The box got sorted. The names got written down. Somebody went through all of it with an aunt and a voice recorder. That’s further than most people get, and it’s where the second stall happens.
Because a sorted, labelled, scanned collection of four hundred photographs is an archive. An archive is a thing you search. A book is a thing you read, and nobody has explained how one becomes the other.
This stage of a family project is a gift and a trap for the people in it. The sorting feels like progress, and it is, but I’ve seen a finished archive become the reason the book never gets made. Choosing forty photographs means deciding what somebody’s life was about, and nobody warns the family that this is the hard part.
How many photos should be in a family photo book?
A photograph book that works uses somewhere between twenty and forty images.
That number surprises people who have just spent two months scanning. It’s not a budget constraint. It’s attention. A reader gives a photograph three to eight seconds before turning the page, and by image sixty they’ve stopped looking and started flipping. Four hundred images at three seconds each is twenty minutes of flipping. Forty images with something worth reading beside them is an afternoon.
The other three hundred and sixty aren’t wasted. They belong in the archive, a separate deliverable with a separate purpose, and I’ve written about when the archive is the right thing to produce.
How do you choose which old photos to use?
Four tests, and most images fail all four. It shows a person doing something. A posed studio portrait of a great-grandmother tells you she owned one good dress. A photograph of the same woman at a kitchen table with her sleeves pushed up tells you how she stood. It contains a detail the text can use. The car at the kerb. What’s on the wall behind them. A shop sign across the street. Those are the images that let you write a paragraph about a decade. Somebody in the family has a story attached to it. If the labelling session produced five minutes of talk about a photograph, that image is in. The talk is the reason. It shows something that no longer exists. A house, a street, a business, a shoreline. Those photographs do work no other source can.
Anything that fails all four goes to the archive.
Cutting will feel like throwing the project away, and I think that feeling kills more of these books than anything else. You’re doing the opposite. Every weak photograph you keep dilutes the ones that tell the story, and the reader pays for your reluctance with their attention.
What does a photo book need besides pictures?
A spine, and the same rule applies here as to any other book. Chronological order is a filing system, and a family photograph book arranged by date reads as a sequence of unrelated moments with an unexplained gap in the 1950s.
Better organizing principles, in rough order of how well they hold up:
One place, across time. A house, a street, a business, a town. Photographs of one location across seventy years make an argument about change without anybody stating it.
One person, from outside. Every surviving photograph of your grandfather, arranged so the reader watches somebody else’s life assemble. This is the version that produces the strongest emotional response and it requires you to be candid about the years you have nothing from.
One question. Why did this family end up here. What did they do all day. What happened in the years nobody talks about. The photographs become evidence.
Work. What people did for money is the most underused frame in family history and the one that tells you the most.
Are captions enough, or do you need to write more?
Captions aren’t text, and I’m sick of seeing a beautiful collection wasted on a book of names and dates. It’s the most common failure in this kind of project, and the family is the one that loses, because the stories that should have gone beside those photographs disappear with the people who could have told them.
A caption identifies. Names, date, place. Necessary, and it does nothing else.
A book needs prose that says what the photograph cannot. Who took it and why. What happened fifteen minutes before. What the person in the frame didn’t know yet. Why the building behind them was new that year.
Test it this way. If your text could be deleted and a reader would lose only the names, you’ve written captions. If deleting it removes the reason the photograph matters, you’ve written a book. The ratio that works is roughly two to four hundred words per image, not twenty. Forty images at three hundred words is twelve thousand words of prose. That’s a real book with real writing in it.
Where does the text for a family photo book come from?
Three sources, and you probably have two of them already.
The recordings from the labelling sessions. If you went through photographs with a relative and kept the recorder running, that audio is your text. It contains the things only that person knew, in their own phrasing. I’ve written about what to do with interview recordings and the same indexing method applies.
The documents. Census entries give you household composition for the year a photograph was taken. City directories give you occupations and addresses. Newspapers give you what was happening on that street. The research piece covers where those live in this region, and the fourth floor in downtown Tampa holds most of it free.
Your own looking. Spend twenty minutes with a photograph and write down everything in it, including the things you’d never mention. The number of chairs. Whether the grass is cut. What season it is. That exercise produces the detail that makes a paragraph feel observed instead of summarized.
How do you date an unlabeled photograph?
The clues are almost never the people. Print format first. Sizes, borders, scalloped edges, square instant prints, the shift from black and white to color in ordinary family snapshots, and the warm cast older color prints acquire. Turn it over. Many prints carry a processing date stamped on the back or the border and most people never look.
Cars are the most reliable object in any street scene, because a model dates the image to no earlier than its production year and people usually drive something a few years old.
Then appliances, televisions, telephones and whatever sits on the kitchen counter. Then clothing and hairstyles, vaguer than people assume but useful together.
And children’s ages are the best internal clock in any collection. Date one photograph confidently and you can date every other image containing the same child.
What resolution should you scan old photos at?
The requirement is different and getting it wrong means rescanning. A photograph reproduced full page in a printed book needs considerably more resolution than anything that looks fine on a laptop. If a print will be enlarged, higher again. Over-scanning wastes disk space. Under-scanning wastes the photograph.
Scan flat, with the print clean, and leave the correction for later. A scan that’s been sharpened and color-adjusted at capture can’t be undone.
Negatives and slides hold more detail than the prints made from them and need a transparency unit or a service. For the twenty to forty images that will make the book, that’s worth doing properly.
Name the files with something searchable as you go. A folder of scan001 through scan400 recreates the original problem on a hard drive.
Who owns the rights to family photographs in a book?
Settle it before anything gets printed. A photographer who held copyright made studio portraits, and for older material that’s usually expired, but not always. A professional photograph from the 1970s may not be yours to publish. Snapshots taken by family members are simpler and still belong to whoever took them or their estate.
The harder conversation is about people. Living relatives appear in these photographs, and some of them will have strong opinions about which images go in a book the rest of the family will read.
I’d much sooner have that argument at the kitchen table before anything is printed. Finding out afterward means a cousin who won’t speak to you and a print run nobody wants to hand out.
For a private family edition none of this is usually a problem. For anything sold or made public, it needs a proper look.
What the finished thing looks like
Concretely, because vagueness is what stops people at this stage.
Roughly a hundred to a hundred and forty pages. Thirty to forty photographs, most of them full page or near it. Twelve to twenty thousand words of prose. Chapters organized by the spine you chose instead of by decade. An appendix listing every image in the wider archive with names and dates, so the book points at the collection instead of pretending to be it.
Printed properly, because photographs on cheap paper look like a photocopy of themselves. This is the one place in self-publishing where the printing choice changes the object. And enough copies for everybody who will want one. That’s always more people than the person making it expects.
When should you hire somebody to make a photo book?
This project has a failure pattern and you should know whether you’re in it.
The pattern: the sorting and scanning gets done, sometimes beautifully, and then the writing doesn’t start, because choosing forty photographs out of four hundred means deciding what matters about people you loved. That’s a hard thing to do alone.
Somebody outside the family can make those cuts without grief attached, then bring them back for approval. That single change unsticks more of these projects than any amount of encouragement.
The other trigger is a deadline with a person attached. An anniversary, a birthday, somebody’s declining health. A project that takes you four years and a professional six months isn’t a comparison of cost. I work with people across Pinellas County, so I can sit at the table while the photographs are on it. That turns out to matter. Half of what gets said about a photograph happens while somebody is holding it.
How should you store old family photographs?
The book uses forty. The other three hundred and sixty need a home and a plan.
Scanned files first, named searchably, stored in two places with one outside the house. A hard drive in a Florida closet is on a clock and so is the closet. The prints themselves go back inside the air-conditioned part of the house. Not the garage, not the attic, not a closet against an exterior wall. Heat and humidity destroy photographs here faster than anywhere I’ve lived, and mould spreads through a box.
Then the question of whether any of it belongs somewhere institutional. If the subject lived in this county and the photographs show streets, businesses or buildings, a local archive may want copies. Several collect exactly this, and the museum piece covers who.
And tell somebody in the family where all of it is. A perfectly organized archive that one person knows about is one house move from being a box again.
What a photograph gives a writer that a document cannot
I’ll state it, because it changes what you look for.
Documents tell you what people did. Photographs tell you what they handled, wore, stood next to and lived in. A census entry gives you a household of six at an address. A photograph of that house tells you the porch was too small for six people to sit on. The useful detail is almost always in the background: the furniture, the wall, the street, the car, the state of the paint.
That’s the same discipline that makes commercial archives valuable to researchers, and I’ve written about it in the piece on twenty thousand photographs of this region, where a studio photographing a storefront also captured the pavement and whoever was walking past.
Your box works the same way at family scale. The picture of a birthday is also a record of a kitchen that no longer exists, and the kitchen makes the paragraph land.
Should a photo book be printed or digital?
Printed, if it’s going to be read, and this is the one place in self-publishing where the choice changes the object. Cheap paper ruins photographs, and there’s no excuse for it. Ink bleeds into uncoated stock and the mid-tones disappear, so a face that was clear in the scan becomes gray mush on the page. The relatives who waited years for the book are the ones who pay for that shortcut.
A digital edition serves a different purpose and it’s worth producing alongside. Relatives across four states can read it immediately and nobody pays shipping. The practical answer is both. A printed run for the people who will keep it on a shelf, and a PDF for everybody else. Photograph books also survive worse as files than as objects, because a format goes obsolete and a book on a shelf doesn’t.
Order a single proof copy before the full run. Colour shifts on press and a proof is the only way to find out what your photographs look like printed instead of backlit.
How do you start a book from family photographs?
If four hundred is paralyzing, choose twelve.
Twelve photographs, three hundred words each, one organizing question. That’s a thirty-page object you can print, hand to somebody, and watch them read.
It’s also a test. If twelve works, forty works. If twelve doesn’t, you’ve learned that at a cost of a weekend instead of two years.
Most of the family history books I’ve seen finish started smaller than the person intended, and the ones that never finished tried to be everything. I’d much prefer twelve good photographs with real writing beside them in a relative’s hands this year over four hundred in a perfect archive nobody reads. The people in that box deserve to be read about while somebody who knew them is still around to hear it.
