Agentic Commerce Has Arrived
The signal in the noise: AI became a real traffic channel
Adobe reports that traffic from AI sources (LLMs) to retail sites rose 693.4% YoY this holiday season. That’s not a rounding error – it’s a new acquisition lane forming in real time.
And early indicators suggest this traffic can be unusually “ready to buy.” On Black Friday, Adobe found shoppers who arrived at U.S. retail sites via an AI chat service were 38% more likely to purchase than visitors from non-AI sources.
The economics: value-hunting + mobile-first + flexible payments
The holiday story also reinforced the fundamentals driving conversion right now:
- Mobile is now the dominant channel: 56.4% of online revenue share, with $145.2B in mobile spend (+10.7% YoY).
- BNPL is mainstream: $20.0B in BNPL spend (+9.8% YoY), including $16.4B of BNPL spend on mobile.
- Discounting stayed aggressive: electronics discounts peaked at 30.9% off listed price (vs. 30.1% in 2024).
- The season also packed more “mega-days”: 25 days crossed $4B+ in online spend (up from 18 days last year).
So what is “agentic commerce,” really?
“Agentic commerce” is the step beyond chatbots and recommendations: AI agents that can anticipate needs, navigate options, and execute transactions – not just answer questions. McKinsey describes it as shopping powered by agents acting on our behalf, moving from discovery to action in multi-step chains.
In practical terms: shoppers are starting to ask (“best winter work boots under $200, wide size, waterproof”) and letting an AI narrow choices, compare, and tee up the purchase – sometimes without ever browsing a category page.
The platforms are racing to own the “buy” moment
This past season, major retailers and tech platforms rolled out assistants and agent-like experiences designed to do more than chat – think personalized recommendations, price tracking, and increasingly, purchase execution. AP highlighted new tools from the likes of Amazon, Walmart, Google, and AI platforms (including ChatGPT), plus agent workflows like calling local stores to check inventory and early “buy for me” style experiences.
What wins in agentic commerce (and what breaks)
As agents become the new “gatekeepers,” the competitive unit shifts from clicks to eligibility – whether your products are the best, most reliable answer for a specific intent.
Brands that win will be the easiest for agents to understand and trust:
- Clean, complete product data (attributes, compatibility, variants, fitment, specs)
- Decision-ready PDPs (clear shipping/returns, FAQs, comparison cues, real photos, honest constraints)
- Real-time signals (availability, lead times, pricing integrity, promotions)
- Structured content that an AI can summarize without guessing
Brands that struggle:
- Thin product pages, incomplete specs, messy variants
- “Hidden” policies (returns/shipping only in footers or PDFs)
- Inconsistent inventory and pricing (agents will route around uncertainty)
A simple January takeaway for you
Agentic commerce isn’t a someday trend – it showed up in holiday performance data this past year. The shift is already underway: mobile is the default, flexible payments are normal, and AI is becoming a measurable acquisition channel with strong buying intent.
The brands that act now won’t just optimize for the next campaign – they’ll be building for a world where the first interaction isn’t with a shopper… it’s with the shopper’s agent.
Sources
Industry Update: Shopify Winter ’25 Editions – the “boring” upgrades that make B2B run better
Shopify’s Winter ’25 Edition (aka “The Boring Edition”) shipped 150+ improvements – and the most meaningful ones aren’t flashy storefront tweaks. They’re operational upgrades that reduce friction for B2B workflows, sales-assisted ordering, and checkout performance. Shopify
On the speed side, Shopify reports cart loading speeds up to 50% faster, and accelerated checkout buttons up to 58.8% faster – the kind of improvements that quietly lift conversion while lowering support tickets.
For B2B teams, the headline is draft orders and accounts getting “real”: Shopify upgraded draft orders so you can lock quoted pricing and offer bundles, and (on Plus) collect deposits and charge vaulted cards – bringing quoting closer to an actual order flow. Customer accounts can now be extended via apps to support quoting workflows and downloadable PDF invoices, plus self-serve returns for B2B – a direct win for customer experience and back-office efficiency.
What to practically do next:
- Revisit your draft order → checkout path (pricing rules, bundles, deposits).
- Use customer account extensions to enable quotes/invoices/returns without custom builds.
- Benchmark checkout speed and prioritize the bottlenecks that impact revenue.
Ready to unlock the power of your business data? In this beginner-friendly video, we’ll show you how to build a Data Lake that merges your ERP, CRM, and website analytics into one powerful hub of insights.
- Learn what a data lake really is
- See how ELT (Extract, Load, Transform) works in practice
- Discover the tools (AWS Athena, Azure Synapse, etc.) that make querying easy
- Understand why data lakes bring scalability, flexibility, cost efficiency, and data democratization to your busines
By the end, you’ll know exactly why a data lake might be your secret weapon – and how to take your first steps.