# Recommendations: Pinning and excluding products for curated results

> Hello Retail's Product Recommendations lets merchants pin products to bought-together and alternatives slots, and blacklist products that should never appear as pairs.

**Author:** Ecaterina Capatina
**Published:** May 21, 2026
**Tags:** product-recommendations, personalization, merchandising, ecommerce

---

Hello Retail's Product Recommendations engine automates product discovery across every page - but some products need a human hand. The platform now lets merchants pin specific products to "bought together" and "alternatives" recommendation slots, and blacklist products that should never appear together. The result: an algorithm you can shape product by product, without rebuilding your entire recommendation strategy from scratch.

## Why automated recommendations still need a curation layer

Product recommendations are one of the highest-leverage touchpoints in ecommerce. <a href="https://www.mckinsey.com/capabilities/growth-marketing-and-sales/our-insights/the-value-of-getting-personalization-right-or-wrong-is-multiplying" rel="nofollow">McKinsey research puts the revenue lift from personalization at 10–15%</a> across the customer journey, and product recommendation widgets on product detail pages are typically where that lift is most concentrated. At the same time, <a href="https://www.epsilon.com/us/about-us/pressroom/new-epsilon-research-indicates-80-of-consumers-are-more-likely-to-make-a-purchase-when-brands-offer-personalized-experiences" rel="nofollow">Epsilon found that 80% of consumers are more likely to purchase from brands that offer personalized experiences</a> - which means the quality of what you surface matters as much as the act of surfacing something at all.

Hello Retail's automated system, backed by Product Intelligence, handles most of this at scale. It reads behavioral signals - views, clicks, purchases - and builds recommendation logic that works even for products with limited purchase history. Ecommerce managers get relevant suggestions across an entire catalog without manually curating every pairing.

The edges are where the algorithm runs short. A fully automated system has no way of knowing that the Hawaiian shirt in your summer collection should never appear next to last season's formal blazers. It doesn't know you've committed to always featuring a brand partner's swimming trunks alongside a specific hero product. Those are business-logic decisions that sit above the data, and until recently there was no clean way to act on them inside a recommendation platform.

## Pinning and excluding: How it works in Hello Retail

The feature is accessible from the Products section of the Hello Retail dashboard. Merchants navigate to any individual product, open its settings, and find overrides for two core recommendation algorithms: **alternatives** and **bought together**.

**Excluding products from alternatives:** If two products share a category but clash in style, brand positioning, or margin profile, you can blacklist one from the other's alternatives feed. The change is immediate - remove a product from the algorithm and it disappears from that recommendation slot. No code change, no waiting for a re-index.

**Pinning products to "bought together":** The bought-together algorithm draws on historical co-purchase data to surface what customers actually buy alongside a given product. That signal is usually reliable, but it takes time to build and it can't capture deliberate merchandising intent. Pinning lets you add specific products to that slot directly, and you can set their display order so your highest-priority pairings appear first in the widget.

Once you save, the curated recommendations go live on the product page. The session demo makes this tangible: pin a pair of swimming shorts to a Hawaiian shirt's bought-together widget, refresh the product page, and those shorts appear in the curated slot immediately - no delay, no deployment.

## When curation earns its keep

The algorithm handles scale. Curation handles intent. For most of a catalog, pure automation is the right default - the system finds the right signals and surfaces relevant products without any manual input. The effort of curating every SKU individually would outweigh the gain.

Where curation pays off is a smaller set of high-priority cases. The session identifies the clearest ones: high-margin movers and high-brand-priority products where the recommendation output is visible enough to warrant direct control. A fashion retailer might guarantee that a statement necklace always appears alongside a specific dress - the kind of styling decision that a buying team would make, not an algorithm. A sports retailer might pin complementary items to a new product launch before co-purchase data has had time to accumulate.

<a href="https://www.salesforce.com/resources/research-reports/state-of-the-connected-customer/" rel="nofollow">Salesforce's State of the Connected Customer report found that 66% of customers expect brands to understand their unique needs and expectations</a> - a bar that keeps rising. A recommendation pairing that feels random or off-brand doesn't just fail to convert. It erodes trust in the widget as a whole. The exclusion tool is as important as the pinning tool: knowing what should never appear together is part of knowing what should.

The two controls together let you apply human judgment selectively, on the products where it adds the most signal, while the algorithm continues to run everything else. Automation handles the catalog at scale; pinning and exclusion give merchandisers surgical control over the handful of decisions that genuinely warrant it.

## Getting started

The feature is available today to all Hello Retail customers with Product Recommendations active in their dashboard. No additional configuration is required - the pin and exclude controls appear under the product settings for any item in the catalog.

The practical workflow is straightforward: identify your seasonal hero products, your key high-margin items, and any pairings that matter to brand partners. Open each product in the dashboard, set the overrides for the alternatives and bought-together algorithms, and save. Changes go live immediately and can be revised as the catalog changes, as promotional periods shift, or as new co-purchase data accumulates and the algorithm's own recommendations improve.

Automation remains the foundation. Curation is the layer that makes it yours.

---

*This content is from the Hello Retail blog. For the full experience with images and formatting, visit [helloretail.com/en/blog/recommendations-pinning-excluding-products](https://helloretail.com/en/blog/recommendations-pinning-excluding-products)*
