We weighed 40 travel backpacks: real weights vs claimed

Pack 39 is a 55 L pack whose maker lists it at 2,100 g. On the bench scale in Manchester it read 2,290 g. That is 190 g over the claim, or 9.0 percent, and it was the heaviest pack of the 40 we weighed. Pack 08, a 28 L pack, went the other way and read 29 g under its claim. The rest sat between those two in grams, and the median pack was 45 g heavier than advertised.

Makers publish weights, shoppers sort by them, and reviews quote them. I wanted to know how far to trust the number, so in July I weighed 40 packs bought for a travel pack comparison and set each reading against the maker's own figure. This is a small data set, and I describe it in full below because the method matters more than any single result.

The 40 packs were bought at retail between late 2025 and July 2026, one unit per model, and nothing was supplied by a brand. I used the bench scale in Owen Whitfield's Manchester workshop over three days, Tuesday 7 to Thursday 9 July 2026. The scale reads to 1 g. Before each session I checked it against a 500 g and a 2 kg check mass, and it never drifted by more than 2 g. Each pack was weighed three times with the scale zeroed between readings, and I kept the middle reading. The three readings of any one pack never differed by more than 3 g.

The one wobble came on the Wednesday afternoon, when the 2 kg check mass read 2,002 g, the largest drift of the three days. I re-zeroed the scale and re-weighed the six packs from that morning to be sure. None of them moved by more than 1 g.

I took each maker's claim from its own product page on 6 July. Two listings had changed since we bought the packs, and I used the July figure. Where a page listed several sizes, I matched by claimed volume. I then sorted the packs into six classes by that claimed volume:

• Under 25 L: 5 packs

• 25 to 30 L: 7 packs

• 31 to 35 L: 8 packs

• 36 to 40 L: 9 packs

• 41 to 45 L: 7 packs

• Over 45 L: 4 packs

Every pack got two weights. As sold is the pack with shipping paper, tags and stuffing removed and everything that came in the box left attached: hipbelt, rain cover, sternum strap, detachable pouch. Stripped is the pack with every removable part taken off, including frame sheets and loose foam inserts. Fifteen packs had parts I could remove, weighing 40 to 260 g in total per pack. The other 25 had nothing to take off, so their two weights are the same.

For three models I weighed a second unit bought later, as a check on unit variation. The second units differed from the first by 6, 11 and 19 g. I treat any gap under 20 g as noise. The pack names in this piece are lab codes, numbered by volume. Owen Whitfield, who runs the workshop, reviewed the protocol and the check masses before I started.

Added up, the 40 makers claimed 53,380 g of backpack and the scale read 55,196 g, which is 1,816 g or 3.4 percent more. The median pack was 45 g over its claim, also 3.4 percent. Twenty-nine packs were heavier than claimed, 11 were lighter and none matched to the gram. The measured weights ran from 611 g to 2,290 g, and eight packs were under 1 kg.

The heavier packs went further than the lighter ones. Six packs were more than 100 g over their claim. The largest miss in grams was Pack 39 at +190 g, and the largest in percent was Pack 31, a 44 L pack, at +152 g or 9.2 percent. The lighter packs stayed close: the furthest under were Pack 08 and Pack 40, a 60 L pack, at 29 g each.

The table shows each class, using the median of the per-pack differences. That median is not the same as the difference between the two medians, which is why a row can look a little off if you subtract by hand.

In grams, the gap grows with the pack, from 10 g in the smallest class to 115 g in the largest. In percent it does not climb in a line: 1.3, 4.1, 2.7, 3.3, 2.9 and 5.9. The largest class has only four packs, and I would not read anything into its 5.9.

Volume does not explain it well. Removable parts do. The 15 packs with something to take off ran a median of 82 g over their claim, or 6.2 percent. The 25 packs with nothing to take off ran a median of 21 g over, or 1.6 percent, which is close to the 6 to 19 g by which two units of one model differed. Every one of the 13 packs more than 5 percent over its claim had removable parts.

Taking the parts off overshoots in the other direction. For 14 of the 15 packs, the stripped weight fell below the claim, by a median of 58 g, so the claim sat between the stripped and the as-sold weight. Pack 18, a 35 L pack, is a clear case. Its maker lists 1,300 g. It read 1,396 g as sold and 1,206 g with a hipbelt and rain cover removed, 190 g in all. Pack 39 shows the same thing at a larger size: 2,100 g claimed, 2,290 g as sold, 2,030 g stripped.

I cannot say what each maker weighs. My reading is that many weigh the pack with some parts left out and do not say which. That is a guess from the pattern, and the listings I read rarely state what the figure includes.

The packs with nothing to remove were close to their claims. The biggest miss among them was Pack 38, a 50 L pack, at 84 g over, or 4.2 percent. All 11 of the packs lighter than their claim were in this group, none by more than 29 g. Some of that is rounding: 35 of the 40 claims were multiples of 50 g, so a maker rounding to the nearest 50 can be off by up to 25 g either way before anyone touches a scale.

I also read the 15 product pages for the packs with removable parts, looking for any note on what the weight covers. Four of the 15 said. One listed the pack weight "without hipbelt", one gave a weight "with rain cover" and two gave two weights, with and without removable parts. The other 11 gave a single figure and no note.

Those four packs were also the closest to their claims of the 15. Their as-sold weights ran 34 to 53 g over, against a median of 111 g over for the other 11. Four listings is too few to prove anything, but it points the same way as the rest of the data: when a maker says what the number includes, the number is easier to trust.

This is one unit of each model, with a second unit for three of them, so I cannot separate a heavy batch from a heavy design. The 40 packs are the ones we bought for a travel test and skew toward popular travel packs, up to 60 L. They are not a random draw from every pack on sale, and ultralight and expedition packs are mostly absent. The class rows are small, from 4 to 9 packs each. Claims come from product pages on one day, and makers revise them.

I weighed packs dry at room temperature. A damp pack will read higher, and a pack that has carried a week of clothes will have picked up some dirt. None of that is in these numbers. The raw table of claimed, as-sold and stripped weights is available on request, and I am happy to share the 6 July page captures behind the claims.

The sentence I would quote from this data: the median travel backpack in our sample weighed 45 g, or 3.4 percent, more than its maker claims, and every pack more than 5 percent over had removable parts. Here is what I would do with that.

If two packs differ by less than 50 g on paper, ignore the weight. Unit variation and maker rounding together are about that size, and the difference will not survive a trip. Look at the harness, the opening and the shape instead, which our backpack guide covers.

If your airline weighs cabin bags, add about 1.5 percent to the listed weight of a pack with nothing to remove, and about 6 percent if it comes with a hipbelt, a rain cover or a frame sheet. For a pack listed at 1,500 g, that is roughly 1,525 g in one case and 1,590 g in the other. Weigh your own pack at home before you trust either figure, with and without the removable parts.

If weight is the reason you are buying, look for a listing that says what the figure includes, and prefer one that gives a weight with and without the removable parts. Of the packs I weighed, the ones with nothing to remove were the easiest to trust. For the packs that fit a budget airline sizer, see our carry-on guide for budget airlines.