Marketing Cloud Intelligence
Marketing Cloud Intelligence is the Salesforce marketing analytics platform that pulls data from many marketing and advertising sources into one model, then turns it into cross-channel dashboards and reports.
Definition
Marketing Cloud Intelligence is the Salesforce marketing analytics platform that pulls data from many marketing and advertising sources into one model, then turns it into cross-channel dashboards and reports. It was the Datorama product before Salesforce acquired the company in 2018 and rebranded it as Marketing Cloud Intelligence in 2022. The platform is built for marketers, with a connector library covering hundreds of ad, email, web, social, and CRM sources.
The point of the product is comparison across channels. Each platform's own analytics shows only its slice, so questions like "did paid search or email drive more revenue" need the numbers sitting side by side. Marketing Cloud Intelligence ingests every feed, harmonizes the fields into a shared taxonomy, and lets analysts build leadership-ready views on top. Salesforce has since folded these ideas into a newer offering, Marketing Intelligence in Marketing Cloud Next, but the Datorama-era product is still in market and widely deployed.
How Marketing Cloud Intelligence turns scattered feeds into one view
From Datorama to a Salesforce product
Datorama started as an independent marketing-data company with a large library of pre-built connectors and a flexible data model. Salesforce acquired it in 2018 and rebranded the product as Marketing Cloud Intelligence in 2022. The rename did not change the underlying engine. Documentation, support articles, and Trailhead modules still cross-reference both names, so a search for either term lands on the same platform. The acquisition slotted the product into the wider Marketing Cloud family, alongside the engagement and personalization tools, and opened paths to feed insights back into other Salesforce clouds. Most of the strengths that made Datorama popular carried over intact: the breadth of the connector library, the marketer-friendly semantic layer, and the dashboard templates aimed at marketing leadership. The heritage matters in practice because older blog posts, community threads, and even some in-app labels predate the rename. Teams adopting the platform today should treat Datorama content and Marketing Cloud Intelligence content as describing the same product, and check the publish date to know which era a given screenshot or field name belongs to.
The connector library and ingestion methods
The connector library is the headline feature. Marketing Cloud Intelligence ships with integrations for advertising platforms (Google Ads, Facebook Ads, LinkedIn Ads, TikTok Ads, Amazon Advertising), web analytics (Google Analytics, Adobe Analytics), CRM (Salesforce, Marketo, HubSpot, Pardot), email, social, and commerce. Building those integrations in-house would take years, which is much of why teams buy the platform rather than wire feeds together themselves. Data enters through three documented methods. API Connectors pull directly from a source system on a schedule. TotalConnect is the marketer-built option for file-based feeds, automating ingestion, cleansing, and mapping for any CSV or report that does not have a native connector. LiteConnect handles lighter file uploads. Each feed becomes a data stream, and streams come in typed flavors such as classification, CRM, ecommerce, generic, and messaging. The variety of methods means almost any source can land in the platform, whether it has a slick API or just exports a spreadsheet. The trade-off is that more feeds mean more streams to maintain, so connector setup is rarely a one-time task.
Data harmonization and the semantic model
Every source arrives with its own field names, currencies, and category labels. One platform calls a metric "spend," another calls it "cost," and a third splits it by region. Marketing Cloud Intelligence harmonizes these into a shared taxonomy so that impressions, clicks, conversions, revenue, and spend line up across channels. The work happens in the Harmonization Center, a no-code area built around several tools that handle naming conventions, classification rules, and field mapping. The platform's data model is dimensional. Dimensions are the descriptive attributes you slice by, including calculated and pacing dimensions, while measurements are the numbers you aggregate, including calculated and filtered measurements. Mapping source fields into this model is configurable, so admins decide how each input lines up with the canonical structure. Harmonization is usually the part of a rollout that takes the longest. The dashboards are quick to draw once the data is clean, but reconciling dozens of feeds into one consistent set of dimensions and measurements is detailed work. Getting it right early pays off, because every later report inherits the taxonomy decisions made here.
Dashboards, widgets, and analysis
Once data is harmonized, analysts assemble dashboards from interactive widgets. Dashboard pages group into collections, and a template library covers the views marketing leaders ask for most: total spend by channel, conversion rate by source, return on investment by campaign, and pacing against budget. Authoring is drag-and-drop, so building a view does not require code. Beyond dashboards, the platform offers pivot tables and reports for deeper analysis, plus an activation center that can trigger rule-based actions when metrics cross a threshold. A common pattern is a single executive dashboard that rolls every channel into one spend-and-return summary, with drill-downs into each source for the analysts who manage them. The discipline that matters here is restraint. It is easy to spin up hundreds of dashboards, and clutter kills adoption. A curated set, one tailored view per audience, tends to drive more usage than a sprawling gallery nobody trusts. Because the widgets read from the harmonized model, a clean taxonomy upstream is what makes the dashboards quick to build and easy to keep accurate over time.
Multi-touch attribution
Channel-by-channel reports have a known flaw. If email and paid search both touched a buyer before they converted, crediting the full sale to each channel double-counts the result. Attribution fixes this by distributing credit across the touchpoints on a path. Marketing Cloud Intelligence supports the standard models: first-touch, last-touch, linear, time-decay, and position-based. Each spreads credit differently, and the choice changes the story the numbers tell. Last-touch flatters the channel that closes, first-touch flatters the channel that opens, and linear treats every touch as equal. Attribution is one of the most requested analyses from marketing executives, because it underpins how budget gets allocated. The model you pick shapes every ROI conversation, so it deserves a deliberate decision rather than a default. A worked example: a customer sees a display ad, clicks a paid search result a week later, then converts from an email. Last-touch gives email all the credit, first-touch gives the display ad all of it, and linear splits it three ways. None is wrong, but reporting only one without saying which can mislead the people setting the budget.
Fitting into the Salesforce stack
The strongest argument for the Salesforce-native option over a standalone tool is the closed loop between analysis and activation. Marketing Cloud Intelligence can feed insights back to other products: audience refinement in the engagement tools, lead-scoring tuning in Pardot, and profile enrichment in Data Cloud. Salesforce also documents pulling Data Cloud (now branded Data 360) data into Intelligence through TotalConnect, so the customer data platform and the analytics platform share information in both directions. That two-way flow means an insight does not just sit on a dashboard. A finding about which audience converts best can shape the next campaign's targeting without a manual export. The product is licensed separately from other Marketing Cloud tools, and pricing usually scales with connectors, data volume, and user count, so the integration value has to be weighed against cost. Looking forward, Salesforce has introduced Marketing Intelligence in Marketing Cloud Next, which carries these ideas into a newer architecture built on data pipelines, Data 360, Tableau, and Agentforce. The Datorama-era product remains supported, but teams planning new investments should understand where the roadmap is heading.
Standing up Marketing Cloud Intelligence
You do not create a single Marketing Cloud Intelligence record. You stand up the platform by connecting sources, harmonizing the data, and publishing dashboards. This is the high-level path an admin or analyst follows; the exact screens depend on your edition and which successor product you are on.
- Connect your data sources
Use an API Connector for sources with a native integration, TotalConnect for file-based feeds that need mapping, or LiteConnect for lighter uploads. Each connection becomes one or more typed data streams.
- Harmonize the incoming fields
In the Harmonization Center, set naming conventions and classification rules so each source's spend, clicks, conversions, and revenue map to the same dimensions and measurements.
- Choose an attribution model
Decide between first-touch, last-touch, linear, time-decay, or position-based before building ROI views, since the model shapes every budget conversation downstream.
- Build a curated set of dashboards
Assemble widgets into dashboard pages and collections, starting from the template library. Keep one focused view per audience rather than a sprawling gallery.
Pulls data directly from a source platform on a schedule; best for sources with a native integration.
The marketer-built method for file-based feeds; automates ingestion, cleansing, and mapping for CSVs and reports with no native connector.
The no-code area where naming conventions, classification, and field mapping align every feed to the shared data model.
The rule that distributes conversion credit across touchpoints; pick first-touch, last-touch, linear, time-decay, or position-based.
- Harmonization, not dashboard authoring, is the slow part of a rollout. Budget time for the data plumbing.
- Datorama and Marketing Cloud Intelligence name the same product; older docs and labels may still say Datorama.
- The platform is licensed separately from other Marketing Cloud products, and cost scales with connectors, data volume, and users.
- Salesforce now offers Marketing Intelligence in Marketing Cloud Next; confirm which product your org is provisioning before you design.
Prefer this walkthrough as its own page? How to Marketing Cloud Intelligence in Salesforce, step by step
Trust & references
Cross-checked against the following references.
Straight from the source - Salesforce's reference material on Marketing Cloud Intelligence.
Hands-on resources to go deeper on Marketing Cloud Intelligence.
About the Author
Dipojjal Chakrabarti is a B2C Solution Architect with 29 Salesforce certifications and over 13 years in the Salesforce ecosystem. He runs salesforcedictionary.com to help admins, developers, architects, and cert/interview candidates sharpen their fundamentals. More about Dipojjal.
Test your knowledge
Q1. What is Marketing Cloud Intelligence, formerly known as Datorama?
Q2. What core problem does Marketing Cloud Intelligence solve for multi-channel marketing teams?
Q3. How does Marketing Cloud Intelligence connect to the many data sources in a marketing stack?
Discussion
Loading discussion…