Web service

Assessment of page views over a period distinguishes user devices, network and hosting. Comparisons between versions rely on the same user journey and delivered function.

Prepare the data

Input	Preferred measurement	Use in the calculation
Page views and journeys	Analytics over a representative period	Service units
Active time	Observed user journey	Device allocation
Transferred bytes	Browser measurements or logs, cache included	Network activity
Hosting	Energy and allocated manufacture over the period	Server contribution independent of reading time

Device profiles can describe a device type or a mobile/computer mix. For audiences spanning different devices or countries, the total is the sum of separately calculated group results. Transfers should include observed third-party resources and requests.

Device energy is hours × power / 1,000. Component-based profiles use manufacture already allocated per second. Custom mode allocates per-device manufacture over calendar lifetime. Both conventions are detailed below.

The D4B network model uses per-GB·km factors and estimated segment distances. This page’s illustration accepts a factor already expressed for the route in kg CO₂e/GB, including manufacture and operation. Without a compatible factor, traffic remains known and network impact remains unassessed. This input simplification does not replace the distance-based model.

Hosting uses allocated IT energy and its PUE. That allocation must account for reserved resources while idle.

Detailed method and special cases

Build a representative journey

Pages form groups by function and behavior: reading, search, transaction, video or dynamic application. Each group is described by views, active duration, actual transferred bytes and device distribution. Observed cold and warm cache proportions determine their representation in the sample. One home page does not represent an entire service.

Web service boundary

User devices

Allocated manufacture and electricity during use.

Networks

Access, transit and CDN delivery within the selected boundary.

Hosting and third parties

Physical resources allocated to applications, VMs and third-party services.

A sum for each indicator, for the same function and period.

A VM is a share of physical hardware. A third-party service already included in a contribution is not added again.

Allocate the device to usage

For $N$ views lasting $t$ seconds, cumulative time is $H=Nt/3600$. A D4B average manufacturing factor $b$ is already in kg CO₂e/s: $I_{manufacture}=Nt\,b$. Electricity is $E=H W/1000$, with $W$ in watts.

The mobile/computer mix weights factors: $b=p b_m+(1-p)b_o$ and $W=p W_m+(1-p)W_o$. $p$ is the mobile share of active time. It equals view share only with equal average durations. Otherwise assess groups separately and sum them.

Component-based profiles

These profiles come from illustrative configurations, not measured averages. Manufacturing is calculated by component, then allocated over calendar lifetime. Parameters and limitations appear in the register.¹

Profile	Scenario power (W)	Allocated manufacturing (kg CO₂e/h)
Computer: 60% laptop, 40% desktop	26.78	0.00465
Mobile: 95% smartphone, 5% tablet	4.85	0.00329
Laptop (illustrative configuration)	17.62	0.00338
Desktop with display (illustrative configuration)	40.53	0.00654
Smartphone (illustrative configuration)	4.73	0.00330
Tablet (illustrative configuration)	7.00	0.00297

Example: 100,000 views of 60 seconds, with 50% mobile time, give 26.36 kWh and 6.61 kg CO₂e of allocated device manufacturing. The electricity factor is then applied to energy.

Custom mode uses device manufacture $M$ and lifetime $L$: $I_{manufacture}=H M/(8760L)$. This replaces the per-second factor; it does not add to it. Electricity still uses active time, not continuous consumption.

Network, cache, hosting and third parties

Traffic is $G=N\times MB_{view}/1000$. Fixed/cellular share is independent of device type: a phone on Wi-Fi uses the fixed path. Network impact is $G_f f_f+G_c f_c$ when factors cover the selected paths. A segment factor is not a full-path factor; estimated distances belong to the per-GB·km model and are not applied again to a route factor.

Browser caching reduces received bytes; a CDN also changes paths and uses storage and servers. Measured cached bytes include the cache effect once. Hosting uses allocated energy and manufacture over the period. Smaller pages do not demonstrate proportional reductions in those contributions.

For external video, APIs or payments, the boundary covers the delivered function and the provider inventory. Independently calculated impacts can only be added when the same networks, servers and devices are not already included.

Extrapolate from samples to a period

Journey groups are summed: $I_{service}=\sum_j I_j+I_{shared}$. Shared resources are allocated once. Measurements remain distinct from audience projections. Per-view results require equivalent content and function for comparison; an arbitrary score out of 100 adds no physical precision.

Limits and details

Prefilled parameters are an exercise. Historical audience prediction, equating reading time with CPU time, and the 0–100 score are not part of this calculation. Their historical versions remain available in v1.

The Native applications case includes native-app installations and updates. External inference falls within Generative AI, with its hosting counted once in the boundary.

The result is framed by boundary and functional unit, allocation and double counting.

Electricity factors come from the D4B 2025 snapshot². Prefilled activity and hosting values are exercise assumptions. Factors and conventions are available below the simulator.

Interactive example

This example applies the calculations and limitations described above.

Ex: 100,000 page views transferring 2 MB each represent 200 GB. At 60 active seconds per view, device time totals 1,666.67 hours. This is not server CPU time.

Example inputs

Page views over the periodActive time per view (s)Transferred data per view, cache included (MB)User deviceMobile / computer mixComputer: 60% laptop, 40% desktopMobile: 95% smartphone, 5% tabletLaptop (illustrative configuration)Desktop with display (illustrative configuration)Smartphone (illustrative configuration)Tablet (illustrative configuration)Custom parametersUsage time split: mobile / computer

Mobile 50% · Computer 50%

User countryFranceGermanyUnited StatesUnited KingdomChinaIndiaPolandSwedenNorwayJapanBrazilAustraliaTaiwanSouth KoreaIrelandHosting countryFranceGermanyUnited StatesUnited KingdomChinaIndiaPolandSwedenNorwayJapanBrazilAustraliaTaiwanSouth KoreaIreland

Adjust assumptions

Allocated IT energy over the period (kWh)Allocated hosting manufacture over the period (kg CO₂e)Site PUECellular traffic share (%)Fixed network factor, complete path (kg CO₂e/GB)Cellular network factor, complete path (kg CO₂e/GB)

Calculated subtotal31.28 kg CO₂e

0.0003128 kg CO₂e / page view

Not assessed: Missing fixed network factor ; Missing cellular network factor ; Unlisted third-party services and building manufacture.

Result and activity breakdown

Contribution	kg CO₂e
Device: allocated manufacture	6.611
Device: electricity · 26.36 kWh	0.8416
Fixed network	Not assessed
Cellular network	Not assessed
Hosting: allocated manufacture	20
Hosting: electricity · 120 kWh	3.831

• Cumulative device time : 1,667 h
• Transfers : 200 GB

Prefilled activity and hosting values are example assumptions. The result covers climate impact; unassessed contributions remain identified.

Simulator factors and conventions

Electricity: D4B 2025 national averages. Use country does not change manufacture. Network factors remain unknown until supplied; they must cover the full path, manufacture and operation included. In this illustration, the distance from the D4B per-GB·km model is incorporated upstream into the per-GB route factor.

Devices: illustrative parametric configurations. Profiles are calculated from components and include their display. Computer and mobile groups assume 60% laptops and 95% smartphones, weighted by active time. These are fictional compositions, not measured averages. Manufacturing per second is already allocated over lifetime.

Current profile : 15.82 W · 0.003967 kg CO₂e/h allocated manufacture.

Manufacturing is allocated over calendar lifetime (8,760 h/year). Custom inputs replace the calculated profile; they are not added to it.

Footnotes

1. D4B convention, illustrative configurations d4b-components-illustration-2026.10.19; historical CPU, RAM and SSD coefficients documented by Boavizta, other coefficients labelled as assumptions in the register. This source does not validate the DDR5 packaging factor 1.1. ↩

2. Digital4Better, D4B 2025 snapshot. Snapshot 77cae4a, 2025 data; reviewed 16 October 2026. ↩