Amboras: The AI-Native Store Builder Bet
Amboras launched in May 2026 as the first AI-native ecommerce platform, a phrase the founders use literally. The three-person Y Combinator company automates storefront operations from design to A/B testing to email flows, and claims, Amboras's website reported, early merchants saw conversion-rate lifts above 80%. That claim, unverified by third parties, has forced a response: Shopify opened its commerce infrastructure to any brand via the Agentic Plan, and BigCommerce rebranded to Commerce.com around agentic tools. Merchants and the engineers who build for them now face a concrete fork: bet on an all-in-one AI platform, or stay on open, composable infrastructure.
The company emerged from Y Combinator's Spring 2026 batch, backed by primary partner Aaron Epstein. Co-founders Imad Mokadem and Amin Mokadem both studied at ETH Zurich, in mechanical engineering and computer science. Before Amboras, they sourced products, wrote ads, and tracked every metric themselves for an educational board-game brand that hit $200,000 in monthly recurring revenue, the Y Combinator launch post showed. No agency. No investors. They lived the problem they now sell a solution for: a dozen apps charging monthly, a CRO agency on retainer, and a cut of every conversion uplift on top.
The platform went live in May 2026. The pitch is all-in-one: storefront hosting, product and order management, payments (Stripe and PayPal wired day one, Apple Pay and Google Pay standard), transactional and marketing email, first-party analytics without pixels, and an AI Business Assistant embedded in every admin page. The prompt editor lets merchants change any section — hero, products, fonts, colors, animations — by describing it. Edits inherit brand tokens and preview live. Generative A/B testing spins up hero, offer, layout, and copy variants around the clock, ships them as live tests, and promotes winners automatically. The more traffic, the faster the conversion rate climbs on its own.
Amboras positions itself as a direct replacement for the Shopify app stack. Its marketing page lists "0+ Shop apps replaced," Amboras's site recorded, and bundles upsells, subscriptions, pricing optimization, SEO, and multi-storefront deployment into the base platform. For merchants running or planning multiple storefronts, the Grow tier's five-stores-for-one-subscription model creates a structural cost advantage over Shopify Basic plus the typical app subscriptions; Klaviyo alone starts at several hundred dollars a month at meaningful list sizes.
The conversion claim is the sharpest edge. Amboras states that early beta merchants saw "over 80% CVR lift on the first version Amboras puts live," Amboras's YC launch post found, a figure repeated on its Y Combinator page and its own site. Founderland.ai and aigregator.com have both cited the number with the qualifier "according to Amboras." No independent A/B study or third-party audit has been published. The company's YC application frames the problem bluntly: most stores plateau at a 2% conversion rate. "What about the other 98%?" The platform's answer is continuous, autonomous optimization: storefronts that learn from every interaction and adapt to every ideal customer profile, without designers, CRO experts, or developers waiting weeks for a single test.
The bet is that merchants will trade the flexibility of Shopify's plugin ecosystem for an opinionated, autonomous stack that compounds learning instead of stacking fees. The next move belongs to Shopify.
Shopify's Counterpunch: Opening the Catalog to Everyone
Shopify's answer arrived in two parts. First, the Universal Commerce Protocol — co-developed with Google and backed by Walmart, Target, Etsy, American Express, Mastercard, Stripe, and Visa — established an open standard for agentic checkout. Then, in March 2026, the company flipped the switch on the Agentic Plan: a global program that lets any brand, on any ecommerce platform, push its product catalog into Shopify's infrastructure and reach shoppers across ChatGPT, Microsoft Copilot, Google AI Mode, the Gemini app, and the Shop App. The storefront stays where it is. The commerce layer moves to Shopify.
The mechanics are deliberate. A merchant on BigCommerce, WooCommerce, or a custom stack signs up for the Agentic Plan, feeds products into Shopify Catalog, and those listings become discoverable in AI channels without a Shopify online store. No monthly fee. Transaction fees match standard card rates (2.9% plus 30 cents) with Shopify taking no extra cut on AI checkouts. OpenAI charges its own 4% on ChatGPT Checkout sales; Google AI Mode and Gemini currently add nothing. Orders flow back into the merchant's existing admin with ChatGPT referral attribution intact. The merchant remains the merchant of record. Customer relationships and data stay with the brand.
Shopify Catalog does the heavy lifting. It structures every product into a universal taxonomy, infers missing attributes, consolidates variants, syncs price and inventory in real time, and distributes the enriched record to partner AI channels. When AI platforms scrape product data themselves, they get HTML soup: wrong currency, last week's sold-out variant, the SEO description from 2019. Catalog replaces that with a clean, machine-readable feed. The same enrichment that helps products surface in AI also lifts organic search discovery, Shopify says.
The numbers explain the urgency. U.S. consumers ask ChatGPT over 84 million shopping-related questions per week. ChatGPT counts 700 million weekly active users. Shopping-related searches on generative AI platforms grew 4,700% between 2024 and 2025. In Q1 2026, AI-driven traffic to Shopify stores grew eight times year over year; orders from AI-powered searches jumped nearly 13 times. By Q2, AI-referred sessions were up 197% and orders tripled. New buyers arriving through AI channels convert at nearly twice the rate of other channels. ChatGPT-referred traffic converts at 11.4% versus 5.3% for organic search: a pattern Shopify calls "buyer journey compression," where AI collapses discovery and consideration into a single conversation and delivers high-intent buyers straight to product pages.
The Agentic Plan is a sidecar model. Average platform migration costs run $25,000 to $500,000, and 76% of B2B sellers are actively looking to switch but cite cost as the top barrier. By decoupling the catalog from the storefront, Shopify removes the switching cost. A brand on Adobe Commerce or Salesforce Commerce Cloud can plug into AI channels tomorrow without replatforming. Over 1 million Shopify merchants — including Glossier, SKIMS, Spanx, and Vuori — were already available on ChatGPT's instant checkout as of March. Thousands more are selling in Microsoft Copilot. Select brands have started appearing in Google AI Mode and the Gemini app via UCP.
This move rewrites Shopify's moat. The platform's historic strength was the all-in-one bundle: storefront, checkout, payments, apps, ecosystem. The Agentic Plan unbundles the commerce layer — catalog, checkout, payments, fraud protection, tax, shipping — and offers it as infrastructure to competitors' customers. Shopify becomes the rails, not just the destination. President Harley Finkelstein called AI the "new frontier" for online retail when OpenAI launched Instant Checkout last fall. Six months later, OpenAI killed Instant Checkout in favor of dedicated retailer apps inside ChatGPT. Shopify's response was to open the catalog to everyone, betting that the commerce layer — not the storefront — is the defensible asset in an agentic world.
Trade-offs exist. The rollout defaulted to opt-in: nearly all eligible Shopify stores were activated for ChatGPT discovery referrals on March 24, 2026, with updated supplemental terms taking effect May 25. No global "turn off Agentic Storefronts" button exists. No fine-grained product-level opt-out. No control over how the AI summarizes a product or which competitor appears next to it in the chat. Direct checkout carves out subscriptions, bundles, customizable products, digital goods, B2B-only items, and local delivery or pickup. Custom pixels don't fire in partner checkout; Google Analytics events fire server-side only. Required account sign-in breaks the guest-checkout model. Merchants with messy catalog data (color in the variant name but no per-variant image, 90 variants crammed into one product, tags full of internal abbreviations) see visible, ugly results in AI channels.
Walmart found conversion rates three times lower for products sold directly in ChatGPT versus rerouting users to its own site for checkout. Etsy called ChatGPT a valuable discovery channel but said purchase volume from Instant Checkout was relatively low. The technology is nascent. But the traffic curve is steep, and Shopify has positioned its catalog as the default data pipeline for AI commerce. UCP will power checkout in Meta experiences next. The geography will expand over the next six to twelve months. The "zero transaction fee" line is a market-share play; once AI channels are entrenched, that clause is exactly the sort that quietly changes in a future revision.
For them, the choice sharpens: stay on your current stack and plug into Shopify's AI distribution, or migrate to an AI-native platform like Amboras that bakes the agentic layer into the storefront itself. Shopify is betting the infrastructure wins.
Commerce.com: The Composable Third Pole
BigCommerce didn't just rename itself. It restructured the entire proposition. In August 2025, the Austin-based public company (Nasdaq: BIGC since its 2020 IPO that raised $216 million) became Commerce, ticker CMRC. The move was deliberate. "When you're named after the platform, it's hard to not think of that," CEO Travis Hess told the Austin American-Statesman in October 2025. "When you're named as a steakhouse, it's hard to imagine you've got other things on the menu." Hess, who joined as president in May 2024 and took the CEO role five months later, described the six-month integration of BigCommerce, Feedonomics, and Makeswift as "building a car as you drive it down I-35 at 90 miles-an-hour."
The rebrand signals the third pole in this platform war. Shopify is opening its catalog. Amboras is selling an all-in-one AI store. Commerce is betting on an open, intelligent ecosystem (three acquired products wired together) that lets merchants compose their own stack while the company handles the agentic layer underneath.
At Commerce Live 2026 in Chicago (the conference formerly known as BigSummit), the company unveiled a slate of updates that make the strategy concrete. The agentic commerce capabilities, announced April 30, 2026, center on five pillars: enriched, agent-ready product data through Feedonomics; distribution across AI-driven discovery surfaces including ChatGPT, Gemini, Microsoft Copilot, Perplexity, PayPal, and Stripe; agent-enabled checkout experiences that let transactions occur while merchants retain customer ownership; conversational search and AI-assisted shopping on the merchant's own storefront; and BigCommerce Companion, an AI assistant that answers questions, surfaces insights, and handles operations.
"Commerce as an industry is entering a phase where the distance between discovery and transaction is shrinking rapidly," said Sharon Gee, senior vice president of product for AI at Commerce. "Our focus is on making sure merchants can reap the benefits of agentic commerce while maintaining control of their data, their customers and their operations."
The platform updates extend beyond agentic features. Checkout load times dropped by one full second: a measurable conversion lever. Multi-language capabilities arrived with translation APIs, localized URLs and sitemaps, and native translation management. Makeswift added batch translation jobs so teams can localize entire sites at scale, not page by page, with full review control before publish. B2B tooling expanded with AI-driven purchase order automation, event-driven webhooks, and cascading price lists. Feedonomics Surface, the self-service feed management product, added Microsoft Ads, TikTok, and Pinterest to its existing Google and Meta channels. Native Hosting for Catalyst, a headless storefront deployment on Cloudflare managed via CLI with log retention and custom domains at no extra cost, enters open beta summer 2026.
Hess sees the shift in buyer behavior as the accelerant. "Websites are not being optimized for humans anymore. They're being optimized for agents," he said. "Organic traffic for a lot of organizations has dropped 20-30% because all the eyeballs are going to the answer engines." Those answer engines (OpenAI, Perplexity, and others) lack the merchant centers, governance, and syndicated-data infrastructure that Google and Microsoft have built over decades. Commerce's play is to be the connective tissue: Feedonomics structures the catalog for agent consumption; BigCommerce handles the transaction; Makeswift renders the experience.
Early signals suggest traction. Existing BigCommerce small businesses using Feedonomics Surface saw approximately 24 percentage points greater year-over-year gross merchandise value growth in November 2025 compared to peers. A 2023 Forrester Total Economic Impact study commissioned by the company cited 211% ROI within eight months, $700,000-plus saved by retiring legacy solutions, and 90% developer savings by year three. Joe Sharplin, head of ecommerce at AS Colour, said automating purchase orders and improving pricing management "has been a significant unlock for our business. We're reducing manual work, minimizing errors and delivering a more seamless buying experience for our customers."
Vipul Shah, chief product officer, framed the releases as "a stronger platform and new offerings and capabilities designed to help merchants turn innovation into measurable growth." The roadmap includes unifying B2B and B2C experiences under a single control panel and one storefront API layer.
Commerce's bet differs from both Shopify's and Amboras's. Shopify is pursuing this strategy, letting any brand plug into its commerce backbone. Amboras is collapsing the stack into a single AI-native platform. Commerce is arguing for composable infrastructure with an agentic intelligence layer on top: open APIs, 600-plus integrations (ERP, CRM, PIM, WMS), 20-plus embedded payment providers without payment fees, and 130-plus payment solutions total. The merchant who wants to build custom AI solutions on an API-first architecture gets that flexibility; the merchant who wants turnkey agentic distribution gets Feedonomics; the merchant who wants visual editing without developers gets Makeswift.
Hess expects the disorientation to compress fast. "I think six months from now, this will maybe be old hat. People are going to be used to all these acronyms and all these terms." The company's job, he said, is readiness: making sure merchants can be discovered, transact, and retain control when the agent becomes the shopper.
The Merchant's Fork: TCO, Conversion Claims, and What to Trust
The decision facing merchants isn't abstract. It's that fork: bet on such a platform that bundles storefront, checkout, email, reviews, A/B testing, and an AI assistant into a single subscription, or stay on composable infrastructure where you assemble best-of-breed pieces yourself. Amboras prices the bundle at $39/month for Basic (one store, one user, 5 million AI tokens), $105/month for Grow (up to five stores, five seats, autonomous A/B testing, multi-currency), and $399/month for Advanced (20 stores, dedicated CSM, private Slack, custom AI fine-tuning). Enterprise is custom. No platform transaction fees: Stripe processing passes through at standard rates (2.9% + $0.30 domestic). A 30-day launch guarantee promises a live store in 24 hours and a full refund if no sales arrive in the first month.
| Plan | Monthly | Stores | Seats | Key Inclusions |
|---|---|---|---|---|
| Amboras Basic | $39 | 1 | 1 | AI designer, image enhancement, email automation, native reviews, 150+ payment methods, 5M AI tokens |
| Amboras Grow | $105 | 5 | 5 | Autonomous A/B testing, advanced analytics, multi-currency, post-purchase upsell, cart recovery |
| Amboras Advanced | $399 | 20 | 20 | Dedicated CSM, private Slack, AI fine-tuning, white-label storefront |
| Shopify Basic | $39* | 1 | 2 | Platform only — apps, email, A/B testing, reviews extra |
*Shopify Basic at $39/month excludes app fees. A well-configured store commonly adds $50–$300/month in app costs. It alone starts at several hundred dollars at meaningful list sizes. Each additional store requires a separate Shopify plan. Amboras Grow's model yields a structural cost advantage that diverges quickly as store count increases.
Shopify's commissioned TCO study (December 2023–January 2024, 250 enterprises, $10M+ revenue, US/Canada) claims 33% better total cost of ownership than competitors, with 23% better platform costs, 19% better operating costs, and 33% better implementation costs. Adobe platform and implementation costs run 42% higher per that study; BigCommerce implementation 88% higher; Salesforce operating 6% higher; WooCommerce operating 41% higher. The research was commissioned by Shopify from an independent consulting firm — not an independent benchmark.
Amboras's conversion claims come from its own beta. Early merchants reported CVR lifts above 80%, Amboras's data shows, with one documented case moving from 1% to 3.2% in 10 days. No third-party verification exists. The median Shopify store conversion rate in 2026 sits around 1.4%; the average is 2.1%, pulled up by high-converting categories (food, beauty, supplements) and optimized social-proof stacks. AI-driven traffic converts 42% better per scattered 2026 studies, though product pages score just 66%. Shopify says Shop Pay recognizes 100 million+ users and accesses 150 million+ shoppers, delivering "significantly higher conversion than standard checkout" (again, Shopify's own figure).
The engineering calculus differs by team. If you have in-house engineers or a dedicated Shopify operator, you extract maximum value from 8,000+ apps and open APIs. If your team is lean and operational overhead is the constraint, Amboras's AI automation targets that directly. Brand precision (typography, animation, checkout flow) still favors Shopify's theme ecosystem and Liquid control. Amboras's prompt-driven design system generates or modifies entire storefronts from descriptions like "minimal layout in our brand colors" or "video and customer reviews above the fold," but you're locked to their rendering layer. Cross-border remains Shopify's moat: Shopify Markets, local payment methods, region-specific fulfillment, and a global partner ecosystem are battle-tested. Amboras's documented user base and case studies concentrate in English-speaking markets as of mid-2026.
Platform risk is real. Checkout reliability, edge cases, payout operations, and integration depth are exactly what Shopify spent 15 years hardening. Amboras integrates ShipBob, ShipMonk, ShipHero, Easyship, and says most major 3PLs connect via webhooks, but no full warehouse management system sits underneath. Order splitting across multiple fulfillment locations lacks clear documentation. Migration off a proprietary platform is a project. Amboras offers free migration from Shopify (products, customers, orders, SEO redirects, 48-hour typical completion) and an A/B comparison mode to run both in parallel, but the reverse path is unproven at scale.
The safe answer for established volume remains Shopify underneath with AI layered on top: tools like Clyro that output real Liquid themes you own and export. The compelling answer for a first store, a test market, or a new product line is Amboras: prompt the store into existence, let the AI run A/B tests and email flows, and see what happens when real traffic hits the checkout. Just don't move the $10M business until you've stress-tested the edge cases yourself.
Why the Market Is Moving Now
The agentic commerce layer isn't a niche: it's a $5.7 billion market that Grand View Research valued at 2025 close, projecting $7.7 billion for 2026 and $65.5 billion by 2033 on a 35.7% CAGR. The broader agentic AI platform market, per Fortune Business Insights, Precedence Research, MarketsandMarkets, and Deloitte, sizes the standalone market at $7–8.5 billion in 2025 with 40%+ CAGRs. Gartner puts agentic AI spending at $201.9 billion in 2026 alone, overtaking chatbot spend by 2027. The spread between forecasts is wide — Fortune Business Insights sees $139 billion by 2034, Precedence Research $199 billion, MarketsandMarkets $93 billion by 2032 — but the vector is unanimous: this layer is absorbing capital faster than any other software category.
Consumer behavior is the accelerant. EMARKETER's December 2025 forecast puts AI platforms at $20.57 billion of U.S. retail ecommerce in 2026 (roughly 1.5% of total), but the growth rates underneath that sliver are violent. Adobe Digital Insights tracked AI-referred traffic across 130 North American retailers: Q1 2026 grew 393% year over year; March alone hit 269%; May 2026, without a seasonal peak, set an all-time high at 1,324% above October 2024 when tracking began. Conversion flipped in a single year. March 2025: AI referrals converted 38% worse than non-AI traffic. March 2026: 42% better. May 2026: 54% better. Salesforce measured $67 billion in global Cyber Week 2025 sales influenced by AI and agents, 20% of all orders, up from $60 billion in 2024.
The trust gap is the brake. Axis Intelligence's AI Commerce Trust Gap Index, derived from YouGov and Checkout.com 2026 data, puts U.S. consumer authorization for autonomous purchasing at 14%, 51 percentage points below the 65% who trust AI to compare prices. Gen Z and millennials reach 29–30% trust for autonomous checkout; the general population doesn't. Checkout.com's June 2026 survey of U.S., UK, and Australian consumers found 33% expect at least 10% of purchases to be AI-driven within a year, but 24% say they'll never delegate, and 27% trust no organization to operate an agent. The mechanics consumers demand are specific: spending caps (30%), instant revocation (29%), easy cancellation (28%). Average authorized spend per autonomous transaction without re-authorization: £177 ($224), below the £200 ($253) merchants expected.
B2B is further along in intent than execution. Forrester's January 2026 survey of 18,000 global buyers: 94% use generative AI as a core research tool. Gartner's March 2026 survey of 646 buyers: 67% prefer rep-free routine procurement. But Boston Consulting Group finds only 18% of B2B organizations at "advanced" AI commerce maturity, and Gartner/Mirakl estimate under 50% have agent-ready real-time pricing APIs. Gartner's November 2025 forecast: 90% of B2B purchases intermediated by agents by 2028, routing more than $15 trillion (roughly half of U.S. GDP). The protocol stack is fragmented. OpenAI and Stripe's Agentic Commerce Protocol (September 2025) and Google's Universal Commerce Protocol (January 2026, backed by Amazon, Meta, Microsoft, Salesforce, Stripe) are not interoperable as of mid-2026. Most B2B suppliers have implemented neither.
Fraud is scaling with autonomy. Darwinium's 2026 research: 97% of organizations reported increased AI-facilitated attacks, average direct loss $4.5 million annually. Visa flagged a 25% spike in malicious bot transactions over six months, 40% surge in the U.S., driven by agent-automated account takeover, inventory manipulation, checkout abuse. Accenture: 78% of financial institutions expect agent-linked fraud to rise; 87% of CTOs and payments heads call trust the primary adoption barrier. The emerging "Know Your Agent" framework (verification of agent identity, permissioned scope, behavioral parameters) is the infrastructure response. McKinsey cites KYA as foundational.
Regionally, North America holds 38.2% of the agentic commerce market (Grand View, 2025). Asia-Pacific runs at 34.88% CAGR through 2031 (Mordor Intelligence); 48% of Indian retailers pilot generative AI; Chinese AI avatars have generated $7 million in seven hours on live commerce streams. Europe's EU AI Act becomes fully applicable August 2026, imposing transparency and human-oversight requirements on high-risk commerce agents. IKEA's governance framework is the cited blueprint. The UK: only 15% of top 100 retailers say payment systems are agent-ready, though 49% invest actively; 58% believe agents have reached their platforms; measured agent transaction share is 3%.
The platforms — Amboras, Shopify, Commerce.com — are fighting for the commerce layer because the traffic, the conversion lift, and the capital formation are all moving there at once. The Mokadem brothers built their board-game brand by hand because no platform automated the work. Now three platforms offer three different automations. The fork is no longer theoretical.
Working in AI? Zero G Talent tracks the openings: see every open Databricks role, browse AI jobs, openings at Anthropic and Harvey AI, and the people building the field.