AI agents helped drive $67B in Cyber Week sales - here's what sales teams can do next
Salesforce reported a strong holiday run: global Cyber Week spending reached about $336 billion. AI-powered agents influenced an estimated $67 billion of that by serving up relevant recommendations and fast conversational support.
Retailers using Agentforce 360 saw sales grow faster than peers. Funko is one example. Josh Smiley put it plainly: "agentic experiences are paramount to our success." He added, "Leveraging the Agentforce 360, we have modernized our customer support model with AI-driven replies to drive higher productivity and operational efficiency. And by going live with our shopper agent, powered by Agentforce Commerce and conversational AI, we can secure new customers with a seamless, personalized shopping experience."
Why this matters for sales leaders
- Always-on reps: AI agents answer pre-purchase questions, handle objections, and keep carts moving.
- More revenue per visit: Better product guidance lifts conversion and average order value.
- Lower costs: Automated support deflects tickets and shortens resolution times.
- Clean data: Every chat becomes structured intent, content gap, and buyer signal data.
A practical Agentforce 360 playbook
- Pre-sales guidance: Use a shopper agent for product fit, comparisons, and bundles.
- Cart rescue: Trigger proactive assistance on hesitation signals (idle time, exit intent).
- Post-purchase support: Automate "where is my order," returns, and exchanges with clear handoffs.
- Routing rules: Send high-intent chats to live reps instantly; keep FAQs with the agent.
- Offer logic: Surface promos and financing only when it improves margin and conversion.
- Feedback loop: Tag unanswered questions, update content weekly, and retrain prompts.
- A/B test flows: Test greetings, product questions, and recommendation prompts; keep winners.
- Guardrails: Enforce compliance, refund limits, and escalation policies inside the agent.
Metrics to track this quarter
- Conversion rate uplift on sessions with agent vs. without
- Average order value and units per transaction
- Cart recovery rate after agent interaction
- Time to first response and time to resolution
- Deflection rate and containment rate (no human needed)
- Revenue attributed to agent interactions
- CSAT/NPS on agent vs. human-only support
- Return rate and margin impact on orders influenced by the agent
30-day rollout plan
- Week 1: Pick 3-5 high-traffic pages. List top 50 buyer questions. Write crisp answers tied to your catalog and policies.
- Week 2: Connect product data, inventory, pricing, CRM, and order systems. Define routing and escalation.
- Week 3: Soft launch to 10-20% of traffic. Track conversion, AOV, CSAT, and deflection. Fix content gaps fast.
- Week 4: Scale to 50-100%. Turn on cart rescue prompts. Start A/B tests. Review guardrails and edge cases weekly.
Team enablement and next steps
Sales, RevOps, and Support should co-own agent goals. Agree on shared KPIs, incentives, and a weekly review to tune prompts, offers, and routing based on real conversations.
If you're evaluating Agentforce, start with a narrow use case and measurable goals, then expand. For product details, see Salesforce's overview of Agentforce here: Agentforce.
Level up your sales skills with AI
Want a structured path to upskill your team on AI for sales and ecommerce? Explore curated training by role here: Complete AI Training - Courses by Job.
Your membership also unlocks:
