Search that forgives typos, builds its own synonyms and searches across brands, specs and categories — a replacement for your store's built-in search in minutes.
A visitor who uses your on-site search is already almost a buyer: they know exactly what they want and are looking for where to get it. That makes site search one of the most valuable spots in a store — and one of the most costly when it fails. Standard search in most CMSs matches exact strings and gives up on the first typo. Someone types "wireles headphons" instead of "wireless headphones" and sees "nothing found", even though the item is in stock.
Looma's quick search closes that gap. It's a Google-grade smart search that installs with a single line of code, replaces your store's built-in search, and understands the customer even when they misspell, type in the wrong layout, or speak instead of type. Here's a full breakdown of how it works and what you can configure.
Quick search is a dedicated search window layered over your site that opens when a customer clicks your usual search field. As they type, they instantly see relevant products, categories and suggestions — with no page reload. Out of the box it can:
The core idea: the customer searches in their own words, not in catalog jargon, and still gets an exact result. Let's go part by part.
Customers rarely type names perfectly. They spell by ear, mix Latin and Cyrillic layouts, truncate words and make typos. Ordinary search answers "nothing found" — and the store loses a sale for no reason, with the product sitting in stock. Looma's search runs on OpenSearch and recognizes intent, not letters:
Notice the left column: search doesn't just return a flat list — it groups results into categories with counts ("Bedroom dressers — 76", "Dressers — 40") and lets the customer narrow down in one click. No need to scroll hundreds of items — they reach the right subset in a second.
The most interesting part starts where even smart search finds nothing. Looma doesn't let that slide: every week it analyzes all queries that returned zero results and decides for itself what's missing.
Here's how it works. Once a week Looma gathers the zero-result queries from the last 7 days — not random one-offs, but the ones that repeated several times (real demand, not noise). Then, for each one, AI picks synonyms: a translation, a fix for a common typo, an alternative name, a brand variant. Those synonyms feed straight into search — and next time the same query finds the product.
You don't need to maintain synonym dictionaries by hand or guess what customers call your products. Search learns from real queries and closes the blind spots in your assortment itself. If needed, you can see the generated synonyms in the dashboard, disable an odd one or add your own — the human stays in control.
Customers don't always search by product name. Often they search by a parameter: brand, size, color, material. Looma extracts these attributes from every product card in advance (via AI Magic), so search works across all fields at once:
Type part of a brand name and Looma completes the rest, lifting the right products to the top of the results.
A query like "160x200" or "table 120 cm" is recognized as a size, filtering out everything that doesn't fit.
Search returns not only products but also relevant categories — the customer jumps straight into the right section.
Latin in a Cyrillic catalog and vice versa — "komod" finds "комод", and "диван" finds "dyvan".
When a customer opens search but hasn't typed anything yet, that's not an empty screen — it's a storefront. Looma shows several blocks there that pull people back toward a purchase:
People return to what they already searched — no need to type the query again.
What people search most on your site — social proof plus a fast way into your best-selling sections.
Right inside search the customer sees items they already looked at, and easily returns to compare and move toward an order.
Show best-sellers automatically — or hand-pick which products to feature on the start screen.
Not everyone likes typing a query, especially on a phone. So search has voice input built in: the customer taps the microphone, says the query in their own words — and instantly sees results. It's faster than typing and more natural for mobile shoppers.
As soon as the customer types the first letters, search shows autocomplete — pick an option with a single tap or with the keyboard arrows, without finishing the query. It speeds up the path to the product and nudges people toward the right phrasing.
And every customer query is saved — and that's more than a convenience history. For the store it's the most honest demand data: what people search most, what they can't find, which products are worth adding. Together with the weekly review of zero-result queries, this turns search into a constant source of insight about your customers.
You configure the look and behavior of search yourself, in the dashboard, with no developers. You can change a lot:
All of it is visible in the dashboard with a live preview: flip the toggles and immediately see how the product card in the results will change. No code, no developers.
And most importantly — right from the search results the customer jumps to the AI assistant in one click to ask about the products they found. Search and chat aren't two separate tools, but one path to purchase.
Looma search doesn't require rebuilding your site and doesn't depend on your CMS:
After that, quick search replaces your store's standard search, and customers start finding products even when they type with mistakes, in Latin script, or by voice.
Site search is where a store meets a customer who's ready to buy. Standard search often ruins that meeting over a single typo. Looma's smart search works differently: it forgives mistakes, builds its own synonyms from real queries, searches across brands, specs and categories, understands voice, and turns even an empty screen into a storefront that sells.
Want to see what else the platform is made of and how search works alongside the AI chat? Read the full Looma overview →
Yes. Looma's search recognizes intent, not an exact letter match: it finds products with typos, truncated words, the wrong keyboard layout, and Latin script in a Cyrillic catalog.
AI generates them itself. Once a week the system analyzes queries that returned zero results and repeated several times, picks synonyms for them (translations, fixes, alternative names) and adds them to search. You can review and adjust them in the dashboard.
Yes. Quick search opens when a customer clicks your usual search field and fully takes over that function — with no rebuild of the site itself.
Brand color, placeholder text, product card size and contents, number of columns, start-screen blocks, pagination mode, badges and even custom CSS — all in the dashboard, no developers.
Yes. The customer taps the microphone, dictates the query — and instantly sees results. It's especially handy for shoppers on phones.
Looma installs in minutes and starts selling for you 24/7.
Try it free →Looma installs in minutes and sells for you 24/7.
