Cyber Week 2025: AI agents pushed record sales - here's what Sales and Support should do next
Cyber Week 2025 set a new high. Global sales hit $336.6B (+7% YoY), with AI agents influencing 20% of all orders and directly driving $67B through personalized recommendations and conversational support.
In the U.S., sales reached $79.6B (+5%). Black Friday brought in $79B worldwide; Cyber Monday delivered $53B. Salesforce based these figures on behavior from more than 1.5B shoppers.
The numbers that matter
- AI impact: 20% of orders influenced; $67B generated by AI-led recs and service.
- Retailers using Agentforce 360 + branded agents: grew sales 32% faster than those without.
- Customer service: AI-led conversations up 55% week over week; agent actions (address updates, returns) up 70%.
- Commerce ops: Agentforce Commerce processed 61M orders; the Order Management System handled 76% more orders; Agentforce Marketing sent 56.3B messages.
- Mobile: 70% of orders happened on mobile (global and U.S.); mobile wallets made up 27% of global orders.
- Social: 15% of global traffic came via social; TikTok up 55% YoY and drove 9% of global social traffic.
- Pricing and volume: Average selling price up 6%; order count up 2% globally and 1% in the U.S.
Why Sales should care
AI is moving from assistive to decisive. Buyers accept conversational help that's fast, specific, and frictionless - especially on mobile. Teams that pair reps with on-brand agents are growing faster and converting warmer traffic at checkout.
- Deploy an AI pre-sales assistant on PDPs and carts to answer objections, compare SKUs, and surface bundles - with clear handoff to a human.
- Make recommendations session-aware: use browse history, cart, and inventory to upsell without slowing the page.
- Track "AI-influenced revenue" as its own metric. Attribute assist credit to prompts, rec modules, and chat outcomes.
- Build offer playbooks per channel: social → limited bundles; email → replenishment; mobile → one-tap add-ons.
- Update comp plans so reps benefit from AI-assisted conversions they touch.
Why Customer Support should care
Support volume didn't spike - it shifted. AI covered routine work so humans handled edge cases. That's how costs dropped while CSAT held.
- Automate high-frequency actions: address changes, delivery updates, returns initiation, cancel/reorder, gift notes.
- Gate AI with smart guardrails: verify identity, confirm intent, summarize before commit, and require approval for risky tasks.
- Set up fast lanes for WISMO and returns with clear policies and instant refunds where risk is low.
- Instrument every bot step: containment rate, average handle time, escalation reasons, and cost per contact.
- Train agents to supervise AI: edit summaries, approve actions, tag gaps for new automations.
Mobile and social changed the funnel
With 70% of orders on mobile and wallets at 27%, checkout speed is revenue. Social now starts a meaningful chunk of sessions, with TikTok growing fast.
- Cut mobile checkout to two steps; default to wallets; prefill everything you can.
- Use link-in-bio and shoppable posts that point to the exact variant on a fast PDP.
- Run SMS or push for cart recovery on mobile; cap to two nudges with a clear incentive.
- Track UTMs by creator and creative. Kill what doesn't convert within 72 hours.
Stack notes from the report
- Agentforce 360 + branded agents: outpaced peers by 32% growth - the combo of recs + conversational checkout matters.
- Order spikes: OMS handled 76% more orders - stress-test your capacity, queues, and retry logic ahead of promos.
- Messaging at scale: 56.3B sends is a reminder: frequency needs guardrails. Tie send logic to inventory and margin.
30/60/90 plan to capture this momentum
- Next 30 days: Ship AI Q&As on top 20 SKUs, enable wallet-first checkout, automate address updates and returns.
- Days 31-60: Add session-aware bundles, SMS recovery flows, and supervised refunds for low-risk orders.
- Days 61-90: Roll out branded agents across PDP, cart, and order tracking; expand automations to exchanges and backorders; implement full AI attribution.
If you want to upskill Sales and Support teams on practical AI playbooks and tools, explore these resources:
For more context on Salesforce's commerce and AI announcements, see the Salesforce Newsroom.
Your membership also unlocks:
