Exact models first. The right accessory second.
Electronics shoppers type model numbers, compare specs and worry about compatibility. AACSearch protects identifiers from typo correction, turns specs into filters, and answers “will it fit?” from your own catalogue.
Queries plain keyword search gets wrong
iphone 15 pro 256model + storageSM-S928Bmanufacturer codeтелевизор 55 дюймов oled до 80000size, panel, priceзарядка для macbook air m2compatibilityпылесос для шерсти кошкиuse case, not a specнаушники с шумодавомfeature slangWhere search loses customers today
Model numbers get “corrected”
Typo tolerance turns “iphone 15” into “iphone 13” and a manufacturer code into noise. The exact product the visitor wants lands on page two.
Specs buried in text
“55 inch”, “256 GB”, “OLED” sit in descriptions instead of facets, so the visitor cannot narrow the list.
Accessories that don't fit
Generic “similar products” suggest a USB-C cable for a Lightning phone. Every wrong suggestion costs trust and creates returns.
Comparison is a support ticket
“What's the difference between these two?” becomes a chat with a human or an abandoned tab.
Use cases, not specs
People search “vacuum for pet hair” or “laptop for video editing”. Keyword search needs the exact words in the title.
How AACSearch solves it, layer by layer
Each card names the capability behind it. Items marked “in development” are designed and scheduled, not live yet.
Finds the right item from the first letters — typos, synonyms, filters, meaning.
Identifier protection
SKUs, model numbers and EANs match exactly and rank first; typo correction is applied to words, not to codes.
Specs as filters
Screen size, storage and price ranges become facets with live counts, including numeric range sliders.
Use-case search
Hybrid semantic search matches “vacuum for pet hair” to products whose descriptions mention fur and brush rolls.
Widens the basket: similar, frequently together, personal.
Alternatives in the same class
Similar products by specs and embedding when a model is out of stock or over budget.
Promote and protect margins
Pin the flagship for its launch query, bury discontinued models, schedule a promo window.
Answers questions a filter can't, citing your own catalogue and docs.
Spec questions answered
“Does it support HDR10+?”, “what's the warranty?” — answered from product pages and manuals with links to the source.
Comparisons
“Compare these two TVs” — a short, cited answer built from both product pages.
Learns how items, queries and topics connect — and leads the visitor to the goal.
Compatibility graph
Will connect products through requires and excludes relations, so “for MacBook Air M2” returns only chargers that fit and never suggests ones that don't.
Bundles that make sense
Learns which accessories are actually bought with each device and suggests them in the cart.
Query refinements
Learns from sessions how shoppers narrow “телевизор” into size and panel type, and offers those steps as refinements.
One visitor, from first query to goal
What the visitor does on the left, what AACSearch does on the right.
- 1
Find
Types “macbook air m2 15”.
The exact model ranks first; the code is not “corrected” into another one.
Instant search - 2
Narrow
Filters by memory and colour.
Facets show only configurations that are in stock.
Facets and filters - 3
Ask
Asks whether it can drive two external monitors.
Answers from the spec sheet with a citation instead of guessing.
AI answers with citations - 4
Basket
Adds the laptop to the cart.
Suggests a compatible charger, sleeve and adapter; incompatible accessories are excluded.
Product graph· in development
Your first two weeks
A typical rollout for this industry. Most teams are live on day one and spend the rest of the time measuring and tuning.
Day 1
Import the catalogue with model numbers and specs as fields.
Day 2
Check exact-match queries for your top models; add spec facets.
Week 1
Index manuals and spec sheets for AI answers; add manual `excludes` for known incompatible accessories.
Week 2
Turn on bought-together rows in the cart and compare attach rate.
What to measure
The numbers to judge a rollout by. Compare them with an A/B split or a before/after window on the same traffic.
- Exact-match success
- Share of model-number queries where the exact product is in the top 3. Should be near 100 %.
- Search conversion
- Purchases per search session — the headline number.
- Accessory attach rate
- Orders with a device and at least one accessory.
- Returns for incompatibility
- The cost of wrong suggestions; compatibility rules should push it down.
Connects to
- 1C-Bitrix
- Shopify
- WordPress / WooCommerce
- YML feed
- CSV / JSON
- REST API
Capabilities used
Coming next
Questions
Search for Electronics & appliances, live this week
Connect a feed or a site, embed one script, and measure the difference on your own traffic.