A wedding guest has a problem that a product database does not. She needs something to wear to an occasion. The database has colors, sizes, names and stock codes. Both are talking about the same dress. They have simply neglected to agree on a language.
At The Aje Collective, search data helped make that disagreement visible. The Australian fashion business saw enough interest in wedding-guest dresses to create a dedicated subcategory. A request typed into a small rectangle became a decision about the shop itself. It is a pleasing reversal: the customer, briefly, gets to arrange the shelves.
- Searchspring helps retailers make products easier to find, rank and recommend.
- It now belongs to Athos Commerce, alongside Klevu and Intelligent Reach.
- The enlarged offering connects on-site discovery with external product feeds.
- The practical lesson: study failed searches before spending more to attract shoppers.
A dress by any other name
Searchspring’s territory is the gap between an available product and a discoverable one. A retailer can own the stock, photograph it beautifully and pay to bring someone through the digital door, only to lose the sale at the search box. The merchandise has done nothing wrong. The route to it has.
Aje’s older tools struggled when queries diverged from descriptions, while manual merchandising consumed time. With Searchspring on Shopify and agency Convert Digital handling the wider site project, the team could examine queries and automate ranking. Its published case study reports two to three working days saved each week. That is a customer-reported result, not a promise for the next store.
The distinction matters because product discovery is easy to confuse with advertising. Advertising brings someone to a shop. Discovery decides whether the shop understands the request. Searchspring sells software for that second encounter: search and navigation, merchandising, personalization and the reporting that lets a retailer inspect what happened.
Its customers include brands such as Chubbies, Moen and Peet’s Coffee. That list is useful precisely because it is varied. Clothing, fixtures and coffee make different demands on a catalog. The common buyer is the ecommerce team responsible for turning a visit into a purchase, often while juggling promotions, inventory and a storefront platform someone else maintains.
The merchant needs a steering wheel
The proposition is partly mathematical and partly editorial. Search has to match intent with products. Merchandising has to decide which matching products deserve attention today. A sale, a launch and a nearly depleted size run can each change the answer. Relevance is a commercial decision as well as a linguistic one.
Athos’s merchandising tools allow teams to influence product placement, set ranking rules and place campaign content. This gives the person running the shop a way to express priorities without treating every rearrangement as a development project. Personalization adds another variable: what this particular visitor appears to want. The two must coexist. A merchant may want to promote a collection; a shopper may have arrived for something quite different.
Consider Absolute-Snow, whose shoppers may need an exact component for a binding. Its custom search required repeated manual attention. The replacement combined search with synonyms, rules and reporting. The retailer reports a conversion increase of 1.1 percentage points and items per transaction moving from 1.4 to 1.6. Its case study says day-to-day management takes about five minutes.
“We needed something that just works”Matt Pyne-Madden / Absolute-Snow
Those figures describe one implementation. Still, the underlying problem travels well: the more a purchase depends on precise attributes or compatibility, the less patience a customer has for a search result that is merely adjacent to the answer. A clever suggestion is no consolation when the part does not fit.
Eleven years, then a wider ambition
Searchspring began in 2007, co-founded by Gareth Dismore and Scott Zielinski. Dismore later described roughly eleven years of bootstrapping before Scaleworks acquired the business in 2019. Peter Messana became CEO. PSG invested in 2022. The company’s present backing is therefore a different chapter from its earlier association with Scaleworks.
Acquisitions widened the job description. Nextopia and 4-Tell joined in 2020, Increasingly in 2023, and Intelligent Reach in 2024. Search became entangled with personalization, bundling and product feeds. In January 2025, the combination with Klevu brought the businesses under Athos Commerce. Searchspring’s website now points visitors to that name.
2007 Find products on the store.
2024 Add product feeds and external channels.
2026 Launch the unified Intelligent Discovery Platform.
The commercial logic is straightforward. A product can be easy to find on a retailer’s own site and poorly represented on a marketplace or shopping channel. Feed management deals with that second problem by preparing and distributing product information. Searchspring’s original concern follows the merchandise outside the store.
This is also where its positioning gets interesting. Constructor, another commerce discovery provider, offers search, recommendations and merchant controls. AI alone cannot distinguish Athos in this market. Its particular pitch combines those store-facing functions with Intelligent Reach’s channel and feed capabilities, sold through a broader discovery offering. Buyers should compare the actual workflow they need rather than count how many times a vendor says “intelligent.”
The calendar gets a vote
The formation of Athos and the delivery of a unified platform were separate events. Early communications described a 2025 rollout. On August 26 that year, Athos revised the timetable, aiming for general availability in the first quarter of 2026 and explicitly prioritizing uninterrupted fourth-quarter trading.
The announcement is revealing without being scandalous. It documents a changed schedule and the company’s stated reason: protect peak season and prepare a commercially ready release. It does not establish a platform failure. For a retailer, however, the lesson is substantial. A migration has to work in the shop’s calendar, not merely in the supplier’s presentation.
By February 2026, public documentation described the Searchspring-to-Athos API transition as ready to implement. Beacon 1.0 tracking had entered maintenance mode, with new development directed to Beacon 2.0. On June 10, at Shoptalk Europe, Athos announced immediate availability of its Intelligent Discovery Platform. The combined business reported serving more than 2,700 brands across over 50 countries.
The release included a Conversational Assistant for shoppers, a Channel Assistant for product feeds and a GEO Assistant for discovery through generative AI platforms. These are different assignments: guide a person, help operate listings, prepare a catalog for new discovery surfaces. Treating them as one magical assistant would obscure what a retailer is actually buying.
The bill includes more than software
Athos sells annual, custom-quoted software plans. Onsite Discovery covers search, merchandising and personalization. Offsite Discovery covers feeds. Complete Discovery combines them, with AI capabilities and assistants available as add-ons. The pricing page offers a quote rather than a public dollar amount. A buyer can begin with one side of the problem.
The implementation also consumes resources. Public documentation offers APIs, JavaScript libraries, Snap templates and managed implementations. A direct API build places responsibility for the interface, filtering, pagination and tracking on the retailer’s team. More control arrives with more maintenance. That is a meaningful purchase decision, not a footnote for the developer.
The open-source Snap SDK makes the integration story inspectable. The partner ecosystem includes supported storefront platforms and agencies. Those relationships matter because discovery software has to meet a real catalog and a real storefront. The cleanest demo in the world cannot answer who will maintain your custom filters six months later.
Copy the questions, then run the experiment
The most portable habit here is modest: look at requests that return nothing. Group them. Ask whether the store lacks the product, labels it differently or makes the route unnecessarily difficult. Each diagnosis suggests a different action. A synonym can fix a vocabulary problem. A new category can make an occasion legible. Neither can manufacture stock.
Then test a change against a defined outcome. Searchspring’s documented campaign experiments randomly divide shoppers between variations and can compare different banners, boosts, pinned products and filter orders. This is a stronger way to judge a merchandising decision than admiring it in the admin screen.
Customer comparisons also need care. Shoppers who use search can have more buying intent than those who browse. A higher conversion rate among search users does not, by itself, show how much the software caused. Keep the distinction between an association and an experiment; your budget will appreciate the courtesy.
As a practical inference, a small, simple catalog may not justify an extensive discovery stack. Weak attributes can undermine matching; sparse traffic can make experiments difficult to read; an integration without an owner can become a maintenance burden. The sensible demonstration uses your awkward queries and your products. Let the supplier show what happens when the answer is hard.
Searchspring’s story is ultimately about listening at an unusually useful moment: when a customer says what she wants. Athos is extending that work across more places to shop. The test remains pleasingly ordinary. Can the person find the thing, understand it and buy it? Every additional feature ought to earn its place in that sentence.