Trainline · Data landscape

Three delivery options, one lake design.

Each option describes a different balance of ORG ownership and Trainline internal co-delivery across Phase 1 and Phase 2. Pick the model that fits your team's availability and appetite for hands-on involvement.

How we think about Phase 1 and Phase 2

Phase 1 - Discover and build architecture

Infrastructure, connections, Daxia installation and a small data pull to populate the lake. The goal is a working Bronze layer in Trainline's Azure tenant with a governed inbound path from AWS and Dynamics 365.

Phase 2 - Transform and serve

Full transformation and pull of data into Silver and Gold, then push into Pigment and Dynamics 365. Daxia underpins data lineage so every served number can be traced back to its raw source.

Option 1

Co-implement, then hand over

What it covers

Discovery of requirements for Silver/Gold transformation, with ORG co-implementing Phase 1 and Phase 2 alongside Trainline internal IT.

Pros

+Shared knowledge transfer: your team learns the lake, pipelines and patterns as they are built
+Lower day-one dependency on ORG once the solution is live
+Phase 1 architecture and Phase 2 transformation are designed together, so the lake is shaped by the same thinking

Resources & capacity

RoleQ1Q2Q3Q4

Enterprise Architect

ORG

100%

50%

25%

Cross-over experience in infrastructure and integration set-up. Full capacity through the first half, tapering as the build stabilises.

Data Engineer

ORG

100%

Integrations, Azure and hands-on build, with a strong finance background so they can talk directly to the finance team.

Data Analyst

ORG

100%

Strong finance experience, main link between SMEs and the build team. Writes SQL and Power BI for reporting streams.

Business SMEs

Trainline

50%

40%

30%

Business subject matter experts. Heaviest at the start while requirements are shaped, lighter once decisions are locked.

Business Testers

Trainline

15%

30%

75%

90%

Business testers. Awareness only at the start, then heavily involved from mid-project through go-live.

Project Manager (optional)

Trainline

50%

Only where the business wants a traditional scrum or waterfall cadence with formal reporting.

Effort split

ORG 60%Trainline 40%

Roughly 60% of the total effort sits with ORG and 40% with Trainline, because your internal IT builds alongside us and your SMEs stay close to every design decision.

Option 2

ORG builds, you operate

What it covers

Discovery of requirements for Silver/Gold transformation, then ORG full implementation of Phase 1 architecture and Phase 2 data transformation, pull in Silver/Gold and push into Pigment/D365, with Daxia providing data lineage.

Pros

+Fastest path to a governed, end-to-end data flow
+Single accountability for architecture, transformation and downstream delivery
+Daxia deployed in your tenant gives finance and operations direct query and lineage capability

Resources & capacity

RoleQ1Q2Q3Q4

Enterprise Architect

ORG

100%

50%

25%

Cross-over experience in infrastructure and integration set-up. Full capacity through the first half, tapering as the build stabilises.

Data Engineer

ORG

100%

Integrations, Azure and hands-on build, with a strong finance background so they can talk directly to the finance team.

Data Analyst

ORG

100%

Strong finance experience, main link between SMEs and the build team. Writes SQL and Power BI for reporting streams.

Business SMEs

Trainline

50%

45%

30%

20%

Business subject matter experts. Heaviest at the start while requirements are shaped, lighter once decisions are locked.

Business Testers

Trainline

10%

25%

70%

85%

Business testers. Awareness only at the start, then heavily involved from mid-project through go-live.

Project Manager (optional)

Trainline

50%

Only where the business wants a traditional scrum or waterfall cadence with formal reporting.

Effort split

ORG 80%Trainline 20%

Around 80% of the effort is ORG and 20% Trainline. Your people are needed for requirements, sign-off and testing rather than for build work.

Option 3

Internal discovery, ORG executes

What it covers

Business leads discovery with internal IT support, documents full Business Requirements and Technical specifications, then hands them to ORG for full implementation or co-implementation.

Pros

+Business owns the requirements before any build starts
+Clearer scope and fewer mid-phase changes
+ORG can start implementation faster because discovery and documentation are already signed off

Resources & capacity

RoleQ1Q2Q3Q4

Enterprise Architect

ORG

75%

100%

60%

25%

Cross-over experience in infrastructure and integration set-up. Full capacity through the first half, tapering as the build stabilises.

Data Engineer

ORG

40%

100%

Integrations, Azure and hands-on build, with a strong finance background so they can talk directly to the finance team.

Data Analyst

ORG

60%

100%

90%

Strong finance experience, main link between SMEs and the build team. Writes SQL and Power BI for reporting streams.

Business SMEs

Trainline

80%

60%

40%

25%

Business subject matter experts. Heaviest at the start while requirements are shaped, lighter once decisions are locked.

Business Testers

Trainline

20%

35%

75%

90%

Business testers. Awareness only at the start, then heavily involved from mid-project through go-live.

Project Manager (optional)

Trainline

50%

Only where the business wants a traditional scrum or waterfall cadence with formal reporting.

Effort split

ORG 50%Trainline 50%

Close to an even 50/50 split. Trainline carries the discovery and documentation effort up front, then ORG carries the build once specifications are signed off.

What every option assumes

These apply whichever option is chosen

Environment access

Whichever option is chosen, ORG needs read access to the AWS Athena endpoint and the D365/Synapse Link path, plus write access to the target ADLS Gen2 storage account and the Pigment/D365 push points.

Shared batch reference

A batch reference tag travels with every load so the AWS side, the lake side and the D365 side can be matched during reconciliation and lineage checks.

Part of the build, whichever option is chosen

Daxia - pipelines, querying and data lineage, deployed in your tenant

Daxia is our own React-based Azure Web App. We deploy it into Trainline's Azure tenant at no licence cost, where it runs the Bronze to Silver to Gold pipelines on cron jobs at a fraction of the usage cost of Azure Data Factory or PySpark clusters. On top of the same lakes it gives finance a direct interface: ask a question in plain English, explore the underlying tables, or pull a pre-built extract as Excel. Because Daxia sits across every layer, it also provides the data lineage that lets you retrace any served number back to the raw record it came from. Azure resource usage is billed locally in your own subscription.

Plain English querying

Finance asks questions directly of the lake, without a BI licence per person and without waiting on a report build.

Data lineage

Trace any number in Pigment or D365 back through Gold, Silver and Bronze to the original AWS or D365 batch.

In your tenant, no licence

Deployed into your Azure subscription at no cost from us; only the Azure resources it uses are billed, to you.