Original research · August 2026
The most common graduation gifts, counted
Gift guides tell you what somebody thinks you should buy. We wanted to know what graduates are actually given, so we read 38 first-hand accounts and counted them. Here is the whole thing, including what it cannot tell you.
The count
The most common graduation gift is something to carry — a bag, a folio or a case — which appeared in 8 of 38 first-hand accounts we read, or 21%. That is more than double any other physical category. Money and jewellery tie for second on 4 accounts each, then pens, books and tools on 3 each.
- Something to carry21%8 of 38
- Money or a card11%4 of 38
- Jewellery11%4 of 38
- Pens8%3 of 38
- Books8%3 of 38
- Tools8%3 of 38
n = 38 first-hand accounts, read 22 August 2026
Bags, folios and cases beat every other physical category by more than double, and no gift guide leads with them. Money ties for second, and not one person described still using it. Everybody who named a bag, a pen or a set of tools described the thing years later.
Gift guides tell you what someone thinks you should buy. We counted instead. Seven category-neutral searches of public discussion threads, 49 first-hand accounts logged and 38 qualifying. No query named a product or a category, so the categories came out of what people wrote rather than out of a guess — which is the whole methodological point. A search for “graduation gift luggage” returns luggage and proves only that we guessed luggage first.
An account qualified if the writer described a gift they personally received, at a graduation, with enough detail to classify it. We excluded second-hand reports, wish lists, anything from a page selling the item, and anything we could not read at its original location.
Read those figures honestly: 38 accounts from one platform, gathered on 22 August 2026, and three of the seven searches asked about things people still use. That biases the count toward things that lasted, which is useful if you are buying but is not the same as what gets given most often. At this sample size a single extra account moves a small category by about three points, so treat the ordering as the finding and the exact percentages as approximate.
The full table
| Category | Accounts | Share |
|---|---|---|
| Something to carry | 8 of 38 | 21% |
| Money or a card | 4 of 38 | 11% |
| Jewellery | 4 of 38 | 11% |
| Pens | 3 of 38 | 8% |
| Books | 3 of 38 | 8% |
| Tools | 3 of 38 | 8% |
| Everything else, one or two accounts each | 13 of 38 | 34% |
That last row is worth reading carefully. At 13 of 38 it is bigger than any single named category, so the six above are not the whole picture — they are the two thirds of it that repeat. The remainder is fragmented rather than small: it is made up of one-off gifts, no two of which matched often enough to reach three accounts. A larger sample would very likely turn some of that tail into categories of its own.
The two findings that surprised us
Nobody leads with bags
Something to carry was the largest category by more than double, and it is close to absent from published gift guides, which tend to open with jewellery, technology or an experience. A folio, a briefcase, a piece of luggage or a decent everyday bag was the single most frequently described gift in the sample, and the accounts describing one almost always tied it to starting work.
Money is common and leaves no trace
Money and gift cards tied for second most given. Not one account described still having it, or what it was spent on. Every account naming a bag, a pen or a set of tools described the object years later. That is not an argument against giving money — it gets used, which is the point of it — but it is a real asymmetry, and it is the kind of thing a count surfaces and an opinion does not. We have written that up separately in how much money to give for a graduation gift.
What this cannot tell you
A count is only worth as much as its limits are visible, so here they are in full rather than in a footnote.
- The sample is 38. Enough to rank the categories, not enough to pin the percentages. Eight of 38 is 21 per cent with a 95 per cent confidence interval of roughly 10 to 37 per cent. Treat the ordering as the finding.
- One platform, one day. Everything was gathered on 22 August 2026 from public discussion threads. A different platform would very likely produce a different mix, and an older or younger population certainly would.
- It is biased toward things that lasted. Three of the seven searches asked about objects people still use. That is useful if you are buying for durability and it is not the same question as what gets given most often. We have not corrected for it, because correcting would mean inventing a weighting we cannot defend.
- We cannot segment it. High school against college, or gifts from parents against gifts from friends, would be the most useful cuts available and 38 accounts will not carry them. That needs a bigger corpus, which is the next piece of work.
- It is self-reported and self-selected. People who post about a gift years later are not a random sample of people who received one.
How the count was made
Seven category-neutral searches of public discussion threads about gifts people were actually given. Neutrality is the whole methodological point: a search for “graduation gift luggage” returns luggage and proves only that we guessed luggage first. None of the seven named a product or a category, so the categories that came out are the ones people raised unprompted.
That produced 49 first-hand accounts, of which 38 qualified. An account qualified if the writer described a gift they personally received, at a graduation, with enough detail to classify. We excluded second-hand reports, wish lists, anything from a page selling the item, and anything we could not read at its original location.
We do not quote or name anybody. Everything published is our own summary of what the accounts said. That is deliberate and the reasoning is on how we count gift research, along with the rest of the method. We keep the full internal record of every source read and the date it was read; if you want to check a specific figure, ask and we will answer.
Using this
You are welcome to cite these numbers. Please link back to this page so a reader can see the sample size and the caveats, because a percentage from this count without its denominator is a misrepresentation of it. If you spot an error, tell us and we will correct it in public.
Common questions
Frequently asked questions
What is the most common graduation gift?
Something to carry — a bag, a folio, a case — in 8 of the 38 first-hand accounts we read, or 21 per cent. That is more than double any other physical category. Money or a gift card and jewellery tie for second on 4 accounts each (11 per cent), followed by pens, books and tools on 3 each (8 per cent). The sample is small and drawn from one platform on one day, so treat the ordering as the finding and the percentages as approximate.
What is a typical graduation gift?
In our count the typical gift falls into one of six categories: something to carry, money or a gift card, jewellery, a pen, a book, or a set of tools. Together those six account for 25 of 38 accounts. What separates them is not how often they are given but how long they survive: every account naming a bag, a pen or a set of tools described still using the thing years later, while nobody who received money described still having it.
What graduation gifts last the longest?
Bags, pens and tools, by a wide margin. Every account in our corpus naming one of those three described still using it years after graduation, including a pen in daily use for more than two decades and a cookbook still in the kitchen four decades after it was given. Be aware of how we found that: three of our seven searches asked specifically about things people still use, so the corpus is biased toward gifts that lasted. It is useful if you are buying for durability and it is not evidence about what gets given most often.
How did you work out what graduates actually get?
We ran seven category-neutral searches of public discussion threads about gifts people were actually given, logged 49 first-hand accounts and qualified 38 of them. No query named a product or a category, so the categories came out of what people wrote rather than out of a guess. We summarise what those accounts said in our own words rather than quoting anyone. The full method, including the exclusions and the limits of a 38-account sample, is on our how we verify page.
Is 38 accounts enough to draw a conclusion from?
It is enough to rank the categories and not enough to pin the percentages. Eight of 38 is 21 per cent, but the 95 per cent confidence interval around that runs from roughly 10 to 37 per cent, and a single extra account moves a small category by about three points. What the count supports is that carry items lead by a wide margin. What it does not support is that they lead by exactly 21 per cent. We publish the count rather than a bigger claim because the count is what we did.