01
0–15 sec
Capture context
The dealer uploads a site photo and enters dimensions, budget, and intent instead of explaining the job across multiple calls.
InputPhoto + dimensions + budget
OutputStructured sales brief
← Back home
From site photo to order in 120 seconds.
ADG Sale AI adds a natural interface over SAP/CRM so 2,000 dealers can find products, configure an option, quote, and order while the customer is still at the counter.
The bottleneck was not the ERP. It was the 15 minutes before a customer left the store.
Complex configurations, hard-to-recall pricing, and fragmented documents forced dealers to wait for sales support. Every minute weakened consultation at the exact point where conversion was decided.
2,000 C1/C2 dealers needing one consultation standard
5–15 min previous point-of-sale response time
< 60 sec target for each quote and order step
One sales flow. Five control points.
The agent does not replace core systems. It assembles the right data, permission, and timing into a flow a dealer can complete in front of the customer.
02
15–40 sec
Recommend a configuration
Sale AI checks catalogues, technical documents, installation constraints, and policy to narrow the right options.
InputKnowledge base + policy
OutputReasoned recommendation
03
40–60 sec
Generate the quote
Dealer-specific pricing and permissions are applied automatically; sales reviews exceptions rather than rebuilding the quote.
InputConfiguration + allowed pricing
OutputCustomer-ready quote
04
60–120 sec
Create the order
Confirmed data becomes an order and syncs to SAP/CRM, preserving the core system as the operational source of truth.
InputApproved quote
OutputSynchronized order
05
Weekly
Measure adoption
The dashboard tracks response time, usage, conversion, and the steps that still require human intervention.
InputOperating events
OutputAgent improvement KPI
AI in the middle. Accountability at both ends.
Sale AI only reads and acts within assigned permissions. Dealers retain customer context; SAP/CRM retains governed data; people approve high-impact exceptions.
Dealer & Sales
Provide context, choose the option, and approve before anything reaches the customer.
Sale AI
Understands intent, retrieves knowledge, applies policy, and orchestrates the workflow.
SAP / CRM
Owns master data, pricing, orders, permissions, and transaction history.
Evidence before promises.
The case study separates design targets, financial assumptions, and post-launch measurement instead of presenting every number in the deck as a verified result.
Operating target
70–80% Target reduction in consultation and quoting time, to be measured after rollout.Financial scenario
3–5× First-year ROI is an estimate based on dealer scale and conversion assumptions.Governance
Human + AI Exceptions, security, and final decisions remain assigned to accountable people.From the stage to the operating model.
Swipe sideways to view the next slides →
An AI Agent matters only when it reaches the transaction.
I designed Sale AI as a measurable operating layer — connecting knowledge, permissions, business actions, and a weekly improvement rhythm.