Algolia and Velou logos side by side announcing the acquisition — Algolia's announcement graphic for the Velou acquisition. Image: Algolia.

Good ecommerce search has always been about speed. Algolia now says the next round will be won on how well a store understands its own products.

On October 6, 2026, Algolia announced it had acquired Velou. The small New York startup turns product photos and descriptions into detailed catalog data. The deal is tiny in team size. Its message for online stores is much bigger.

I'm Nitin from the Debugged Pro team. We build and maintain online stores for clients in India and the USA, and search is the feature they complain about most. Here is what this deal means for store owners and the agencies that support them.

What Algolia Announced About Ecommerce Search

According to Algolia's official press release, Velou uses multimodal AI to extract product data from catalog text and images. That means it reads both the words and the pictures.

The key facts are simple:

Buyer: Algolia, which says it handles nearly two trillion searches a year for more than 18,000 customers.
Target: Velou, which already powers search experiences for more than 50 retail and premium brand enterprise customers.
Price: Not disclosed.
Team: Velou's six staff members join Algolia.
Customer impact: Existing APIs, SDKs and integrations stay the same.

Business news site citybiz, which also covered the deal, described the core problem well. Product catalogs often lack enough structured information to answer detailed customer searches.

Why Ecommerce Search Breaks on Thin Product Data

Most site search problems are not search engine problems. They are data problems.

Suppliers send product feeds in different formats. Titles are short. Key details live only in the photo or in a PDF size chart. The search box can only match what the catalog actually says.

The "Blue Dress" Problem

Algolia used an example that every store owner will recognise. A shopper types "machine-washable navy midi dress for a fall wedding." The catalog simply says "blue dress."

The store has the right product. The shopper never sees it, or it sits behind dozens of other blue items.

Teams usually patch this by hand. They add synonyms, boosts, redirects and manual tags. In his blog post on the deal, Algolia CEO Stephen Lynch wrote that this work "takes months, and it never finishes, because the catalog changes every day."

Shoppers notice the gaps. Baymard Institute has studied site search usability on large retail sites for years. Its ecommerce search research is a useful reference when you audit a client store.

What Velou Adds to Algolia Ecommerce Search

Velou has spent eight years building two things. The first is a retail taxonomy and product graph, which is a structured map of how products, categories and attributes relate. The second is an engine that reads descriptions and photos to pull out specific attributes.

For a dress, that could mean colour, length, fabric weight, care instructions and occasion. Each attribute is tied to evidence in the source text or image.

That evidence trail matters. Lynch put it bluntly in his post: "A confident wrong attribute is worse than a missing one." A wrong tag can cost a sale or cause a return.

Algolia says Velou's data will sit underneath several products at once:

Search: Better results for long, specific queries.
Facets and filters: Shoppers can filter by details such as fit, fabric or occasion once those attributes exist.
Recommendations: Stronger suggestions based on what products actually are.
New arrivals: Products can rank before they have any click history.
Agent Studio: Algolia's tool for building AI agents gets richer product records to work with.

Algolia also expects fewer hand-written merchandising rules. For busy store teams, that may be the most practical benefit.

The Numbers Behind the Deal

Algolia shared two customer results from Velou deployments. Both are UK fashion retailers.

Bar chart of results reported by Algolia and Velou: Get The Label site search revenue up 60 percent, Everything5pounds catalog attributes up 85 percent and search conversions up 33 percent, and AI-referred US retail traffic up 62 percent — Figures reported by Algolia and Velou on October 6, 2026. Chart: Debugged Pro.

Get The Label added more than 190,000 product attributes over six months. Revenue from its site search rose 60%.
Everything5pounds grew its catalog attributes by more than 85%. Its search conversions rose 33%.

Treat these as vendor-reported case studies, not independent research. Lynch himself wrote that every catalog is different and that he would not promise those numbers to anyone. We think that is the right way to read them.

Why AI Shopping Agents Raise the Stakes for Ecommerce Search

The deal is not only about the search bar on a store's own site. It is also about AI assistants that shop on a customer's behalf.

Algolia notes that shoppers increasingly start in ChatGPT, Microsoft Copilot, Muse and Google Gemini. It cites Adobe Analytics data showing AI-referred traffic to US retail sites rose 62% year over year in July.

Here is the catch. These agents do not browse pages the way people do. They read structured product data. If a detail is missing, the product is close to invisible to them.

Lynch summed it up in the release: "An AI agent can only successfully recommend a product that the data accurately describes."

IDC analyst Heather Hershey made the same point. "Product data has long been treated as back-office plumbing," she said. "AI shopping agents change that, because they pick products based on what the data says, not how a page looks."

This links straight to SEO. Google already asks stores to describe products with structured data. Its product structured data guidelines explain how price, availability, reviews and other details can appear in search results.

Algolia homepage showing an AI agent shopping assistant and a news banner about the Velou acquisition — Algolia's homepage, featuring its agentic search pitch and a banner about the Velou deal. Image: Algolia, via citybiz.

What This Means for Agencies in India and the USA

Agencies that build stores sit right in the middle of this shift. We set up catalogs, connect search tools and write the product templates. So the data quality problem often lands on our desk.

If Your Clients Already Use Algolia

There is nothing to migrate today. Algolia says Velou's enrichment will become available inside the platform. It also plans a webinar with a live walkthrough soon.

Until then, a few sensible steps:

Ask your Algolia contact when enrichment reaches your client's plan and region.
Check how pricing works, since Algolia has not shared it yet.
Pull a baseline now: zero-result rate, search conversion rate and revenue from search.

If Your Clients Use Another Search Tool

The lesson still applies. Whether a store runs on Shopify, Magento, WooCommerce or a custom stack, richer attributes help every search engine. They also help AI agents and Google.

Many Indian agencies run catalogs for brands selling to US buyers. Those shoppers search in very specific terms, such as fabric, fit and use case. A thin catalog leaves those sales on the table.

If your team is also exploring meaning-based search, our breakdown of semantic search with EmbeddingGemma 2 covers the model side of the same problem.

A Practical Ecommerce Search Checklist for Product Data

You do not need an acquisition to start. Here is the checklist we use when a client says "our search is bad":

Read the zero-result report. Export the last 90 days of searches that returned nothing. This is your to-do list.
Pick the attributes that matter. For fashion that might be fabric, fit, length and occasion. For electronics, it could be ports, battery life and compatibility.
Normalise values. "Navy," "dark blue" and "midnight" should map to one colour family that filters can use.
Fill gaps from images carefully. AI tools can suggest attributes from photos. Have a person review a sample before you publish them.
Add structured data. Make sure product pages carry accurate product markup for Google and other systems.
Measure monthly. Track search conversion and zero-result rate after each change.

A Simple Example

Say a Jaipur handloom brand sells cotton kurtas to customers in India and the USA. Its titles read "Blue Kurta" with a size chart image.

A US shopper searches "breathable cotton kurta for summer, machine washable." Nothing useful comes back. After the team adds fabric, care, sleeve length and season attributes, the same kurta can now match that query. It can also appear in a filter for "machine washable."

None of this needs fancy tooling. It needs a clean catalog and a habit of checking what shoppers actually type.

Risks and Open Questions

The deal looks promising, but a few points are still unclear:

Pricing and timing: Algolia has not said when enrichment arrives or what it costs.
Retail focus: Velou's taxonomy is built for retail. Other industries may see less benefit at first.
Accuracy: AI-generated attributes still need review, especially for size, safety and compliance details.
Lock-in: Enriched data that lives only inside one vendor's platform is harder to reuse elsewhere. Ask whether you can export it.

Search and SEO also work together here. Our guide to SEO optimization for ecommerce websites covers the page-level basics that sit alongside good site search.

Our Take on Algolia, Velou and Ecommerce Search

This is a small acquisition with a clear point. Fast search is no longer enough if the catalog cannot describe what is for sale. That applies to the search bar, to Google and to AI shopping agents.

For agencies, the work does not change much. It just matters more. Clean product data used to be a nice extra. Now it decides whether an AI assistant can recommend a client's product at all.

If you want help auditing a client catalog or planning search for a new store, our software development consulting team can help you map the work.

Key Takeaways

Algolia acquired Velou on October 6, 2026 to add AI product data enrichment to its platform. Terms were not disclosed.
Velou reads product text and images to extract attributes tied to evidence in the source.
Existing Algolia APIs, SDKs and integrations stay the same.
Reported results include a 60% rise in site search revenue at Get The Label and a 33% rise in search conversions at Everything5pounds.
AI shopping agents rely on structured product data, so thin catalogs are harder for them to recommend.
Agencies can improve ecommerce search today by fixing attributes, values and structured data.

Frequently Asked Questions

What is Velou and why did Algolia buy it?

Velou is a New York-based startup that uses multimodal AI to read product text and images and turn them into structured product attributes. Algolia bought it to add that product understanding under its search, recommendations, personalization and Agent Studio tools.

Do Algolia customers need to change their integration?

No. Algolia says its APIs, SDKs and integrations stay the same. Velou's enrichment will become available inside the platform, so there is no migration project.

How much did Algolia pay for Velou?

The price was not disclosed. Algolia confirmed that Velou's team of six is joining the company.

Why does product data matter for AI shopping agents?

AI assistants such as ChatGPT, Microsoft Copilot and Google Gemini read structured product data rather than browsing pages like people do. If a detail is missing from that data, the product is much harder for an agent to find or recommend.

How can I improve ecommerce search without Algolia?

Start with the catalog. Add specific attributes such as material, fit, care and use case, clean up inconsistent values, add product structured data to your pages and review your zero-result searches every month.

About the Author

Nitin Kumar is the Lead Project Coordinator at Debugged Pro. Nitin works with clients in India and the USA on web, mobile, ecommerce and AI projects, turning new tech news into practical plans that teams can actually use.