they can submit fewer. It is included in the Professional and Advanced plans. Capping the count up front is a legitimate design. We do the opposite, and neither approach is obviously right.

ShootProof's labels are defined by the photographer and applied by one Linked Contact who enters an access code, and that contact can apply as many labels to an image as they want. It is a tagging system for a single privileged viewer rather than a two-tier rank for an ordinary client.

One more difference matters for album work. On Pixieset the Lightroom Copy List creates a list of all images in a Favorite List formatted for Lightroom's search function, which you paste into Library, Find, with Any Searchable Field and Contains, and the Pixieset plugin itself uploads photos to Pixieset directly from Lightroom Classic. A filename list finds your photos. It does not tell you which of them the client loved.

When this method does not work

This method assumes the client will do two passes. Some will not. They do one pass, love everything they like, star nothing, and send. Plan for that rather than resent it.

Be clear about what we do not do here. We enforce neither counts nor deadlines. There is no cap that stops a client at 70 loves and no timer that closes the gallery. If a hard cap is what your workflow needs, Pic-Time's Selection Request is built for exactly that and you should use it.

What to do when one pass comes back too long:

  • Do not send it back whole. Returning 180 loves with "please cut this to 70" recreates the original wall.
  • Re-cut by chapter. Give the target per chapter instead of per album. Twelve for preparation, twenty for the ceremony. Small numbers are decidable.
  • Design a first draft yourself. Lay out spreads from their loves, send the draft, and let them cut from a book instead of from a grid. Removing a spread from a sequence is a far easier decision than ranking a thumbnail.
  • Sell the overshoot. If they genuinely want 180, that is a larger book or a second volume, and it should be priced rather than argued about.

A client who sorts on one pass with no stars at all is fine too. A flat list of loves still arrives as 5 stars and Pick flags. You only lose the substitution pool.

Sources

clientgallery.io product behaviour was verified in the product source on 5 September 2026. All competitor pages were fetched on 5 September 2026.

Frequently asked

How do I ask a client to narrow down photos for an album without overwhelming them?

Split the question into two passes. First pass: star anything you would be sad to lose, quickly, without comparing. Second pass: look only at the starred set and love the frames that earn a page in the book. One pass asks an emotional question and an editorial question at the same time, which is why single-mark galleries stall.

How many photos should a client pick for their album?

Derive the number from the book you sold rather than from a rule of thumb. Count the spreads in their package and multiply by the number of photographs you typically place on a spread in your own design style. State that product as a single number, not a range, and tell the client what happens to everything they do not love.

What is the difference between a star and a love?

A star is level one and a love is level two, on the same photo in the same grid. A single tap stars and a double tap loves, and the client-facing labels are Star and Love. Back in Lightroom Classic a love arrives as 5 stars plus a Pick flag, and a star arrives as 3 stars with no flag.

Do both pick levels survive the export to a text file?

No. The Export picks (.txt) file is filenames only, one per line, with no header and no level column. Loves are listed first and stars follow, but the file does not label which is which, so the two tiers do not survive as data. The counts arrive in the pick email instead, which states how many were loved and how many were starred.

Should the client keep photos that are not album-worthy?

Yes, as stars. Coverage matters more than peaks in an album, and the ordinary connective frames, the empty room, the hands on the door, the wide of the tables, are what make a book read as a day rather than a highlight reel. Tell the client to star those even when they are not loves, so you have connective tissue and a substitution pool at layout time.

What if the client picks far too many photos anyway?

Expect it, and do not send the whole list back with a smaller number attached. Re-cut chapter by chapter with a small target per chapter, or design a first draft from their loves and let them cut spreads from a sequence rather than thumbnails from a grid. If they genuinely want all of them, that is a larger book and should be priced.

Can I force a maximum number of picks or a deadline?

Not on clientgallery.io. We do not enforce counts and we do not enforce deadlines, so the method depends on the client actually doing two passes. If a hard cap is what your workflow needs, Pic-Time's Selection Request sets a maximum number of images the client can select and is included in its Professional and Advanced plans.

Your client galleries, under your name

Unlimited galleries, your branding, one-click Pixieset import, and a Lightroom plugin, all on a flat 10 $/month with everything included. Your first gallery is free.

Create your studio

Related guides

How to Help Clients Narrow Down Photos for Their Album