Frequently bought together: turning order history into offers
Frequently bought together reads your store's own orders, finds the products that ship together, and offers the pair to the next shopper. No guessing: the orders already voted. Show the pair as a full-price recommendation or as a discounted bundle, on the product page, in the cart, or in search results.
Somewhere in your order history, two products keep leaving in the same box. Shoppers wrote that pattern, order after order, until it became a fact. Frequently bought together (FBT, from here on) is the discipline of finding those pairs and putting them in front of the next buyer. The raw material is already yours.
What is frequently bought together?
Frequently bought together shows a shopper the products other customers bought alongside the item on screen. The pairs come from your store's own orders, and that sourcing is the entire point.
A generic "related products" widget guesses from categories: shirt, therefore more shirts. Co-purchase data answers a sharper question: what did people who bought this shirt put in the same order? Often it is not a shirt at all. It is the trousers, the belt, the wash kit. Your customers drew the map. FBT reads it back to them.
How do you read co-purchase patterns?
Look for pairs that repeat across many orders, then sort them into two kinds: confirmations and surprises. Both earn; the surprises earn from territory only your store can see.
Confirmations match what you would have guessed: camera and lens, planner and pen. The surprises are why order history beats intuition: the candle that ships with the bath salts, the dog leash that travels with the car seat cover. No category logic predicts those pairs. Orders record them anyway.
In Sledge Recommendations, frequently bought together is one of thirteen recommendation types, built from your orders automatically. Manual picks override any pairing, product by product, so evidence does the heavy lifting and you keep the veto.
Where does frequently bought together work best?
On the product page first, then the cart, then search results. The pair follows the shopper to wherever the buying decision is happening.
The product page is the classic seat: the shopper studies the camera, and the lens appears below with its case. In the cart drawer, the pairing reads as one last nudge: the watch is in the cart, here is the strap, one tap before checkout. In search results, the pair rides along with the searched product, reaching shoppers the moment they have named what they want.
Sledge places offers across instant search, search results, collections, product pages, the cart drawer, the cart page, and the thank-you page. Sell where shoppers decide. Start on the product page, prove the pair earns, then add seats.
Should the pair be a bundle or a recommendation?
Show a recommendation when the pair already sells itself; build a bundle when the pair needs a price to close. Same two products, different amounts of persuasion.
A recommendation asks for nothing: the strap appears beside the watch at full price, and convenience does the selling. Full margin, nothing to discount. If your orders show the pair selling together unprompted, this is all it needs.
A bundle adds a deal: both products priced together, the combined compare-at price struck through, ten to twenty percent off. Reach for it when the pair makes sense but does not happen on its own, or when the attach is stock you want moving. The working order: run the recommendation first, since showing it costs nothing, and promote stubborn pairs to bundles.
Can Growth Intelligence find the pairs for you?
Yes. That is the job. Built-in Growth Intelligence reads your orders, names the specific pair worth acting on, and attaches a dollar estimate to the suggestion.
A finding reads like advice from someone who studied your store: shoppers who buy the planner often add the pen refills, offer them together, worth about $180 a month. Click it and the offer opens pre-filled: review, save, live. Then comes the part guessing never delivers: Growth Intelligence measures the offer against real sales and shows the estimate beside the measured result. Every suggestion is checked against real sales, so you always know what's working. And why.
How do you know the pairs are earning?
Read revenue per offer and per placement. The product-page pairing and the in-cart pairing report separate lines, so each seat proves itself.
Read the lines the way you read payroll: each placement either earns its spot or it does not. A/B testing settles the close calls on the offers themselves: this bundle against that one, with revenue per variant deciding.
The FBT checklist
- Pairs come from your orders, not category guesses.
- Confirmations ship as-is; surprises get a quick sense check, then ship.
- Wrong pairings get overridden by hand, product by product.
- Product page first; cart and search results once the pair proves out.
- Pairs that sell unprompted stay full price as recommendations.
- Stubborn pairs get promoted to a discounted bundle.
- Every placement reports its own revenue line.
Related reading
Recommendations · Sledge Upsell · What is frequently bought together? · Bundle pricing strategy · Sledge Bundles
Questions, answered
How is frequently bought together different from related products?
Related products guess by category: a shirt suggests other shirts. Frequently bought together reads real orders: the shirt suggests whatever customers added to the same purchase. Evidence over resemblance.
How many orders do I need before frequently bought together works?
Enough for pairs to repeat, and your bestsellers gather co-purchase data fastest. Wherever the data runs thin, set pairings by hand; in Sledge, manual picks override suggestions either way.
Should frequently bought together products be discounted?
Not by default. A pair that already sells together at full price needs placement, not a markdown. Save the discount, and the bundle format, for pairs that make sense but need a nudge.
Where should I show frequently bought together first?
The product page, beneath the item being considered. It is the seat with the clearest buying moment. Once that placement earns its revenue line, add the cart drawer and search results.
Your orders already voted.
Install free. By tomorrow, built-in Growth Intelligence reads your order history and names the pairs worth offering, each with a dollar estimate attached.
14-day free trial · No code · Works with any theme · Uninstalls clean