Adobe Implementation Playbook

Plan

Plan the whole project before building

Phases 1–4 · Initiation → Design

Each phase ends with a client sign-off. Nothing moves to the next phase until its gate is passed.

Discovery sign-offDesign (SDR/SDD) sign-offBuild complete sign-offUAT/QA sign-offGo-live sign-offHypercare exit sign-off
0 of 0 done

Phase 1 · Initiation

Gate to pass: RFP response; sponsor and team named

Step 1.1

RFP (Request for Proposal)

What it is
Client document describing business need, scope, and evaluation criteria; vendors/SIs respond with approach, team, cost, timeline.
Who
Client procurement + business stakeholders
When
Pre-sales, before any statement of work
Practice
Write a one-page RFP response outline for SkyLumen Airways or Cartwell: approach, team, tools in scope, timeline.
RFP (Request for Proposal)
Screenshot RFP (Request for Proposal)Add screenshots/phases/step-1-1.png

What each tool needs in this phase

Tool1. Initiation (RFP / kickoff)
AEPConfirm licences, sponsor, sandbox count, sources in scope
Adobe AnalyticsReport suites, current implementation, stakeholders
Launch (Tags)Sites and apps in scope, current tag manager, CMP
MarketoInstance, CRM, partitions, package
RTCDPConfirm edition (B2C/B2B/B2P) and package; destinations licensed
AEMCloud Service vs. managed, sites, languages
TargetEdition, sites, testing goals
AJOPackage, channels, IPs, sandbox
CJALicensed data volume, stakeholders, questions

Interview questions tap to reveal a model answer

What is an RFP and how would you respond to one?

A client document describing need, scope and evaluation criteria. A response covers understanding of the problem, approach, team, timeline, assumptions and cost.

What do you confirm at project kickoff?

Sponsor, decision makers, licences and packages, sandboxes, systems in scope, success criteria and the communication cadence.

Why do licences matter so early?

They decide which features, data volumes and destinations are available, so scoping without them leads to promises the product cannot keep.

Phase 2 · Discovery

Gate to pass: Current State Assessment; Future State Blueprint — Discovery sign-off

Step 2.1

Current State Assessment

What it is
Audit of existing tools, data flows, tagging, consent posture, and pain points.
Who
Solution Architect / Delivery Lead
When
Discovery phase
Practice
Run Adobe Experience Platform Debugger on a real airline site and a real retail site. List the Adobe tools, tags and consent banner each uses.
Current State Assessment
Screenshot Current State AssessmentAdd screenshots/phases/step-2-1.png
Step 2.2

Future State Blueprint

What it is
Target architecture, data flows, and capability roadmap once the implementation is complete.
Who
Solution Architect
When
Discovery phase, informs SOW scope
Practice
Sketch the future-state flow for SkyLumen: website/app → Launch → Edge Network → Analytics and AEP → RTCDP audiences → AJO messages.
Future State Blueprint
Screenshot Future State BlueprintAdd screenshots/phases/step-2-2.png

What each tool needs in this phase

Tool2. Discovery (current & future state)
AEPAudit sources, identifiers, consent, data quality; draw future-state data flow
Adobe AnalyticsCurrent SDR audit, KPI inventory
Launch (Tags)Tag audit, data layer gaps, page-speed baseline
MarketoMAP/CRM audit, lifecycle, database health
RTCDPCurrent audiences (DMP/AAM), destinations, consent rules by region
AEMContent audit, current CMS, integrations
TargetCurrent implementation, past tests, traffic levels
AJOExisting ESP and journeys, deliverability, consent sources, event availability
CJAAvailable datasets, identity coverage, Analytics usage

How discovery changes by domain

DomainProfileCompliance focus
Travel & AirlinesB2C, seasonal, multi-currency/localeGDPR (EU travelers), loyalty program data, real-time pricing/inventory sensitivity
E-commerce & RetailB2C, high volumePCI-DSS (payment data), GDPR/CCPA, cookie consent, high-cardinality product catalogs
HealthcareB2C (patients) + B2B (providers, payers)HIPAA, HITRUST, consent-first data collection, no PHI in standard analytics tools without a BAA-covered environment
BFSIB2C + B2BPCI-DSS, SOX, GLBA, strict PII handling, fraud/KYC constraints on data sharing
Media & TelecomB2C subscription + B2B advertisingConsent for ad personalization, churn-sensitive data models, high-volume streaming events
Manufacturing / B2BB2B, long sales cyclesAccount-based data models, CRM/ERP integration depth, smaller but higher-value data volumes

Interview questions tap to reveal a model answer

What goes into a current-state assessment?

An audit of existing tools, tags, data flows, identities, consent and pain points, with evidence such as Debugger captures and data samples.

What is a future-state blueprint?

The target architecture: how data will flow from touchpoints through collection, AEP and applications to activation, plus a capability roadmap.

How do regulations change discovery in healthcare or banking?

You confirm what data is allowed (PHI, KYC), required agreements such as a BAA, and access controls before designing anything.

Phase 3 · SOW & plan

Gate to pass: SOW signed: scope, deliverables, timeline, price

Step 3.1

SOW (Statement of Work)

What it is
Contractual scope, deliverables, timeline, resourcing, assumptions, and price. Legally binding.
Who
Delivery Lead + Client sponsor (signed)
When
After discovery/estimation, before build starts
Practice
Draft a one-page SOW for Cartwell: Wave 1 use cases, tools in scope, deliverables, assumptions and a week-by-week timeline.
SOW (Statement of Work)
Screenshot SOW (Statement of Work)Add screenshots/phases/step-3-1.png

What each tool needs in this phase

Tool3. SOW & plan
AEPScope sources, schemas, Wave 1 use cases; check volumes vs. licence
Adobe AnalyticsVariables, suites, dashboards
Launch (Tags)Templates, extensions, migration scope
MarketoPrograms, scoring, CRM sync scope
RTCDPAudience count, destinations, governance scope
AEMTemplates, components, migration volume
TargetImplementation plus first activities
AJOWave 1 journeys, channels, templates
CJAConnections, data views, projects

Interview questions tap to reveal a model answer

What makes a good SOW?

Clear scope and out-of-scope, deliverables tied to milestones, assumptions, client responsibilities, timeline, team and price.

How do you estimate an AEP implementation?

Count sources, schemas, use cases, destinations and environments, apply effort per item from past projects, then add QA, hypercare and contingency.

What do you do when the client asks for more mid-project?

Log it, assess impact on time and cost, and handle it through a change request rather than absorbing it silently.

Phase 4 · Design

Gate to pass: SDR + SDD — Design sign-off

Step 4.1

SDR (Solution/Source Design Reference)

What it is
The master mapping of business requirements to technical configuration — e.g., variables, schemas, events, fields. Living document during build.
Who
Technical Lead / Architect
When
Design phase, updated through build
Practice
Build an SDR spreadsheet for SkyLumen: business question → event → XDM field → Analytics variable → identity.
SDR (Solution/Source Design Reference)
Screenshot SDR (Solution/Source Design Reference)Add screenshots/phases/step-4-1.png
Step 4.2

SDD (Solution Design Document)

What it is
Detailed technical design: architecture diagrams, data flows, environment topology, integration specs, naming conventions.
Who
Solution Architect
When
Design phase, precedes build
Practice
Write an SDD outline: architecture diagram, data flow, environments, integrations and naming conventions.
SDD (Solution Design Document)
Screenshot SDD (Solution Design Document)Add screenshots/phases/step-4-2.png

What each tool needs in this phase

Tool4. Design (SDR / SDD)
AEPSDR: schemas, field groups, namespaces, datasets, mapping. SDD: data flow, sandboxes, governance
Adobe AnalyticsSDR, suite settings, classification plan
Launch (Tags)Data layer spec, rule and data-element design, naming, consent matrix
MarketoLifecycle and scoring model, field mapping, program templates
RTCDPMerge policies, audience catalog, destination matrix, labels and policies
AEMIA, component catalog, fragment models; SDD for environments, CDN, integrations
TargetRoadmap, audiences, A4T, properties/workspaces
AJOJourney designs, event specs, channel configs, rule sets, consent model
CJAConnection and data view spec, stitching, derived fields

Solution architecture: how the tools fit together

Customer touchpoints

  • Websitepages served by AEM
  • Mobile appMobile SDK
  • Inbox & devicesemail, SMS, push, in-app
  • Stores & call centrePOS and contact-centre systems
  • Sales teamCRM (Salesforce / Dynamics)

Collection & content

  • AEM Sites & Assetspages, fragments, assets; Tags via Cloud Services
  • Tags + Web / Mobile SDKdata layer → XDM events, consent
  • Edge Network & datastreamsroutes events to AEP, Analytics, Target
  • Source connectorsbatch and streaming from CRM, POS, warehouse
  • Marketo EngageB2B leads, Munchkin, CRM sync

Experience Platform foundation

  • Data lake: schemas & datasetsXDM schemas, datasets, Data Prep
  • Identity Servicenamespaces, identity graph
  • Real-Time Customer Profilemerged profiles, merge policies
  • Query ServiceSQL validation, derived datasets
  • Data governancelabels, policies, consent

Applications

  • Adobe Analyticsfed by the datastream Analytics service
  • Customer Journey Analyticsconnections and data views on AEP datasets
  • Real-Time CDPaudiences and destinations
  • Journey Optimizerjourneys, channels, decisioning
  • Targettests and personalization at the edge

Activation & insight

  • Dashboards & BIWorkspace, Tableau
  • CRM & sales alertscustom actions, Marketo, RTCDP
  • Ad platforms & partnersRTCDP destinations
  • Email · SMS · Push · In-appAJO channel delivery
  • Web & app personalizationTarget and AJO web/code-based

Master configuration sequence (the build order)

All tools in dependency order. "Depends on" lists the row numbers that must be finished first.

#ToolConfigurationDepends onOutputPhase
1Admin Console / AEPProduct profiles and roles; Dev and Prod sandboxes—Users can work in the right sandboxBuild
2AEPIdentity namespaces (email, phone, CRM ID, loyalty ID, ECID, account ID)1Namespace list in SDRBuild
3AEPSchemas and field groups for profiles, events and accounts2Published schemasBuild
4AEPDatasets with Profile toggle and retention3Datasets ready for ingestionBuild
5Data CollectionDatastreams per environment with AEP, Analytics and Target services3, 4Datastream IDsBuild
6Adobe AnalyticsReport suites, props, eVars, events, bot rules— (parallel)Report suite IDs for the datastreamBuild
7TagsProperty, Web SDK extension, data elements, rules, consent5Events reaching AEP and AnalyticsBuild
8AEPBatch and streaming sources with Data Prep mapping (CRM, POS, warehouse)4Dataflows runningBuild
9MarketoAdmin, email authentication, Munchkin via Tags, CRM sync7Marketo database synced with CRMBuild
10AEPMarketo source connector into AEP3, 9Leads and activities in AEPBuild
11AEPIngestion validation with Query Service (counts, nulls, duplicates)7, 8, 10Validation reportQA
12AEP / RTCDPIdentity graph checks; merge policies11Trusted unified profilesBuild
13AEPData usage labels and governance policies3Policies enforced on activationBuild
14RTCDPAudiences and evaluation methods12, 13Audience libraryBuild
15RTCDPDestinations and activation (ads, Marketo, Target, CRM)14Audiences flowing to channelsBuild
16AEMProgram, pipelines, templates, components, fragments, DAM, workflows— (parallel from start)Site ready for contentBuild
17AEM / TagsTags configured on AEM via Cloud Services7, 16Tracking on every AEM pageBuild
18TargetTarget service in datastream; properties; A4T; first activities5, 14Tests and personalization liveBuild
19AJOSubdomains, IP pools, channel configurations, rule sets1Channels ready to sendBuild
20AJOUnitary and business events; data sources; custom actions5, 8Triggers available in journeysBuild
21AJOTemplates, decisioning, journeys14, 20Journeys ready for testBuild
22CJAConnection (web, CRM, AJO step events, Analytics data); data views; projects11, 21Cross-channel reportingBuild
23All toolsEnd-to-end test: one synthetic person from web visit → profile → audience → journey → report1–22End-to-end test logQA / UAT
24All toolsGo-live in waves; hypercare monitoring across dataflows, activations, journeys, tags23Go-Live and Hypercare Exit sign-offsGo-live & hypercare
25All toolsHandoff: runbooks, SDR/SDD final versions, access list, training, backlog24Closure and handoff packClosure

Configuration objects per tool, in build order

ToolObjects
AEPSandbox Dev and Prod; product profiles and roles → Identity namespaces Display name, code, type (cross-device / device) → Schemas Class (Profile / Event), standard field groups, _tenant custom fields → Identity fields Namespace per field; one primary; Profile toggle on schema → Datasets From schema; Profile toggle; retention → Sources Connector, auth, dataflow schedule (batch / streaming) → Data Prep mapping Source → XDM; calculated fields for format fixes → Profile & merge Default merge policy; identity graph checks → Query Service Validation SQL; CTAS derived datasets → Governance Data usage labels; policies on marketing actions → Monitoring Dataflow alerts; ingestion dashboards
RTCDPMerge policies Stitching type; timestamp or dataset precedence; default → Audiences Audience Builder rules, time constraints → Evaluation Batch, streaming or edge per audience → Labels & policies Marketing actions mapped to policies → Destination connection Catalog card, auth, marketing actions → Activation dataflow Audiences, schedule, identity + attribute mapping → Monitoring Dataflow runs, match rates, failures
Tags + Web SDKDatastream Event schema; services: AEP dataset, Analytics RSID, Target; ID settings → Tag property Domains; Dev / Stage / Prod environments → Web SDK extension Datastream per environment; consent default; ECID → Data elements Data layer readers; XDM object / variable → Rules Event + conditions + Send event action → Consent CMP → setConsent; rule categories gated → Library & publish Build Dev → approve Stage → publish Prod → Embed code Environment-specific script on each site → Validate Debugger, Assurance, dataset preview
Adobe AnalyticsReport suite RSID, time zone, currency, go-live → Props & eVars Names, allocation, expiration → Success events Counter / currency / numeric; serialization → Bot & IP rules IAB filter; internal IP exclusion → Data collection Datastream Analytics service or extension; XDM → variables → Processing rules Server-side copy / set / clean → Classifications Classification sets for codes → Components Segments, calculated metrics → Workspace Projects, scheduled reports → Data Feeds Hit-level export to warehouse
AJOSubdomain delegation DNS delegation; SPF, DKIM, DMARC → IP pools & warm-up Pools per message type; warm-up plan → Channel configurations Marketing vs. transactional; sender; tracking → Events Unitary (identity) or business; schema; event ID condition → Data sources & custom actions External APIs, auth, payloads → Business rule sets Channel caps, quiet hours, journey entry caps → Templates & fragments Personalization tokens with fallbacks → Decisioning Items, eligibility rules, selection strategies → Journeys Entry, conditions, waits, actions, exit, re-entry → Test & dry run Test profiles, proofs, dry run → Publish & report Live report; journey_step_events
CJAConnection Sandbox, datasets, person/account ID, backfill → Lookup & profile data Keys matched to event datasets → Stitching Field-based stitching where needed → Data view Containers, session timeout, time zone → Components Dimensions, metrics, derived fields, attribution → Filters & metrics Reusable filters, calculated metrics → Workspace Projects, flow, fallout, cohort → Audiences Publish to Real-Time CDP → Access Product profiles per data view
TargetImplementation Datastream Target service (Web SDK) or at.js; pre-hiding snippet → Properties & workspaces Enterprise permissions → Analytics for Target A4T reporting source → Audiences Target rules or RTCDP audiences → Activity A/B, XT, MVT, AP; URL or location → Experiences VEC or form-based composer → Targeting & goals Allocation; primary metric → QA links Every experience, every device → Activate & report Split check; significance
AEMCloud Manager program Program and roles → Environments Author, Publish, Preview; Dev / Stage / Prod → Repo & pipelines Git; quality gates; deploy → Project code Archetype; Core Component proxies; client libraries → Dispatcher & CDN Cache rules, filters, redirects → Templates & policies Editable templates; allowed components → Content Fragment Models Models; GraphQL endpoints for headless → Assets Processing profiles; metadata schemas → Workflows & permissions Approval models; groups → Cloud Services Tags, Target, Analytics configs → Publish & monitor Replication, logs, uptime
MarketoUsers & workspaces Roles; workspaces and partitions → Email authentication SPF, DKIM, DMARC; branding CNAME → Munchkin Tracking code via Tags → CRM sync Sync user, Sync Fields, start sync → Field management Custom fields; mapping → Smart lists & segmentations Core lists; dynamic content segments → Program setup Channels, tags, period costs → Scoring & lifecycle Change Score campaigns; stage transitions → Templates, pages, forms Progressive profiling → Engagement programs Streams, cadence, transitions → Reporting & AEP Program reports; AEP source / destination

Interview questions tap to reveal a model answer

SDR vs. SDD?

The SDR maps business requirements to technical configuration (variables, schemas, fields); the SDD describes the technical design: architecture, data flows, environments, integrations and naming.

Who signs off design and why does it matter?

The client sponsor and technical owners. Sign-off freezes scope for the build, so later changes go through change control.

How do you design identity for an airline?

Loyalty ID as the main person ID, plus email, phone and ECID; booking references stay as fields, and agents' shared emails are excluded from stitching.