How to Find Out What Something Is Worth

Turn it over first. Marks live on the base, inside the lid, under the seat, on the back of a movement, along the inside edge of a lining, and on the copyright page of a book.

Identification comes before valuation and cannot be skipped. A number attached to the wrong object is worse than no number at all.

Identifying it precisely: marks, numbers, labels, stamps

Hallmarks on silver and gold carry a maker, a standard, an assay office and often a date letter. Ceramics carry a backstamp that changed over the years the factory operated. Machines carry a model and serial plate. Furniture carries a stencil, a label under a drawer or a stamp on the underside of a rail. Clothing carries a union label or a care tag whose format dates it. Books carry an edition statement and a number line.

Photograph the mark straight on, in daylight, filling the frame. Move the camera closer rather than cropping a distant shot, and turn the flash off - raking light from a window shows an impressed mark that direct flash flattens.

Then search the mark rather than the object. Put the exact string of numbers or letters in quotation marks. Reverse image search the photograph of the mark on its own. Write down every character even where it looks like noise, because a single letter is often the date.

Reading completed sales, not hopeful asking prices

An asking price is one person's hope. A completed sale is what somebody actually paid on a particular day for a particular condition.

On auction and marketplace sites, filter to sold or completed listings - the filter exists and it is the single most useful control on the page. Read the ones that sold. Look at how many bidders there were, because one bid means somebody set a low start and got lucky, and a long bidding war means real competition.

Discard the extremes at both ends and read the middle. Then read the condition of the items in that middle: a chip, a repair, a missing lid or a replaced part makes a different object from yours.

If nothing comes back as sold, that is information rather than an absence of it. Either almost nobody sells one, or nobody looks for one, and those two look identical from your kitchen table. Turning any of this into an actual number is the next job: turning research into an asking price.

Free valuation routes and the bias in each

A dealer's free valuation is an offer. The number reflects what they can resell it above, plus their risk of being wrong, and there is nothing dishonest about that as long as you know which number you are being given.

An auction house valuation day wants consignments. Estimates there are set to attract bidders and to be beaten, so ask what the seller's commission is, what the lotting and photography fees are, whether there is a minimum, and what happens if it does not sell.

Online valuation forms are lead capture. Pawnbrokers value the loan rather than the object. Insurance valuations answer what replacement costs, which is a different question from what a buyer pays.

The test that cuts through all of it: ask each person what happens next if you say yes. Whoever ends up owning the object has told you what their number was for.

Asking a collector community so they answer

Specialist forums, subject subreddits with an identification thread, collector groups and, for genuinely odd things, a museum curator or a specialist society will answer questions from strangers. Museums identify and will not value; that is a policy, not a brush-off.

Read the pinned rules before posting, because some groups take identification questions and refuse valuation ones outright, and posting the wrong one gets you removed rather than answered.

Then ask properly. One object per post. Photographs of the whole thing, the base, and every mark. Dimensions and weight. What you know about where it came from and how long it has been in the family. A direct question at the end. Vague photos get vague answers.

Do not post your address, your town or the fact that you live alone with something valuable. And be careful with anyone who replies asking you to message privately so they can take it off your hands: people who offer to appraise it and then buy it.

What the sell-it-cheap-and-find-out approach costs

You only get to find out once. A cheap listing produces a fast first message, and the fastest messages come from people who buy to resell - that is the entire craft of buying things to resell at a profit, and they are better at this research than you are.

The honest other side: research is not free either. An object sitting in a cupboard while you keep meaning to look into it is not paying anything, and the looking-into-it can absorb a lot of evenings with no answer at the end.

So bound it. Decide before you start how many evenings you are giving this, and decide now what you will do when they run out - list it, take the dealer's offer, or put it back in the cupboard on purpose rather than by default.

If you list it and nothing happens, the silence itself is data about the title, the photographs or the price: what to change when something did not sell.

Objects where nobody can tell you

Unattributed paintings, unmarked jewelry without a stone report, one-off handmade pieces, tools and furniture made by someone locally, and anything whose value rests entirely on a story that has no paperwork behind it. These have no comparable sales because there is nothing comparable.

Authentication and grading services exist for some categories, and they charge a fee for an opinion rather than for a good answer. Ask what the fee covers, how long the queue is, whether you get a written report if the answer is negative, and what happens to the item while they hold it.

There is one more case worth naming. Something that matters to you more than it would to any buyer has no market answer, and selling it to discover the number is the one experiment you cannot run twice.

Photograph every mark before anything leaves the house. If the question comes back later, those photographs are all you will have.