<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Infinite360 - guides and student questions</title><link>https://infiante360techacademy.com/</link><description>Infinite360 - guides and student questions from Infinite360: Adobe MarTech, AEP, AJO, CJA and Agentic AI.</description><language>en</language><atom:link href="https://infiante360techacademy.com/feeds/guides.xml" rel="self" type="application/rss+xml"/><item><title>AEP schemas: designing XDM schemas step by step</title><link>https://infiante360techacademy.com/adobe-experience-platform/xdm-schemas.html</link><guid isPermaLink="true">https://infiante360techacademy.com/adobe-experience-platform/xdm-schemas.html</guid><category>Guides</category><pubDate>Wed, 28 Oct 2026 09:00:00 +0000</pubDate><description>An AEP schema is an Experience Data Model (XDM) definition of how data is structured. Each schema is built from a class (usually XDM Individual Profile for attributes or XDM ExperienceEvent for time-stamped events) plus field groups. Mark identity fields, choose a primary identity, and enable the schema for Profile only when the data should feed Real-Time Customer Profile.</description></item><item><title>AEP datasets: setup, enablement and troubleshooting</title><link>https://infiante360techacademy.com/adobe-experience-platform/datasets.html</link><guid isPermaLink="true">https://infiante360techacademy.com/adobe-experience-platform/datasets.html</guid><category>Guides</category><pubDate>Wed, 28 Oct 2026 09:00:00 +0000</pubDate><description>An AEP dataset is a storage and management container for data that conforms to one XDM schema. Data lands in the data lake first; if the dataset is enabled for Profile and Identity, records also feed Real-Time Customer Profile and the identity graph. Datasets show batch status, errors and ingested record counts for monitoring.</description></item><item><title>AEP data ingestion: batch, streaming and sources</title><link>https://infiante360techacademy.com/adobe-experience-platform/data-ingestion.html</link><guid isPermaLink="true">https://infiante360techacademy.com/adobe-experience-platform/data-ingestion.html</guid><category>Guides</category><pubDate>Wed, 28 Oct 2026 09:00:00 +0000</pubDate><description>AEP ingests data in two modes. Batch ingestion loads files or scheduled extracts through source connectors or the Batch Ingestion API. Streaming ingestion sends records in near real time through the HTTP API or the Edge Network via Web SDK and Mobile SDK. Data Prep maps and transforms incoming fields into the target XDM schema.</description></item><item><title>AEP identity: how identity resolution works</title><link>https://infiante360techacademy.com/adobe-experience-platform/identity-service.html</link><guid isPermaLink="true">https://infiante360techacademy.com/adobe-experience-platform/identity-service.html</guid><category>Guides</category><pubDate>Wed, 28 Oct 2026 09:00:00 +0000</pubDate><description>AEP Identity Service links identifiers that belong to the same person into an identity graph. Each identifier is tagged with a namespace such as ECID, Email, Phone or a custom CRM ID. When records sharing an identifier are ingested into Profile-enabled datasets, their identities are linked, allowing Real-Time Customer Profile to merge fragments into one view.</description></item><item><title>AEP profiles: Real-Time Customer Profile explained</title><link>https://infiante360techacademy.com/adobe-experience-platform/real-time-customer-profile.html</link><guid isPermaLink="true">https://infiante360techacademy.com/adobe-experience-platform/real-time-customer-profile.html</guid><category>Guides</category><pubDate>Wed, 28 Oct 2026 09:00:00 +0000</pubDate><description>Real-Time Customer Profile merges profile fragments from Profile-enabled datasets into one view per person, using the identity graph to decide which fragments belong together and a merge policy to decide which values win. Profiles combine attributes and behavioural events and are used for segmentation, personalisation and activation.</description></item><item><title>AEP segments and audiences: a practical guide</title><link>https://infiante360techacademy.com/adobe-experience-platform/segmentation-audiences.html</link><guid isPermaLink="true">https://infiante360techacademy.com/adobe-experience-platform/segmentation-audiences.html</guid><category>Guides</category><pubDate>Wed, 28 Oct 2026 09:00:00 +0000</pubDate><description>AEP segmentation evaluates Real-Time Customer Profiles against rules to build audiences. Batch evaluation runs on a schedule for complex rules; streaming evaluation updates qualification as events arrive; edge evaluation qualifies profiles instantly on the Edge Network for same-page personalisation. Audiences can also be imported or created from other sources.</description></item><item><title>AEP activation: sending audiences to destinations</title><link>https://infiante360techacademy.com/adobe-experience-platform/activation-destinations.html</link><guid isPermaLink="true">https://infiante360techacademy.com/adobe-experience-platform/activation-destinations.html</guid><category>Guides</category><pubDate>Wed, 28 Oct 2026 09:00:00 +0000</pubDate><description>Activation sends AEP audiences and selected profile attributes to destinations such as advertising platforms, email and CRM systems, cloud storage and Adobe applications. Streaming destinations update as profiles qualify; batch (file-based) destinations export on a schedule. Data governance policies and consent are enforced before export.</description></item><item><title>AEP architecture: an enterprise reference model</title><link>https://infiante360techacademy.com/adobe-experience-platform/architecture.html</link><guid isPermaLink="true">https://infiante360techacademy.com/adobe-experience-platform/architecture.html</guid><category>Guides</category><pubDate>Wed, 28 Oct 2026 09:00:00 +0000</pubDate><description>A sound AEP architecture separates environments with sandboxes, defines a single identity strategy, models data in XDM around use cases, enforces governance with data usage labels and consent, and integrates sources and destinations through monitored dataflows and APIs. It is documented so that every dataset, audience and destination has an owner.</description></item><item><title>AEP implementation: an end-to-end guide</title><link>https://infiante360techacademy.com/adobe-experience-platform/implementation-guide.html</link><guid isPermaLink="true">https://infiante360techacademy.com/adobe-experience-platform/implementation-guide.html</guid><category>Guides</category><pubDate>Wed, 28 Oct 2026 09:00:00 +0000</pubDate><description>An Adobe Experience Platform implementation moves from prioritised use cases to data model, identity strategy, ingestion, profile and audiences, activation, and finally QA and go-live. Successful projects deliver one end-to-end use case first, prove data quality, then scale to more sources and channels.</description></item><item><title>AEP interview questions and answers</title><link>https://infiante360techacademy.com/adobe-experience-platform/interview-questions.html</link><guid isPermaLink="true">https://infiante360techacademy.com/adobe-experience-platform/interview-questions.html</guid><category>Guides</category><pubDate>Wed, 28 Oct 2026 09:00:00 +0000</pubDate><description>Strong AEP interview answers explain the end-to-end data flow, justify design decisions and show how you would troubleshoot. Expect questions on XDM schemas, datasets, ingestion, identity, profiles and merge policies, segmentation methods, activation, governance and how AEP connects to Real-Time CDP, AJO and CJA.</description></item><item><title>AEP career path: roles, skills and roadmap</title><link>https://infiante360techacademy.com/adobe-experience-platform/career-path.html</link><guid isPermaLink="true">https://infiante360techacademy.com/adobe-experience-platform/career-path.html</guid><category>Guides</category><pubDate>Wed, 28 Oct 2026 09:00:00 +0000</pubDate><description>Common Adobe Experience Platform career paths are implementation specialist, AEP developer or data engineer, AEP consultant, and AEP solution architect. Each builds on the same core skills: XDM data modelling, ingestion, identity, profiles, segmentation, activation and governance, plus adjacent skills such as SQL, APIs, Web SDK, AJO and CJA.</description></item><item><title>AEP vs Real-Time CDP: what is the difference?</title><link>https://infiante360techacademy.com/guides/aep-vs-real-time-cdp.html</link><guid isPermaLink="true">https://infiante360techacademy.com/guides/aep-vs-real-time-cdp.html</guid><category>Guides</category><pubDate>Wed, 28 Oct 2026 09:00:00 +0000</pubDate><description>Adobe Experience Platform (AEP) is the underlying data foundation: schemas, ingestion, identity, profiles, segmentation and governance. Real-Time CDP (RTCDP) is an application built on AEP that adds customer data platform capabilities, most notably activating audiences and profiles to marketing and advertising destinations.</description></item><item><title>CJA vs Adobe Analytics: use cases and trade-offs</title><link>https://infiante360techacademy.com/guides/cja-vs-adobe-analytics.html</link><guid isPermaLink="true">https://infiante360techacademy.com/guides/cja-vs-adobe-analytics.html</guid><category>Guides</category><pubDate>Wed, 28 Oct 2026 09:00:00 +0000</pubDate><description>Adobe Analytics is a digital analytics product built on report suites, mainly for web and app data. Customer Journey Analytics (CJA) analyses any AEP dataset, online or offline, joined on a person ID, enabling cross-channel journey analysis. Many organisations migrate to CJA with Web SDK, while weighing reporting continuity and feature differences.</description></item><item><title>What is MCP, and how can it be used safely with enterprise MarTech?</title><link>https://infiante360techacademy.com/guides/what-is-mcp.html</link><guid isPermaLink="true">https://infiante360techacademy.com/guides/what-is-mcp.html</guid><category>Guides</category><pubDate>Wed, 28 Oct 2026 09:00:00 +0000</pubDate><description>The Model Context Protocol (MCP) is an open standard that lets AI applications connect to external tools and data through MCP servers. In enterprise MarTech, MCP can give AI agents controlled access to documentation, configuration checks or approved data, provided access is least-privilege, actions are approved and every call is logged.</description></item><item><title>Marketo Engage and Adobe Experience Platform</title><link>https://infiante360techacademy.com/guides/marketo-engage-aep.html</link><guid isPermaLink="true">https://infiante360techacademy.com/guides/marketo-engage-aep.html</guid><category>Guides</category><pubDate>Wed, 28 Oct 2026 09:00:00 +0000</pubDate><description>Marketo Engage is Adobe&#x27;s B2B marketing automation application for lead management, programs, scoring, email and CRM alignment. Connected with Adobe Experience Platform, Marketo data can feed AEP profiles and AEP audiences can be activated to Marketo, so B2B lead nurture uses the same governed customer data as other channels.</description></item><item><title>AEP student questions, answered</title><link>https://infiante360techacademy.com/learn/student-questions.html</link><guid isPermaLink="true">https://infiante360techacademy.com/learn/student-questions.html</guid><category>Guides</category><pubDate>Wed, 28 Oct 2026 09:00:00 +0000</pubDate><description>These guides answer questions learners asked during live Infinite360 batches - about schemas, classes, identities, hashed email, ingestion, eVars and props, backfill, merge policies, Web SDK, data elements, journeys and consent. Each answer is written in plain language, anonymised and linked to Adobe Experience League for verification.</description></item><item><title>The MarTech implementation backlog and dependency tracker</title><link>https://infiante360techacademy.com/guides/martech-implementation-backlog.html</link><guid isPermaLink="true">https://infiante360techacademy.com/guides/martech-implementation-backlog.html</guid><category>Guides</category><pubDate>Sun, 11 Oct 2026 09:00:00 +0000</pubDate><description>A MarTech backlog is the prioritised list of everything an Adobe program must deliver - data requirements, schemas, identities, sources, audiences, journeys, destinations, tags and integrations - with a dependency for each item. It starts in pre-planning and discovery, and keeps steering delivery, migration and support. Use cases must be final before architecture, schema, journey and audience design.</description></item><item><title>How to integrate OneTrust consent with Adobe Experience Platform</title><link>https://infiante360techacademy.com/guides/onetrust-consent-aep.html</link><guid isPermaLink="true">https://infiante360techacademy.com/guides/onetrust-consent-aep.html</guid><category>Guides</category><pubDate>Sun, 11 Oct 2026 09:00:00 +0000</pubDate><description>To use OneTrust or any consent management platform with AEP, collect consent at the edge with the CMP, send the consent signal to Adobe Experience Platform through Web SDK, store it in a consent field in your schema, enable data usage and privacy policies so AEP enforces it, and keep it updated when people change preferences. Consent is individual permission; governance labels classify data.</description></item><item><title>Hashed email identity and profile stitching in AEP</title><link>https://infiante360techacademy.com/guides/hashed-email-identity-aep.html</link><guid isPermaLink="true">https://infiante360techacademy.com/guides/hashed-email-identity-aep.html</guid><category>Guides</category><pubDate>Sun, 11 Oct 2026 09:00:00 +0000</pubDate><description>Yes - AEP can stitch profiles on a hashed email. Stitching only needs the same identity value in every dataset, so if web, CRM and mobile all send the same lowercase SHA-256 hash under the Email_LC_SHA256 namespace, AEP links them into one profile. The raw email is not required, but hashing must be identical everywhere.</description></item><item><title>AEP ingestion methods: batch vs streaming</title><link>https://infiante360techacademy.com/guides/aep-ingestion-methods-batch-streaming.html</link><guid isPermaLink="true">https://infiante360techacademy.com/guides/aep-ingestion-methods-batch-streaming.html</guid><category>Guides</category><pubDate>Sun, 11 Oct 2026 09:00:00 +0000</pubDate><description>AEP has three common ingestion routes. Source connections can be batch or streaming depending on the source - cloud storage is batch, Kafka-style streams are streaming. Workflows are batch because they ingest files such as CSV or Parquet. Data collection through datastreams is streaming because Web and Mobile SDKs send events in real time.</description></item><item><title>eVars vs props in Adobe Analytics</title><link>https://infiante360techacademy.com/guides/evars-vs-props-adobe-analytics.html</link><guid isPermaLink="true">https://infiante360techacademy.com/guides/evars-vs-props-adobe-analytics.html</guid><category>Guides</category><pubDate>Sun, 11 Oct 2026 09:00:00 +0000</pubDate><description>In Adobe Analytics, props are traffic variables: they describe what happened on a hit and expire immediately, which makes them useful for pathing. eVars are conversion variables: they persist beyond the hit, so they can be credited for later success events such as purchases or sign-ups. Use eVars with success events to measure conversion.</description></item><item><title>What backfill months means in AEP and CJA</title><link>https://infiante360techacademy.com/guides/backfill-months-aep-cja.html</link><guid isPermaLink="true">https://infiante360techacademy.com/guides/backfill-months-aep-cja.html</guid><category>Guides</category><pubDate>Sun, 11 Oct 2026 09:00:00 +0000</pubDate><description>Backfill months tells Adobe how many months of historical data already available to the platform to process and make usable. It does not pull data from outside systems. In AEP it makes older data available to Profile, audiences and destinations; in Customer Journey Analytics it loads that many months of past data into the connection for reporting.</description></item><item><title>Using the Adobe Analytics source connector in AEP</title><link>https://infiante360techacademy.com/guides/adobe-analytics-source-connector-aep.html</link><guid isPermaLink="true">https://infiante360techacademy.com/guides/adobe-analytics-source-connector-aep.html</guid><category>Guides</category><pubDate>Sun, 11 Oct 2026 09:00:00 +0000</pubDate><description>When you add Adobe Analytics from the AEP sources catalog, Analytics becomes a data source and its report-suite data lands in AEP as Experience Events. You map it to datasets, then use it for Profile, audiences, CJA and activation. Profiles unify because Analytics events carry an identity - usually ECID - that matches identities in other datasets.</description></item><item><title>Record schemas vs the XDM Individual Profile class</title><link>https://infiante360techacademy.com/guides/record-vs-xdm-individual-profile.html</link><guid isPermaLink="true">https://infiante360techacademy.com/guides/record-vs-xdm-individual-profile.html</guid><category>Guides</category><pubDate>Sun, 11 Oct 2026 09:00:00 +0000</pubDate><description>A record schema is not automatically an XDM Individual Profile schema. Record is a data behaviour: it stores attributes about an entity and does not expire. A schema becomes an XDM Individual Profile schema only when you choose that class. The class defines meaning; the behaviour defines how data changes over time.</description></item><item><title>Adobe Journey Optimizer: campaigns vs journeys</title><link>https://infiante360techacademy.com/guides/ajo-campaigns-vs-journeys.html</link><guid isPermaLink="true">https://infiante360techacademy.com/guides/ajo-campaigns-vs-journeys.html</guid><category>Guides</category><pubDate>Sun, 11 Oct 2026 09:00:00 +0000</pubDate><description>In Adobe Journey Optimizer, a campaign sends a one-time or scheduled message to an audience - a festive email blast, for example. A journey is an automated, multi-step flow that reacts to each person&#x27;s behaviour, such as welcome, follow-up and reminder. Use campaigns for fixed-date broadcasts and journeys for event-based, always-on experiences.</description></item><item><title>Web SDK troubleshooting: events not reaching AEP</title><link>https://infiante360techacademy.com/guides/web-sdk-troubleshooting.html</link><guid isPermaLink="true">https://infiante360techacademy.com/guides/web-sdk-troubleshooting.html</guid><category>Guides</category><pubDate>Sun, 11 Oct 2026 09:00:00 +0000</pubDate><description>When a website sends Web SDK events but nothing appears in AEP, work from the browser to the platform: confirm interact requests reach the Edge Network, check the datastream ID and that the datastream is active with the Experience Platform service enabled, then verify the dataset mapping, XDM schema, identity namespaces and ingestion monitoring for failures.</description></item><item><title>How to capture ECID in an Adobe Tags data element</title><link>https://infiante360techacademy.com/guides/ecid-data-element-adobe-tags.html</link><guid isPermaLink="true">https://infiante360techacademy.com/guides/ecid-data-element-adobe-tags.html</guid><category>Guides</category><pubDate>Sun, 11 Oct 2026 09:00:00 +0000</pubDate><description>To use ECID in Adobe Tags, call Web SDK getIdentity, store the returned ECID with _satellite.setVar in a rule, and create a custom-code data element that returns _satellite.getVar(&#x27;ECID&#x27;). Reference it as %ECID% in other rules, and validate it in the browser console with _satellite.getVar and the Tags debugger.</description></item><item><title>Adobe Tags data elements: enterprise best practices</title><link>https://infiante360techacademy.com/guides/data-elements-best-practices.html</link><guid isPermaLink="true">https://infiante360techacademy.com/guides/data-elements-best-practices.html</guid><category>Guides</category><pubDate>Sun, 11 Oct 2026 09:00:00 +0000</pubDate><description>Good data elements read from a standardised data layer, are reused across rules, return safe defaults instead of null or undefined, and are documented with an owner and definition. In single-page apps, update the data layer first and then fire a custom event so rules read fresh values. Changes are versioned and tested so one team cannot break another.</description></item><item><title>How to validate large batch ingestion in AEP</title><link>https://infiante360techacademy.com/guides/validate-large-batch-ingestion.html</link><guid isPermaLink="true">https://infiante360techacademy.com/guides/validate-large-batch-ingestion.html</guid><category>Guides</category><pubDate>Sun, 11 Oct 2026 09:00:00 +0000</pubDate><description>To confirm a large load - say ten million records - reached an AEP dataset, compare batch metadata record counts with the source, run Query Service counts and distinct-key counts, check for duplicates and nulls, and validate each batch when loads arrive in parts. Deduplicate and clean files upstream so problems are not carried into Profile.</description></item><item><title>Profile unification strategy and merge policies in AEP</title><link>https://infiante360techacademy.com/guides/profile-merge-policies-unification.html</link><guid isPermaLink="true">https://infiante360techacademy.com/guides/profile-merge-policies-unification.html</guid><category>Guides</category><pubDate>Sun, 11 Oct 2026 09:00:00 +0000</pubDate><description>A unification strategy defines how AEP joins data from different datasets into one customer profile: which identities link records, which datasets are trusted for Profile, and which merge policy wins when two sources disagree - for example CRM name overriding web name. Debug problems by checking identity configuration, data quality, profile enablement, relationships and merge policies.</description></item><item><title>Adobe Journey Optimizer: enterprise implementation best practices</title><link>https://infiante360techacademy.com/guides/ajo-implementation-best-practices.html</link><guid isPermaLink="true">https://infiante360techacademy.com/guides/ajo-implementation-best-practices.html</guid><category>Guides</category><pubDate>Sun, 11 Oct 2026 09:00:00 +0000</pubDate><description>Implement Adobe Journey Optimizer as an orchestration layer on a governed AEP data foundation, not as isolated campaigns. Define use cases and KPIs, establish XDM and identity, configure channels and consent, build reusable modular journeys, test in non-production, publish controlled versions, then monitor and optimise continuously.</description></item><item><title>Adobe Advertising Analytics: frequently asked questions</title><link>https://infiante360techacademy.com/guides/advertising-analytics-faq.html</link><guid isPermaLink="true">https://infiante360techacademy.com/guides/advertising-analytics-faq.html</guid><category>Guides</category><pubDate>Sun, 11 Oct 2026 09:00:00 +0000</pubDate><description>Advertising Analytics brings impression, click and cost data from Google Ads and Microsoft Advertising into Adobe Analytics so paid search can be analysed with on-site behaviour. It is available with Analytics Select, Prime and Ultimate, does not incur server calls and is set up from the Analytics Admin menu. Answers here summarise Adobe&#x27;s documentation; confirm current details there.</description></item><item><title>Customer engagement architecture: AEP plus agentic AI</title><link>https://infiante360techacademy.com/guides/agentic-customer-engagement-architecture.html</link><guid isPermaLink="true">https://infiante360techacademy.com/guides/agentic-customer-engagement-architecture.html</guid><category>Guides</category><pubDate>Mon, 19 Oct 2026 09:00:00 +0000</pubDate><description>A safe agentic architecture keeps Adobe Experience Platform as the system of record for identity and profile, Real-Time CDP for audiences, Journey Optimizer as the single execution engine and Customer Journey Analytics for measurement. AI agents read profile and journey state, propose actions through narrowly scoped MCP tools and a guardrail check, and never send messages directly.</description></item></channel></rss>
