Constructor Review: Enterprise AI Product Search and Discovery

Verdict (TL;DR): Constructor is one of the most capable AI product search and discovery platforms available, backed by a 2026 Gartner Magic Quadrant Leader position and 82% customer growth in FY26. It is purpose-built for large retailers — the kind running their own web platform with high search volume and complex product catalogues. For an Australian enterprise retailer with those characteristics, it deserves a serious look. For a Shopify-based store or any retailer without an enterprise budget, it is not the right fit. Sales-led, US-stored, and not built for SMBs.

Score: 13/25 — ⭐⭐⭐ Good

Last reviewed: 26 June 2026 by Adam.

What Constructor is

Constructor (constructor.com) is an AI-powered product search and discovery platform for e-commerce retailers. It was founded in San Francisco and is now headquartered in the US, with offices in multiple countries. The platform is designed to replace or augment a retailer's native search functionality with AI-driven search, autosuggest, browse, product recommendations, and a growing suite of shopping agents.

Constructor's central claim is that its algorithms are tuned for e-commerce business outcomes — specifically revenue-per-session and conversion rate — rather than abstract search relevance metrics. This distinguishes it from general-purpose search engines (like Elasticsearch or Solr) that optimise for relevance without commercial context.

The platform's scale is significant: in FY26, Constructor reported powering 322 billion product discovery interactions, a 266% increase over two years. Customer growth was 82% in FY26. The company was named a Leader in the Gartner Magic Quadrant for Search and Product Discovery in 2026 — a meaningful independent validation in the enterprise software market.

Constructor's product suite as of June 2026 includes:

• Search and Autosuggest — AI-driven on-site search optimised for commercial KPIs
• Browse — personalised category page ranking
• Recommendations — product discovery widgets across the site
• AI Shopping Agent — conversational natural-language search for shoppers
• Product Insights Agent — answers shopper questions on product detail pages
• Merchant Intelligence Agent (MIA) — conversational AI for internal merchandising teams to query and act on search and discovery data
• Attribute Enrichment — AI-generated product attribute enrichment (similar to Lily AI's core offering)
• Retail Media — sponsored product placements within search and browse results
• Collections and Quizzes — curated product pages and guided discovery experiences
• Cross-Channel Discovery — extending discovery AI to email, SMS, mobile, and in-store

This breadth is genuinely impressive. Constructor has moved well beyond pure search into what it calls an "agentic commerce" platform.

What it costs in Australia

Constructor does not publish pricing. It is a sales-led enterprise product. Pricing is usage-based — tied to the volume of search interactions or sessions generated — and is scoped per customer through a commercial conversation.

Independent sources (Vendr, OMR Reviews) suggest average annual contract values are substantial, with some sources citing contracts in the range of US$150,000–$300,000 per year for mid-to-large retailer engagements. These figures are from third-party procurement databases, not from Constructor's own pricing page. (treat as a rough proxy only; actual pricing depends on volume, modules, and negotiation. Contact constructor.com for a commercial discussion.)

For Australian retailers:

• Pricing will be in USD. There is no documented AUD pricing mechanism.
• The minimum viable engagement is likely a six-figure annual USD commitment, which at AUD ≈ 1.40 translates to A$230,000+ per year before overages. (based on third-party sources; verify directly.)
• Constructor offers a complimentary Search Experience Audit as an entry point — a good way to assess whether the platform's approach fits your situation before committing to a sales process.
• GST: As a US-incorporated vendor, Constructor is unlikely to be registered for Australian GST on its own account. Confirm your obligations with your accountant.

How it works in practice

Constructor integrates with a retailer's existing e-commerce platform via API. The integration can be substantial — you are replacing or augmenting your search layer, which touches product feeds, front-end search UI, and analytics infrastructure. Constructor's documentation is thorough (docs.constructor.com), and the company provides implementation support.

Once integrated, the platform ingests a retailer's product catalogue and behavioural data — what shoppers searched for, what they clicked, what they bought. The AI then uses this behavioural signal to rank search results and autosuggest queries in ways that optimise for conversion and revenue, not just keyword match.

The Commerce Reasoning Engine (Constructor's name for its central AI) interprets each shopper's context — session behaviour, past interactions if recognised, query intent — and personalises the experience in real time.

The Merchant Intelligence Agent (MIA, released in 2026 as beta) gives internal merchandising teams a conversational interface to query their search data. Rather than pulling reports from a dashboard, a merchandiser can ask "why is the search for 'blue jeans' converting 30% below average?" and get an AI-generated explanation with recommended actions. This is a meaningful improvement in how merchandising teams interact with their own data.

What it does well

The Gartner Leader position is earned. Being named a Magic Quadrant Leader in Search and Product Discovery 2026 — evaluated across nine critical capabilities and four use cases — is not a marketing claim. It is an independent analyst verdict based on documented customer outcomes and platform capability.

The scale of deployment is real. 322 billion shopping interactions per year is not a startup number. These are live production deployments at significant retailers. The platform is proven at scale.

Revenue-outcome orientation is the right framing. Most site search tools optimise for relevance. Constructor optimises for revenue. For a retailer, that distinction matters: a search that returns the right product but not the best-margin product is still a sub-optimal business outcome. Constructor's KPI alignment with commercial outcomes is a genuine differentiator.

The MIA agent is a useful innovation. Conversational merchandising intelligence — asking questions of your own data in natural language — reduces the need for a specialist data analyst to extract insights from search dashboards. For large retailers with merchandising teams, this speeds up decision-making.

Attribute Enrichment extends the value. Constructor's GenAI attribute enrichment module (similar to Lily AI's core offering) means the platform can also improve product data quality, not just search ranking. This closes a loop that typically requires a separate tool.

SOC 2 Type 2 and ISO 27001 certifications. Independent security audits confirm Constructor's security posture — relevant for any enterprise retailer with customer data obligations.

What it doesn't do well

It is not built for Shopify. Constructor's platform is designed for enterprise retailers running their own web infrastructure. If you are on Shopify — even Shopify Plus — Constructor is not the right fit. Shopify has its own search infrastructure, and integrating a third-party search layer at the Constructor level is not a standard Shopify implementation pattern.

Pricing excludes most Australian retailers. If the proxy pricing from third-party sources is in the right ballpark (six-figure USD annual contracts), the realistic market for Constructor in Australia is a small number of large retailers — the equivalent of department stores, major grocery chains, and large national specialty retailers. The mid-market and SMB are excluded by cost.

No documented Australian data residency. Constructor's security page documents SOC 2 Type 2 and ISO 27001 compliance, data anonymisation practices, and GDPR compliance mechanisms, but does not specifically reference Australian data residency or AWS/Azure Australia region options. Data appears to flow through US and European infrastructure. (no explicit AU data residency confirmed; verify with Constructor's privacy team before signing a contract.)

Australian Privacy Act considerations apply. Constructor processes shopper behavioural data — clicks, searches, session behaviour — and uses it to personalise results. For large Australian retailers, the volume of personal (and potentially identifiable) data flowing through Constructor warrants a thorough privacy assessment under the Australian Privacy Principles. The data anonymisation practices Constructor documents (stripping IP octets, hashing identifiers) are helpful but should be reviewed against your specific obligations.

No transparent pricing. The absence of even a ballpark pricing range on the website makes it hard for procurement teams to assess fit without initiating a full sales process. This is standard for enterprise SaaS but frustrating.

Australian fit

Constructor is technically capable of serving Australian retailers — the platform is cloud-based and API-driven, and geography does not limit functionality. The real AU constraints are:

• Data residency: No documented Australian-region data storage. For retailers with Privacy Act obligations at scale, this requires careful contractual and compliance assessment.
• No AU-specific localisation documented: Currency, GST, or AU-specific search behaviour handling is not documented on the Constructor website. Large AU retailers would need to verify that AU-specific requirements are handled in the integration design.
• No documented AU client base: Constructor's published case studies and customers do not include named Australian retailers. This is not evidence they don't have AU clients — but it means there is no public track record of AU deployment to reference.
• USD pricing: AUD will be a contractual conversion from USD with FX risk.

Australian Fit score: 2/5. Technically capable; no AU data residency, no documented AU clients, and enterprise pricing that excludes the vast majority of Australian retailers.

Alternatives

For Australian retailers evaluating product search and discovery AI:

• Nosto — full-site personalisation platform for mid-to-large Shopify and custom-platform retailers. Overlaps in recommendations and discovery; does not have Constructor's depth in search ranking.
• Lily AI — product attribute enrichment AI for fashion, home, and beauty retailers. Complementary to Constructor rather than a direct competitor — Lily AI enriches the product data that Constructor then searches and ranks.
• Shopify Magic — free, built-in, and AUD-native. Not enterprise search, but the right starting point for any Australian Shopify store that has not yet outgrown native Shopify search.

Bottom line

Constructor is a genuinely excellent enterprise search and discovery platform. If you are a large Australian retailer with a custom or enterprise e-commerce platform, meaningful monthly search volume (hundreds of thousands of search queries per month), and a budget that supports a six-figure annual investment, Constructor should be on your evaluation shortlist. The Gartner recognition, the scale of deployments, and the revenue-outcome orientation are all real.

For everyone else — which is the vast majority of Australian online retailers — the honest answer is that this tool is not built for you. It is enterprise infrastructure, priced and scoped as such. Start with Shopify Magic if you are on Shopify, or with the native search on your existing platform, and revisit Constructor when your scale justifies it.

Score

Frequently asked questions

What does Constructor do?

Constructor is an AI-powered product search and discovery platform for e-commerce retailers. It replaces or augments native site search with AI that optimises for revenue and conversion rather than simple relevance. The platform includes search, autosuggest, browse page personalisation, product recommendations, AI shopping agents, and a merchandising intelligence agent. It was named a Gartner Magic Quadrant Leader in Search and Product Discovery in 2026.

Does Constructor work with Shopify stores?

Constructor is designed for enterprise retailers on custom or enterprise e-commerce platforms. It is not a Shopify App Store application and is not a standard integration for Shopify-based stores. Australian Shopify merchants should look at Shopify Magic (free, built-in) or other Shopify-native tools for search and discovery improvement.

How much does Constructor cost?

Constructor does not publish pricing. It is a sales-led enterprise product with usage-based pricing scoped per customer. Independent third-party sources suggest annual contracts in the range of six figures USD. Australian retailers should contact Constructor directly for pricing. AUD pricing is not documented; expect USD contracts with FX risk.

Sources

• Constructor website — product suite, security and compliance: https://constructor.com/ — verified 26 June 2026 — Wayback 3 September 2026
• Constructor Security and Compliance page: https://constructor.com/security-and-compliance — verified 26 June 2026 — Wayback 25 July 2026
• Constructor FY26 growth and 322 billion interactions press release: https://www.prnewswire.com/news-releases/constructor-reports-82-customer-growth-in-fy26-with-its-platform-powering-322-billion-shopping-interactions-302711286.html — PR Newswire, 2026
• Constructor Merchant Intelligence Agent (MIA) launch: https://www.prnewswire.com/news-releases/constructor-unveils-merchant-intelligence-agent-mia-bringing-instant-insight-and-faster-action-to-ecommerce-merchandising-302723004.html — PR Newswire, 2026 — Wayback 3 July 2026
• Gartner Magic Quadrant for Search and Product Discovery 2026: https://constructor.com/ (referenced on homepage) — independent Gartner research, 2026 — Wayback 3 September 2026
• Pricing: (no public pricing; third-party sources suggest six-figure annual contracts; verify at constructor.com)
• AUD exchange rate used: US$1 ≈ A$1.40 as of September 2026 (estimate)

The AI Directory Australia is an independent editorial publication. We don't accept payment for positive reviews. Read our Methodology and Affiliate Disclosure.