White paper

Modern
Analytics, AI,
and Governance
at Scale
Learn how a strategic framework for data is
the foundation for AI innovation
 Bring on the era of AI with Microsoft 2

03 / 09 /
Executive summary Microsoft Fabric powers MA2G
11 Enterprise Data Governance
14 Microsoft Purview provides a unified
data governance solution

04 / 15
19
Data Management Foundation
Domains and Data Products
When your data is siloed,
your organization is siloed
22 /
Copilots reduce the heavy lifting

06 /
What will it take to make
your business AI-ready?
23 /
LLM capabilities power your
generative AI applications

24 /
Get ready for the era of AI
 Bring on the era of AI with Microsoft 3

Executive summary Data has long been the linchpin for

making digital transformations possible.
Now, after many years into this digital
future, we know simply having data isn’t
enough. Organizations need a system to
unlock the value of their data, to power
analytics and artificial intelligence (AI) that
help sharpen their competitive edge. They
also need to adapt their culture, people,
and processes so the organization as a
whole is maximizing data.

Based on hundreds of engagements

and conversations with customers,
Microsoft has developed a framework
for organizations to adopt a unified
analytics and AI ecosystem. Known as
Modern Analytics, AI, and Governance at
Scale (MA2G), this framework addresses
hurdles like siloed data, poor governance,
and manual data management. Built
on Microsoft Fabric, MA2G helps
organizations make sure their data
is AI-ready and allows business units
to find relevant data assets, without
compromising enterprise requirements.

Keep reading to learn more about MA2G

and what it can do for your organization.
 Bring on the era of AI with Microsoft 4

When your data is siloed,

your organization is siloed
In most cases, organizations struggle to implement an end-to-end analytics and AI ecosystem,
not because of the technology, but because they do not plan for, nor address, the role culture,
people, and processes play in bringing about digital transformation. In fact, the MIT Technology
Review says nearly every time—92%—an organization struggles with data it links back to an issue
with the company’s data strategy, data governance, and/or data management.1

This is because many organizations approach a data challenge with a technology solution—but
that only solves a fraction of the problem. Organizations need a wider and deeper focus on
solutions that drive cultural changes and align people and processes with technology.

Throughout hundreds of engagements with organizations worldwide to help them become data
driven, Microsoft has seen the following top challenges over and over. Here are three common
problems associated with culture, people, and processes that impede a unified analytics and
AI ecosystem.

Lack of data strategy leads to a siloed ecosystem

After years of implementing different analytics projects in the cloud, organizations continue to
build out their ecosystem in a reactive, piecemeal way. Without a well-defined data strategy,
solutions become siloed and technical debt increases, which stands in the way of bigger data
and analytics innovation. The lack of an analytics foundation for all data also inhibits a thriving
ecosystem.

For example, imagine a data warehouse migration that lands data in proprietary data formats.
At the same time, Internet of Things (IoT) data is streamed into a data lake store. Data from these
two separate projects lands in separate data stores, causing siloed data. To become data driven,
the entire organization must be able to build meaningful insights from both sources, regardless of
the boundaries between business unit, so they can access data of all attributes and types.

1
Building a high-performance data and AI organization, MIT Technology Review, April 2021
 Bring on the era of AI with Microsoft 5

Poor governance prevents democratized data

Data governance—which includes a clear set of policies, processes, and controls—is critical for
organizations to find, manage, and consume data across the business. A lack of data governance
and an incomplete understanding of the data can stymie analytics projects. Failure points can run
the gamut. Some projects can’t access data fast enough, others can’t integrate data due to the
lack of data relationships, and others require additional data engineering work to be completed
before being able to train machine learning models.

All these problems can be avoided if the organizations have enterprise data governance that
provides the inventory and context of all data, automated processes to streamline workflows, and
policies that automatically manage data access. The goal is to implement robust data governance
and data management that enables different analytics projects for different business units.

Manual data management slows time to insights

For data to be consumable, there’s a ton of ingestion and data engineering work that must
happen. It must be cleaned with approved data quality, integrated to reveal new business insights,
and aggregated to become a proper data product. Most organizations tend to build manual data
engineering workflows on a case-by-case basis aligned to specific projects.

However, this creates fragmented data engineering solutions that become harder to maintain as
they grow to include thousands on top of thousands of pipelines. The worst part is, most data
engineering tasks are manual. Many organizations use their best people to perform these manual
tasks when it should be a process change instead. By implementing proper data management
processes with automation, organizations can reassign data engineers to more meaningful work
of business data modelling, data aggregations, and calculations.
 Bring on the era of AI with Microsoft 6

What will it take to make

your business AI-ready?
From chatbots supporting service interactions to generative AI creating sales content, AI is
everywhere. And demand for these capabilities is soaring. Just one year after ChatGPT launched,
54% of companies had already implemented generative AI in some area of their business.2

However, a paradox still exists when it comes to AI. While 78% of executives agree that AI is a
top business priority,3 they also recognize that data problems will likely stand in their way of
achieving their AI goals. This is because many organizations see AI and analytics as initiatives that
can be adopted by investing in technology solutions—but that’s just a small piece of the puzzle.
Organizations that successfully deploy AI and analytics also change processes, adapt their culture,
and support their people to use these new capabilities in effective ways.

AI priorities for executives

Unifying our data

Scaling AI/ML use cases
platform for analytics
to create business value
78%
agree
is a top priority 68% agree
and AI is crucial to our
enterprise data strategy
Disagree 8%
Disagree 12%
Neutral 14%
Neutral 20%

We favor a multi-cloud Data problems are the

approach as a flexible most likely factor to
72%
agree
foundation for AI/ML 72% agree
jeopardize our AI/ML goals
Disagree 12% Disagree 9%
Neutral 16% Neutral 19%

2
2024 AI Business Predictions, PWC
3
CIO perspectives on generative AI, MIT Technology Review, July 2023
 Bring on the era of AI with Microsoft 7

Modern Analytics, AI, and Governance at Scale

The question every leader is asking themselves right now is: How can my organization seize the
full potential of AI, while safeguarding my business, data, and employees?

After hundreds of engagements with organizations worldwide, Microsoft has developed

a framework to not only adopt AI and analytics, but also use these technologies to their fullest
extent. It’s known as Modern Analytics, AI, and Governance at Scale (MA2G) and it outlines the
foundational elements organizations need to ensure their data is AI-ready.

Get started with MA2G

MA2G was created to give organizations a framework for understanding what it takes to deploy a
successful end-to-end analytics and AI platform. It’s built on the following three pillars, which
describe the people, process, and culture considerations organizations need to keep in mind for
AI and analytics to take root.

Enterprise Data Governance includes the set of policies and practices used to discover,
describe, and manage data to accelerate responsible data democratization. Data
governance ties together the data and analytics stack and automates data operations,
such as cataloging, classification, creating lineage, and applying security through policies.
Without it, organizations limit their ability to innovate and unlock new insights.
• Data Management Services • Data Order Service
• Governance • Rapid Access to Data
• Quality • On Premises or Azure
• Policy
• Lineage
• Classification
• Catalogue
 Bring on the era of AI with Microsoft 8

Data Management Foundation involves the practices and processes that help you
create efficiencies with ingesting, storing, protecting, and ultimately serving data to
different domains in the organization.
• Self Service - Automation • Automated Data Operations
• Domain Provisioning • Data Virtualization: Shortcuts and Mirroring
• Workspace Provisioning • Ingestion at Scale Solution
• Data Onboarding • Data Engineering Acceleration
• Automation
• Lakehouse and open data formats

Domains and Data Products describes the environments and services that enable
your business units to fully use their data. Allowing departments to self-serve data and
analytics enables non-technical users to access, analyze, and build data insights or data
products on their own.
• Federation • AI Copilots
• Autonomous Business Units • Empowering Data Practitioners
• Domains, Workspaces, Capacities • Accelerating Report Creation
• Data Sharing / Collaboration • Enable Data Exploration
• AI Guided Insights
• Data Integration

Together, these three solution pillars combine to help organizations achieve MA2G. By first
setting business priorities as the guiding North Star, then implementing aspects from each
solution pillar, your organization can shift into a whole new paradigm of doing business that
empowers everyone to work toward a common goal fueled by data.
 Bring on the era of AI with Microsoft 9

Microsoft Fabric powers MA2G

Based on the Microsoft Intelligent Data Platform, the MA2G architecture brings together the
best of Azure into a unified SaaS solution powered by Microsoft Fabric and Microsoft Purview,
and built on an open and governed data lakehouse. It includes support for Delta Parquet, and
emerging open standards like Hudi and Iceberg, to accommodate diverse investments.

Microsoft Fabric

Enterprise Governance,
Security, and Compliance
IT Finance Marketing
Fabric domain Fabric domain Fabric domain

OneLake Security

OneLake
Shortcuts Fabric storage Mirroring
Existing data lakes (Delta/Parquet) for proprietary data stores
Azure, AWS, Google

Innovation HR Operations
Fabric domain Fabric domain Fabric domain

A hub for all your data

Most organizations have analytics systems that are a labyrinth of specialized and disconnected
services. As businesses aim to adopt modern data capabilities, it’s increasingly important to
integrate all disparate data into a unified source. OneLake, incorporated into Fabric, unifies all
your data in a single, accessible location for comprehensive data management. Plus, it’s backed
by built-in security and governance, ensuring the protection of all your data.
 Bring on the era of AI with Microsoft 10

Simplified data preparation

For AI and machine learning models to be as accurate as possible, they must be built with clean
data and in a semi-structured way. Fabric simplifies data preparation and transformation so you
can quickly get your data ready for custom AI deployments. Fabric includes analytics experiences
such as Data Engineering, Data Factory, and Data Warehouse built into its SaaS-based data
platform so that different teams can all find the tools they need and work together.

AI-ready data that’s accessible to domains

Once you have a cleaned, robust dataset, you can start building AI and generative AI experiences
on top of your data. Fabric allows companies to organize data into domains where data
consumers can filter and find content they need for AI or analytics. It also enables federated
governance so that each business unit or department can define its own rules and restrictions.
 Bring on the era of AI with Microsoft 11

Enterprise Data Four ways to improve

enterprise data governance

Governance Create metadata: Well-governed

data includes metadata about its
Governance is essential to every lineage, profile, quality, business
organization—regardless of the framework context, and classification so it’s
or solution implemented—because it lays trusted and useful for all end
the bedrock for responsibly democratizing parties.
data. Data governance translates your data
strategy into data ownership, rules, and Map data assets: With a data
policies that improve data discoverability, map, data consumers can easily
confidence, security, compliance, and and visually inspect all data
operational efficiencies. assets across all domains whether
it’s physically stored on premises
Over the years, data access and use or in the cloud.
has spread to all corners of companies
with individuals making critical business Catalog data assets: Data
decisions that affect organizations, consumers use a data catalog
customers, and shareholders. You might to find all datasets, see if they’re
see the finance department visualizing complete, and learn the relevant
billions of rows of risk data, while analysts business context associated with
in marketing identify customers for a new the data asset.
product. The range of disparate uses,
abuses, and copies of data is leading Automate governance and
to widespread confusion and risk. Data security: Instead of manually
governance provides the glue that ties implementing data security for
together all the data in the analytics each data asset—which is prone
stack and ensures the right data is easily to errors and highly inefficient—
accessible by the right people. automated governance keeps
data secure while still ensuring
access of data to the right users.
 Bring on the era of AI with Microsoft 12

Within Fabric, you can access a Purview Democratize data through

Hub Insights dashboard that shows you: self-service features
To truly democratize data, organizations
• Overview report: See an overview
need to implement a solution where data
of distribution, use of endorsement,
is discoverable through a data catalog
and sensitivity labeling.
and domain users can request access
• Endorsement report: Drill down without opening a ticket. Microsoft
and analyze distribution and use of Fabric and Microsoft Purview make this
endorsement. possible. Domain teams (data analytics,
data products developers, data owners,
• Sensitivity report: Drill down and
etc.) can browse a data catalog in Fabric or
analyze distribution and use of
Purview to discover new data relevant to
sensitivity labeling.
their use case.
• Inventory report: Get details
about labeled and endorsed items, Fabric displays information about the
then narrow down the results data, such as metadata with classification,
through date ranges and filters for lineage, business terms, related assets,
workspace, item type, etc. and data owners to help domains make
decisions about which sources to explore.
• Items page: Find insights about the Once they select data sources, Fabric
distribution of items throughout makes it easy to grant access to domains
your organization and endorsement if they do not have pre-established
coverage. permission. By combining Fabric and
• Sensitivity page: Discover Purview, you can govern your entire estate
insights about sensitivity labeling and lineage of data. From data source
throughout your entire organization. down to Power BI reports, Purview and
Fabric work together seamlessly so you
can store, analyze, and govern your data
without piecing together services from
multiple vendors.
 Bring on the era of AI with Microsoft 13

Govern and protect your data with integrated services

Governance is not only crucial to empowering your data
consumers, but also gaining their trust while meeting security and Microsoft Purview
compliance requirements. Fabric includes a set of capabilities that provides a unified
help you know, protect, manage, and monitor your organization’s data governance
sensitive information. It works in tandem with Purview to help solution to help
govern and manage your entire data estate. Fabric governance manage and govern
and compliance is tightly integrated with Purview, allowing you your data estate.
to create a holistic, up-to-date map of your data landscape Learn more >>
with automated data discovery, sensitive data classification, and
end-to-end data lineage.
 Bring on the era of AI with Microsoft 14

Microsoft Purview provides a unified

data governance solution
Microsoft Purview allows you to create a holistic, up-to-date map of your data landscape with
automated data discovery, sensitive data classification, and end-to-end data lineage. Enable
business units to access valuable, trustworthy data management, and take advantage of the
following capabilities:

• Support for multi-cloud data estates: Automatically scan and catalog all data assets—
including machine learning models and Power BI reports—across the organization, whether
they’re on premises, in Azure, or running on other public clouds.

• Governance experience: Develop clear role definitions for administrators, domain creators,
data health owners, and data health readers.

• Business-friendly terminology: Assign language that follows the data governance experience
through data products, domains, quality assessments, and reports.

• Data scan and search: Find the data you need across your entire estate and profile data at the
source to indicate attributes like min, max, average, and thresholds.

• Data quality scores: Generate data quality scores once rules and policies are applied, giving
you insights into your data quality relative to your business rules.

• Metadata analysis: Capture metadata and data lineage to help personas to decide if data is
usable, then use profiling or data quality scans for recommendations.

• Data health controls: Ensure your rules and indicators reflect the unique standards of your
organization with a set of cloud data management controls.

• Summarized insights: Showcase the overall health of your governed data estate with built-in
data governance reports.

• Pre-built integrations: Extend the value of Purview with integrations for solutions related to
master data management and data lineage.
 Bring on the era of AI with Microsoft 15

Data Management Foundation

The purpose of data management is to ensure that data is properly collected, stored, processed,
analyzed, and used in a secure and efficient manner to support an organization’s goals and
objectives. Yet, less than one-quarter of organizations have a consistent, global data management
strategy in place.

Implementing automation, frameworks, and services can help organizations bolster their data
management practices. With a data management foundation that uses Microsoft Fabric and
OneLake, you gain both an open and governed data lakehouse for storing data, as well as
automated data virtualization that efficiently sends data to domains without overburdening
IT teams.

Microsoft Fabric

Admin Console: Fabric Domains

Domain Provisioning IT Finance Marketing

Workspace Provisioning Innovation Operations

Data Onboarding
OneLake
Shortcuts allow instant linking of data
already in Azure and other clouds, Shortcuts Fabric Storage Mirroring
Existing Delta and for proprietary
without any data duplication and
data lakes Parquet data stores
movement.
Azure, AWS, Google
Mirroring is a feature that offers
continuous and seamless access to and
replication of data from database or
data warehouse with no ETL required.
 Bring on the era of AI with Microsoft 16

Every workload works with OneCopy and open formats

All compute engines—including Spark, T-SQL, and KQL—automatically store their data in OneLake
in a single common format. Once data is stored in the lake, it’s accessible to all engines and does
not have to be imported or exported. Each compute type has been fully optimized to work with
Delta and Parquet as their native format and a shared.

An easier way to onboard data

Fabric helps eliminate data pipelines through capabilities such as shortcuts and mirroring, which
bring your data into one platform, without the legwork. In addition, you can also use partner
solutions that work with Fabric connectors to move data between stores.

Shortcuts: OneLake shortcuts let you easily onboard data by instantly linking data that already
exists in Azure or other clouds through a unified namespace. This eliminates data duplication or
movement, reducing latency associated with data copies and staging.

• A shortcut is a symbolic link which points from one data location to another

• Shortcuts make data from a warehouse part of your lakehouse

• You can consolidate data across items or workspaces without changing the data ownership

• Data can be reused multiple times without data duplication

• Existing ADLS Gen2 storage accounts and Amazon S3 buckets can be managed externally to
Fabric and Microsoft while still being virtualized into OneLake with shortcuts

• All data is mapped to a unified namespace and can be accessed using the same APIs, including
the ADLS Gen2 DFS APIs
 Bring on the era of AI with Microsoft 17

Industry solutions: Fabric includes pre-built, industry-specific solutions that help organizations
integrate data from different sources and use rich analytics. Data solutions combine data
integration services and, in some cases, machine learning support, so organizations can face
industry-specific data challenges. These solutions include retail, healthcare, sustainability,
and more.

Mirroring: Fabric offers a mirroring feature that provides continuous and seamless access to—
and replication of—data from databases or data warehouses, without ETL. Any database can be
accessed and managed via Fabric without having to switch database clients. By just providing
connection details, your database is instantly available in Fabric as a Mirrored database.

• A full editing experience of the source database is available for the Mirrored database

• Data is replicated into OneLake in Delta format and kept up to date in near real time

• All the Fabric experiences instantly work with the OneLake replica

• Analysts and data scientists can work with real-time data

• The replica protects operational databases from analytical queries

Automated data management

For any sources that don’t have mirroring or shortcuts, you can still automatically ingest data
into the ecosystem. Automated data services and templates can help improve efficiency around
data ingestion, standardization, quality, metadata registration, and access provisioning. These
enterprise-level capabilities allow data foundation teams to minimize repetitive, manual work, and
they create a foundation in OneLake for domain teams to self-serve data.
 Bring on the era of AI with Microsoft 18

Data-agnostic ingestion service

• Pull/push
On premises Azure • Format agnostic
or other Data-agnostic ingestion service • Data agnostic
cloud provider • Metadata driven

Data standardization service

Pull • Analytics format conversion
Data sources Bronze Silver Gold
• Versioning
• Merging
• PII handling
Push
Generic
Data pipeline Drop Zone • Data quality
Confidential • Common Data Model
• Master data unification
• Synchronous processing
• GDPR

Data-agnostic ingestion
Automatically ingest data regardless of its attribute, format, and the domain it belongs to.
Organizations can push or pull data from different sources then process it. Metadata-ingestion
frameworks or Kafka-based solutions are sample solutions that can be implemented to automate
this process.

Data standardization
As your data gets ingested, you can standardize it through processes such as format conversions,
versioning, merging, PII handing, and master data management. Use Apache Spark notebooks
within Fabric to quickly implement data standardization practices. Additional services related
to data quality management address issues such as deduplication, threshold identification, and
alignment with master data. Without proper checks on data quality, you run the risk of slowing
down time to insights.

Metadata registration and access provisioning

With your data in OneLake, the next step is registering the new data assets in your data catalog
so they’re instantly discoverable. As a safety net, data governance scheduled scanning of the
data hub should register these new assets. Once data is added, another automated service can
provision access according to the data classification of data being ingested.
 Bring on the era of AI with Microsoft 19

Domains and Data Products

Organizations are shifting their approach from running a center of excellence—where everything
is centrally controlled—to using federated domains to provide departments more control and
autonomy.

With Microsoft Fabric, every data workload is available in

one SaaS experience for all personas in the organization.

Domains and workspaces makes collaboration happen

Domain provisioning: Fabric allows you to group data into a domain so that users can find
the resources they need that are relevant to their field. For instance, you can create domains
by business department such as HR or finance, allowing those teams to manage their data
according to their specific regulations, restrictions, and needs.

Workspace provisioning: You can create a workspace to collaborate with teammates in your
domain and create collections of items such as lakehouses, warehouses, and reports.

Microsoft Fabric experiences promote org-wide collaboration

Because it’s built on a single, unified SaaS platform, Fabric brings together Power BI, Azure
Synapse, and Azure Data Factory so data teams can collaborate in a single workspace, on
the same copy of data. Fabric provides each domain with core experiences designed to work
together seamlessly.
 Bring on the era of AI with Microsoft 20

Each experience is tailored to a specific persona and a specific task, allowing different domains to
find the tools they need to create their own data products.

Microsoft Fabric Domains

Data Factory Synapse Data Synapse Synapse Data Synapse Real- Power BI Data Activator
Engineering Data Science Warehouse Time Analytics

Workspace 1 Workspace 2 Workspace 3

Fabric Capacity Fabric Capacity

OneLake

Data Factory offers a modern data integration experience to ingest, prepare, and
transform data from a rich set of data sources. Data Factory brings Fast Copy capabilities
to both dataflows and data pipelines so you can move data between your lakehouse and
data warehouse in Fabric at blazing speed.

Synapse Data Engineering provides a world class Spark platform with great authoring
experiences, enabling data engineers to perform large scale data transformation and
democratize data through the lakehouse. The Spark integration with Data Factory also
enables notebooks and Spark jobs to be scheduled and orchestrated.

Synapse Data Science allows you to build, deploy, and operationalize machine learning
models within your Fabric experience. It integrates with Azure Machine Learning to
provide built-in experiment tracking and model registry.
 Bring on the era of AI with Microsoft 21

Synapse Data Warehouse provides industry-leading SQL performance and scale. It fully
separates compute from storage, allowing independent scaling of both the components.
Additionally, it natively stores data in the open Delta Lake format.

Synapse Real-Time Analytics gives you a way to focus and scale up your analytics
solution while democratizing data for both citizen data scientists and advanced data
engineers. As a fully managed big data analytics platform, Real-Time Analytics utilizes
a query language and engine so you can search structured, semi-structured, and
unstructured data.

Power BI provides business owners the ability to access all their data in Fabric quickly
and intuitively to make better decisions with data. This experience allows organizations
to turn unrelated data sources into coherent, visually immersive, and interactive insights.

Data Activator monitors data in Power BI reports and automatically takes actions when
certain patterns or conditions are detected. This allows you to build a digital nervous
system that acts across all your data, at scale and in a timely manner.
 Bring on the era of AI with Microsoft 22

Copilots reduce the heavy lifting

Microsoft Fabric includes several copilots that act as interactive aides, lightening the load on
engineers, scientists, and analysts so they can expedite the journey from raw data to meaningful
insights. From data preparation to report building, Copilot and other generative AI features offer
new ways to analyze data, generate code, and create visualizations in Fabric and Power BI.

Copilot for Data Science and Data Engineering provides intelligent code completion,
automates routine tasks, and supplies industry-standard code templates to facilitate
tasks like data enrichment and the creation of analytical models. Copilot offers
contextual code suggestions and prompts that adapt to specific tasks, helping you code
more effectively and with greater ease.

Copilot for Data Factory supports both citizen and professional data wranglers in
streamlining their workflow. It provides intelligent code to transform data, as well as
code explanations to help you understand complex tasks.

Copilot for Power BI allows you to create Power BI reports automatically. You can
generate summaries of existing reports or ask for suggestions on which reports to create
based on your data. Prompts like “Create a page to examine next month’s forecast” yield
visualizations that help you spot trends and patterns quickly.
 Bring on the era of AI with Microsoft 23

LLM capabilities power your

generative AI applications
As a platform for analytics and AI, Microsoft Fabric is well suited to support the use of large
language models (LLMs) for the creation of generative AI applications. Fabric and SynapseML offer
unique LLM capabilities so you can build solutions that can handle question-and-answer tasks or
document summaries, for example.

Extract insights from unstructured data: Use Fabric to tap into information stored in
unstructured documents like PDFs. You can load PDF documents into a Spark DataFrame, read the
documents using the Azure AI Document Intelligence in Azure AI Services, and use SynapseML to
split the documents into chunks.

Integrate Azure OpenAI: Apply LLMs at scale by integrating Azure OpenAI Service
and SynapseML. Azure OpenAI can be used to solve natural language tasks by prompting
the completion API. Through SynapseML, you can use Apache Spark distributed computing
framework to easily process millions of prompts.

Generate embeddings: Connect Azure OpenAI Service and use SynapseML to generate
embeddings in a distributed manner that allows you to efficiently process large volumes of data.
You can also store the embeddings in a vector store using Azure AI Search and search the vector
store to answer users’ questions.
 Bring on the era of AI with Microsoft 24

As organizations forge ahead in the era

Get ready for of AI, clean data from well-managed and

highly integrated analytics systems is

the era of AI
critical. Frameworks like MA2G make sure
your systems support governance, data
management, and domains, allowing
organization to create customized AI and
analytics experiences. Through Microsoft
Fabric, all the data and analytics tools you
need are available in one, end-to-end
platform.

Interested in learning more

about MA2G and Fabric?

Reach out to an Azure sales

specialist or your Microsoft
sales representative for
best practices on analytics,
help getting started with Fabric, and more.
Or ask about visiting a local Executive
Briefing Center or Microsoft Training Center.

©2024 Microsoft Corporation. All rights reserved. This

document is provided “as-is.” Information and views
expressed in this document, including URL and other
Internet website references, may change without notice.
You bear the risk of using it. This document does not
provide you with any legal rights to any intellectual
property in any Microsoft product. You may copy and
use this document for your internal, reference purposes.