In e-commerce, the customer sees a simple journey: browse, add to cart, pay, receive the parcel, and return it if necessary. Behind that experience, however, sits a complex value stream involving websites, payment gateways, order management systems, warehouses, carriers, customer service, finance and inventory planning.
Value stream mapping makes this hidden system visible. It shows how products, orders, information and decisions move from the first customer interaction through delivery and, where necessary, reverse logistics. For retail operations, this end-to-end perspective is essential because a delay in one area often originates somewhere else. A slow dispatch may be caused by payment-release rules, poor slotting, inaccurate inventory or excessive batching rather than the packing team itself.
The fundamental purpose of VSM is to distinguish activities that create customer value from those that consume time, space, labour or cash without improving the customer outcome. This aligns closely with the Lean principle that value is defined by the customer’s willingness to pay.
Scope the E-Commerce Value Stream From Cart to Return
Before drawing a map, define the product family and boundaries. A practical scope could be:
- Start: Customer submits a paid order on the website or mobile app.
- End of forward flow: Delivered parcel with confirmed proof of delivery.
- Start of reverse flow: Customer submits a return request.
- End of reverse flow: Refund issued and product restocked, refurbished, liquidated or disposed of.
This scope includes both the material flow: products, totes, cartons and parcels: and the information flow that controls it.
A cross-functional mapping team should include representatives from:
- E-commerce and customer experience
- Payments and fraud prevention
- Inventory planning
- Warehouse operations
- Information technology and systems support
- Carrier management
- Customer service
- Finance and returns processing
For a structured improvement project, use a clear boundary statement such as: “Improve the standard domestic parcel order stream for stocked products, from payment confirmation to delivered order, including customer returns.” A Project Scope Boundary Calculator can help define what is inside and outside the project.

Build the Current-State Map With Real Operating Data
A current-state map should be based on observation and verified records, not assumptions. Conduct a gemba walk across the digital and physical process. Follow actual orders through the order management system, warehouse management system and carrier portals.
Capture the following at each step:
- Cycle time and wait time
- Queue size and work in process
- Batch or wave size
- First-pass yield and error rate
- Number of manual touches
- System triggers and handoffs
- Inventory held between activities
- Rework, cancellation and exception volume
Worked example: a standard e-commerce order
The following figures are an illustrative operating case for a retailer processing 12,000 standard orders per day.
| Process step | Processing time | Average waiting time | Key performance data |
|---|---|---|---|
| Payment and fraud review | 2 minutes | 38 minutes | 94.5% released automatically |
| Order release to warehouse | 1 minute | 2 hours 12 minutes | Four release waves daily |
| Picking | 7 minutes | 46 minutes | 97.2% first-pass pick accuracy |
| Packing and label creation | 4 minutes | 31 minutes | 95.8% first-pass pack accuracy |
| Dispatch staging | 3 minutes | 1 hour 25 minutes | 1,850 parcels average WIP |
| Carrier transit | : | 28 hours | 93.1% delivered within promise |
| Customer returns intake | 8 minutes | 3.6 days | 1,440 returned units awaiting inspection |
| Inspection and disposition | 14 minutes | 2.4 days | 82% restockable after inspection |
| Refund approval and payment | 3 minutes | 18 hours | Two manual approval checkpoints |
The total hands-on processing time for the forward order is approximately 17 minutes, excluding carrier transit. Yet the order spends about 33.5 hours moving through release queues, picking queues, packing queues and dispatch staging before delivery.
This produces a process cycle efficiency of approximately:
17 minutes ÷ 2,010 minutes × 100 = 0.85%
That does not mean every waiting period can be eliminated. Carrier transit, fraud controls and customer-requested delivery windows may be necessary. Nevertheless, the contrast between 17 minutes of processing and more than 33 hours of elapsed time reveals substantial improvement potential. The Process Cycle Efficiency Calculator can support this analysis.
Inventory holding also deserves attention. Assume the retailer holds:
- 420,000 units in the central fulfilment centre
- 38,000 units in pick-face locations
- 21,500 units in inbound and outbound staging
- Average inventory value of $8.4 million
- Annual carrying-cost rate of 24%
The estimated annual carrying cost is therefore:
$8.4 million × 24% = $2.016 million
A value stream map helps show where inventory is being held as a deliberate buffer and where it is simply compensating for unreliable information, replenishment delays or uneven flow.
Map the Information Flow and Its System Triggers
In e-commerce, information flow can be more influential than physical movement. A typical sequence may look like this:
- The customer submits an order through the website.
- The platform triggers payment authorization and fraud screening.
- The order management system confirms stock availability.
- A release rule sends eligible orders to the warehouse management system.
- The WMS creates a pick task based on wave, zone or priority logic.
- A scan confirms item selection and reduces available inventory.
- Packing completion triggers label creation and a shipment notification.
- Carrier acceptance triggers tracking visibility for the customer.
- Delivery confirmation triggers completion of the order.
- A return request creates a return merchandise authorization, label and expected-inbound record.
- Inspection results trigger refund, inventory adjustment and disposition.
During mapping, ask whether each trigger is automatic, delayed, duplicated or dependent on a manual decision. Common issues include orders waiting for a scheduled batch release, inventory being reserved in one system but unavailable in another, and refunds requiring approval even after objective return conditions have been met.
Identify the Eight Wastes Across the Retail Value Stream
The eight DOWNTIME wastes: Defects, Overproduction, Waiting, Non-utilised talent, Transportation, Inventory, Motion and Extra-processing: appear in different forms across order management and fulfilment.
- Defects: Incorrect product descriptions, payment errors, mis-picks, damaged cartons and wrong return classifications.
- Overproduction: Buying or replenishing stock beyond demand, or picking orders before delivery capacity is available.
- Waiting: Orders held for wave release, totes waiting for packing, parcels waiting for carrier collection and customers waiting for refunds.
- Non-utilised talent: Warehouse specialists spending their time correcting system exceptions instead of improving standard work.
- Transportation: Unnecessary movement between receiving, storage, picking, packing and dispatch zones.
- Inventory: Excess safety stock, slow-moving products, staged parcels and returned units awaiting inspection.
- Motion: Long walking routes, repeated scanning, searching for packaging and manual data entry.
- Extra-processing: Duplicate quality checks, repeated address validation, manual order re-keying and unnecessary approval loops.
The objective is not to remove every control. The objective is to understand whether a control prevents a meaningful risk or merely continues because the process has never been redesigned.

Design the Future State: Flow, Pull and Fewer Touches
A future-state map should convert improvement ideas into a coherent operating design. In this example, the retailer could:
- Replace four large release waves with smaller, demand-levelled releases every 30 minutes.
- Introduce real-time order release for low-risk, in-stock orders.
- Re-slot the top 20% of stock-keeping units that generate 65% of order volume.
- Use scan-based confirmation to eliminate manual order re-keying.
- Create standard pack cells with point-of-use packaging materials.
- Set a supermarket replenishment trigger for pick faces.
- Link delivery exceptions directly to customer-service work queues.
- Use return reason codes and inspection outcomes to trigger automated refund rules.
- Establish a daily visual management review for backlog, WIP and service failures.
These changes apply the Lean principles of continuous flow, pull, standardised work and takt alignment. If available warehouse time is 600 minutes per shift and demand is 1,200 orders per shift, the approximate takt time is:
600 ÷ 1,200 = 0.5 minutes per order
That figure does not dictate every individual task time, but it provides a reference for balancing work across picking, packing and dispatch.
Current-state versus future-state targets
| Metric | Current state | Future-state target | Improvement |
|---|---|---|---|
| Payment-to-release wait | 38 minutes | 10 minutes | 74% reduction |
| Order-release queue | 2 hours 12 minutes | 30 minutes | 77% reduction |
| Pick accuracy | 97.2% | 99.2% | +2.0 percentage points |
| Pack accuracy | 95.8% | 99.0% | +3.2 percentage points |
| Dispatch staging WIP | 1,850 parcels | 700 parcels | 62% reduction |
| Order-to-dispatch elapsed time | 5.2 hours | 2.1 hours | 60% reduction |
| On-time delivery | 93.1% | 97.0% | +3.9 percentage points |
| Return inspection wait | 2.4 days | 8 hours | 72% reduction |
| Refund completion time | 4.35 days | 1.5 days | 66% reduction |
| Inventory value held | $8.4 million | $7.1 million | $1.3 million release |
The future state should also specify ownership, performance measures, escalation rules and control-plan frequency. Without those elements, the map remains a visual exercise rather than an operating system.

Prioritise Kaizen Improvements With Evidence
Not every opportunity should be implemented simultaneously. Prioritise initiatives according to customer impact, financial benefit, risk, effort and speed of implementation.
A practical sequence is:
- Stabilise data and definitions: Agree on order cycle time, return cycle time, WIP, yield and on-time delivery calculations.
- Remove obvious information delays: Automate release triggers, status updates and exception routing.
- Reduce physical travel and batching: Re-slot products, redesign pack stations and lower wave sizes.
- Strengthen replenishment: Set pull signals using demand, takt and inventory limits.
- Redesign returns: Standardise inspection categories and define refund rules by condition and risk.
- Control the gains: Use daily visual management, weekly trend reviews and updated standard work.
A Project Prioritisation Matrix Calculator can help compare initiatives objectively, while the Business Case Financial Calculator can quantify labour savings, released working capital, avoided rework and service benefits.
Build Capability Through Lean Six Sigma Training
Value stream mapping is most powerful when it connects directly to DMAIC. During Define and Measure, the team establishes scope, customer requirements and baseline performance. During Analyse, it investigates root causes using process data, Pareto analysis, capability measures and visual evidence. During Improve, it designs the future state and tests changes. During Control, it sustains the gains through standard work, visual controls and ongoing measurement.
For professionals in retail operations, logistics, supply chain, IT or customer experience, structured Lean Six Sigma training develops the capability to turn complex customer journeys into measurable improvement projects. Practical certification is particularly valuable when it includes real-world simulations, worked examples, dummy data and complete DMAIC case studies.
Start your Lean Six Sigma certification journey today and learn how to map, analyse and improve the e-commerce value streams that shape customer experience and business performance.
Kaizen. Kai-Care. Kai-Done. ( Lean Six Sigma)








