How to Design a Packaging Insert That Gets Reorders
I ordered two candles from a small shop last month. The box was lovely — kraft paper, wood-wool packing, a wax seal. Inside, on top of everything, sat a folded card that said: thank you for your order, then the founder's story, then a request to follow on Instagram, then a request to leave a review, then a 10% code, then the care instructions, then a QR that went to the homepage.
Seven things. I read none of them. The card went in the recycling with the wood-wool.
That's the packaging insert problem in one paragraph. The card is the only piece of print you get to put directly in a customer's hands at their single most positive moment — the box is open, the thing is real, they already like you — and most sellers use that moment to say everything at once, which is the same as saying nothing.
A packaging insert has exactly one job
Pick it before you design anything.
There are really only three jobs a card in a box can do, and they compete:
- Get a review. Best for new shops with under ~50 reviews, where social proof is the actual bottleneck.
- Get a reorder. Best for consumables — candles, coffee, skincare, dog treats — where the customer will genuinely need more in six weeks.
- Get a follow. The weakest of the three, and the one people default to. A follow is a maybe. A reorder is money.
You can only ask for one. Not because of some law of design, but because the card is read for about four seconds, standing up, holding a candle in the other hand. Two asks means the reader picks neither — they put the card down "for later," and later is the recycling.
My rule: if the product runs out, ask for the reorder. If it doesn't run out, ask for the review. Everything else on the card is there to support that one ask or it comes off.
The box already said some of it
The second thing that bloats inserts is repetition. Before you write a word, list what the customer has already seen by the time they read the card:
- your brand name (on the shipping label, the tape, the tissue)
- what they bought (it's in their hand)
- your product's ingredients or care info (that belongs on the product label, not on a loose card they'll throw away)
- the founder's story (they read it on the listing page — that's why they bought)
Cut all of it. What's left is the thing they haven't seen: the reason to come back, and how.
A worked example: the candle card, rebuilt
Here's the card I'd send that shop, at 3.5 × 5 inches, printed both sides on 350gsm.
Front — the ask, and nothing else.
Your next one is 20% off. Code: BURN20 · expires in 60 days ash-and-ember.com
That's it. Four short elements on a mostly empty card. The headline is the biggest thing on the front by a wide margin — around 34pt, roughly 3× the body copy. The code is the second biggest, in a bordered box or a solid panel so the eye lands on it even in a bad-lighting unboxing photo.
Why 60 days and not "no expiry"? Because a candle burns down in about six weeks, and a deadline that lands right when the product runs out is the whole mechanism. A code with no expiry is a code with no urgency, and it never gets used.
Back — the support.
Make it last 40+ hours Burn 2–3 hours the first time, until the pool reaches the edge. Trim the wick to 5mm before every burn.
[QR] → How we make them (60-second video) @ashandember
The back is where the care tips and the social handle live, because they're genuinely useful but they're not the ask. Anything on the back is a bonus if it's read and costs nothing if it isn't.
Notice what's gone: the founder story, the "thank you for supporting a small business" paragraph, the review request. The review request is gone because this shop already has 400 reviews. If it had four, I'd flip the card — review on the front, discount gone entirely.
Two details that quietly decide whether it works
Use a code that only exists on the card. Not your newsletter code, not the sitewide one. BURN20 should be unique to the insert, because it's the only way you'll ever know whether the card is worth printing. Run 500 cards, watch the redemptions for 90 days, and you have a real number instead of a vibe. Most sellers never do this and then argue about card stock.
Size the QR for arm's length, not for a poster. An insert is read at about 40cm, which means a 2cm code scans fine — much smaller than what a flyer needs. What matters more is quiet space and contrast: leave a clear margin around it and never put it on a photo or a dark tint. I went into the scan-distance maths in the QR-on-a-flyer post, and the same rules apply, just at a shorter range. And point it at a specific page — a video, a reorder page — never your homepage. A QR that dumps someone on a homepage is a dead QR.
Building one in Ridvay Studio
Open Studio and type the card, both sides, in one prompt:
A 3.5 × 5 inch packaging insert card for a small candle brand called Ash & Ember. Front: a large headline "Your next one is 20% off", a code block reading BURN20 with "expires in 60 days" underneath, and the website ash-and-ember.com small at the bottom. Back: a heading "Make it last 40+ hours", two short care lines, a QR code, and the handle @ashandember. Warm cream background, deep charcoal text, one amber accent. Lots of empty space.
What comes back isn't a flat image you have to re-prompt until it's right. It's an editable design — the headline, the code, the QR, the background panel are all separate layers you click and change. That matters here more than on a social post, because a print card lives or dies on a handful of millimetre decisions.
Three edits I'd make immediately on the result:
- Click the headline and push the size up until it's clearly three times the body copy. AI layouts tend to land conservatively; the one-job rule only reads if the ask dominates.
- Select the code and put a solid panel behind it — amber block, cream text — so it survives a badly lit unboxing photo. This is the single highest-value edit on the card.
- Delete a line. Whatever the fourth element on the front is, remove it. The empty space is doing work.
Then use the QR element with your reorder URL — it generates as a real scannable vector, not a picture of a QR code, so it stays crisp when the printer scales it.
When the front is right, duplicate the page for the back rather than starting a second design, so the type sizes and the margin stay identical across both sides. If you've set up a brand kit, apply it and the cream, charcoal and amber come in from your palette instead of being eyeballed.
Open Studio with the candle-card prompt already filled in → — then swap the brand name for yours.
Print specs, briefly
3.5 × 5in is the sweet spot — big enough to hold a headline, small enough to sit on top of the product without bending. 350gsm minimum; anything flimsier reads as a receipt and gets treated like one. Keep everything important 4mm inside the trim edge, and export at 300 DPI. If your printer asks for bleed, extend the background panel 3mm past the edge on all sides.
One card, one ask, one code you can actually measure. That's the whole thing.