A question that keeps coming up in the Shopify Community is some version of: “Do I actually need a paid recommendations app, or does Shopify’s built-in tool already do this?” It’s a fair question, because the “Recommended Products” feature that ships inside the free Search & Discovery app looks, on the surface, a lot like what personalisation apps charge a monthly fee for.
The honest answer is that both can produce a “customers also bought this” style module on your product pages, but they arrive at those suggestions differently, and the gap between them widens as your catalogue, traffic and sales history grow. Some stores genuinely don’t need to spend a cent beyond what’s already in their Shopify plan. Others are leaving conversion rate on the table by sticking with the native tool for too long.
This post walks through how Shopify’s native recommendation engine actually works, where its manual override option covers most of its weaknesses, what third-party personalisation apps add on top, and a practical framework for deciding which camp your store falls into before you install (or pay for) anything.
What “product recommendations” means on Shopify
On a Shopify store, product recommendations usually show up in one of three places:
- The “You may also like” or “Related products” block near the bottom of a product page
- Cross-sell prompts in the cart drawer or on the cart page
- Post-purchase or thank-you page upsells
The native Search & Discovery app powers the first of these out of the box, and with some theme or app support can extend into the cart. Third-party personalisation apps typically cover all three, plus additional surfaces like homepage “recommended for you” grids, email and on-site pop-ups, and quiz-based recommendation flows.
How Shopify’s native recommendation engine works
Shopify’s free Search & Discovery app includes a “Recommended products” feature that most themes reference via the product-recommendations section (built on Shopify’s recommendations.products object in Liquid). It generates suggestions using two inputs:
- Sales and browsing pattern data, Shopify looks at what’s frequently bought or viewed together across your store’s order history to identify products that tend to convert well alongside each other.
- Product catalogue attributes, when there isn’t enough sales history for a product (a new launch, for example), Shopify falls back to matching on product type, vendor, tags and price range to find reasonably similar items.
This is a genuinely useful default. It requires no setup, no app install beyond what’s already free in your Shopify admin, and no ongoing subscription. It also improves over time as more orders come through, because the “bought together” signal gets stronger with more data.
Manual override: the underrated part of the native tool
What a lot of merchants don’t realise is that Search & Discovery lets you manually curate recommendations on a per-product basis. In the app, under the Recommendations tab, you can search for a product and hand-pick which items should appear as its related products, overriding Shopify’s automatic pick entirely for that product. This matters more than it sounds. It means you’re not stuck with whatever the algorithm decides; you can steer recommendations toward higher-margin items, seasonal stock, or products you know pair well together even if the sales data hasn’t caught up yet.
In our experience auditing Shopify stores, this manual override is one of the most under-used features in the entire admin. Merchants either don’t know it exists or assume they need a paid app to control recommendations at all, when for a huge number of stores, ten to twenty minutes curating your highest-traffic product pages achieves most of what a personalisation app would.
Where the native engine falls short
The native tool has real limits, and it’s worth being upfront about them rather than pretending Shopify’s free option does everything a $50-plus-a-month app does.
- No true 1:1 personalisation. Recommendations are the same for every visitor looking at a given product. There’s no adjustment based on an individual shopper’s browsing history, location, past purchases, or where they came from (a Google Ads visitor vs an email subscriber, for instance).
- Limited surfaces. Out of the box it’s product-page focused. Getting recommendations into the cart drawer, checkout, or post-purchase flow generally needs custom theme development or an app, native tooling doesn’t cover this comprehensively on its own.
- No behavioural real-time logic. It won’t adapt in-session, for example, showing a different recommendation set because a shopper just added a specific item to their cart, or because they’ve viewed three products in the same category in the last two minutes.
- Basic reporting. You can see that recommendations exist; you can’t easily see click-through or attributed revenue by recommendation without pulling that data together yourself in GA4 or another analytics tool.
- No testing framework. There’s no built-in way to A/B test one recommendation logic against another.
None of these are dealbreakers for every store. They’re simply the trade-offs of a free, general-purpose tool versus a paid, purpose-built one.
What personalisation apps actually add
Paid Shopify personalisation and recommendation apps (the category includes tools like Rebuy, LimeSpot, Nosto and various bundling/upsell apps with recommendation features) generally layer in some combination of:
- Behavioural, session-based logic, recommendations that shift based on what a specific shopper has viewed, added to cart, or purchased before, sometimes using machine learning models trained across a much larger dataset than one store’s order history.
- Multi-surface placement, the same recommendation engine can populate the homepage, product page, cart drawer, checkout (on Shopify Plus, or via post-purchase extensions), abandoned cart emails and pop-ups from a single dashboard.
- Merchandising rules layered on top of AI, for example, “always prioritise items with more than 20 units in stock” or “never recommend products from a discontinued collection,” combined with the algorithm’s suggestions.
- Built-in A/B testing, the ability to run one recommendation algorithm or layout against another and measure the lift in add-to-cart rate or revenue per visitor.
- Reporting dashboards, attributing revenue specifically to recommendation widgets, which makes it much easier to judge whether the tool is paying for itself.
The trade-off is cost (typically an ongoing monthly subscription, often scaling with order volume or store traffic), an extra app affecting page load, and another piece of software your team needs to configure and maintain.
A framework: when native is enough vs when a paid app earns its keep
Rather than defaulting to “more tech is better,” it helps to run through a short checklist before adding a personalisation app to your stack.
Stick with the native Search & Discovery tool if:
- Your catalogue is under roughly 200-300 SKUs and product relationships are fairly obvious (a shopper looking at a jacket is reasonably well served by “other jackets” or “matching accessories”)
- You’re a newer store still building order history and traffic volume
- You haven’t yet manually curated recommendations on your top 20-30 product pages by traffic
- Site speed and simplicity are a priority (fewer apps generally means a leaner theme and fewer third-party scripts to manage)
- You don’t yet have the analytics maturity to interpret a recommendation-specific reporting dashboard
A paid personalisation app is genuinely worth evaluating if:
- You have a large or complex catalogue where “bought together” data alone doesn’t capture real cross-sell logic (multi-variant products, bundles, complementary categories that aren’t obviously linked)
- You’re running meaningful email and on-site personalisation elsewhere and want recommendations to feel consistent across channels
- You want to A/B test recommendation placement or logic rather than guess
- Your average order value would benefit meaningfully from stronger cart-level upsells (a small percentage lift on AOV at real volume can outweigh the app’s monthly cost)
- You’ve already exhausted manual curation on the native tool and are still seeing recommendation click-through underperform
When to bring in a specialist
Deciding between “free and manual” and “paid and automated” isn’t really an app decision in isolation, it’s a conversion rate decision. The right call depends on your current add-to-cart rate, where shoppers are dropping off, your margin structure, and whether recommendations are even the highest-leverage lever to pull right now versus, say, checkout friction or page speed.
This is exactly the kind of trade-off a Shopify conversion rate optimisation engagement is built to work through, testing whether a recommendation change (native curation or a paid app) actually moves add-to-cart and revenue-per-visitor, rather than installing a tool on faith and hoping. If you’re not sure which side of the checklist above your store falls on, that’s a reasonable point to get a second, structured opinion before committing to a monthly subscription.
A practical rollout checklist
However you decide to proceed, this is the sequence we’d recommend:
- Audit your current product pages, is Search & Discovery’s “Recommended products” section even installed in your theme?
- Identify your top 20-30 product pages by traffic (Shopify Analytics or GA4) and manually curate recommendations for those first
- Track baseline click-through and add-to-cart rate on recommendation widgets for at least 2-4 weeks
- If manual curation plateaus and you have a genuine case (complex catalogue, multi-channel personalisation, AOV upside), trial one personalisation app rather than several at once
- Set a review point (30-60 days) to compare recommendation-attributed revenue against the app’s subscription cost before deciding to keep it
FAQ
Does Shopify’s free plan include product recommendations?
Yes. The Search & Discovery app is free on all Shopify plans and includes the “Recommended products” feature, which most current Shopify themes can display via a built-in section. You don’t need a paid plan tier or a separate app to get basic recommendations working.
Can I control which products Shopify recommends, or is it fully automatic?
You can override the automatic recommendations. Inside the Search & Discovery app, the Recommendations tab lets you search for any product and manually choose which items should appear alongside it, which takes priority over Shopify’s algorithm-generated picks for that product.
Will a personalisation app slow down my store?
Any additional app adds some JavaScript and, depending on how it’s implemented, additional network requests. The impact varies significantly between apps and how well they’re built, so it’s worth checking page speed before and after installing one rather than assuming it will be negligible.
How do I know if my recommendations are actually working?
At minimum, track click-through rate on the recommendation module and whether sessions that engage with it have a higher add-to-cart rate than sessions that don’t. Native Search & Discovery doesn’t report this natively, so you’ll likely need GA4 event tracking or a personalisation app’s built-in reporting to see it clearly.
Is it worth paying for a personalisation app if I have a small catalogue?
Usually not straight away. With a small, simple catalogue, manually curated native recommendations tend to perform close to what a paid app would deliver, without the ongoing cost. It’s generally worth revisiting once your catalogue, traffic or AOV grow to a point where behavioural personalisation could plausibly move the numbers.
Ready to find out what’s actually holding back your conversion rate?
Recommendations are one lever among several, checkout friction, page speed and merchandising often matter just as much. A Shopify audit will show you where your store is actually losing conversions, or you can book a call to talk through your specific setup with our team.
