The complete martech guide to CRM, CDP, DMP and more

The modern retail martech stack explained: CDP, CRM, ESP, and data warehouse

The retail martech landscape has changed more in the last three years than in the previous decade. Third-party cookies are gone. AI-generated search is reshaping discovery. And the number of tools claiming to solve your data problems has grown faster than most teams' ability to evaluate them.

This guide cuts through the noise. It explains what each major platform in the retail martech stack actually does, where each one ends, and how they work together. If you're evaluating your current stack or building a business case for a new investment, this is the map.

The four platforms every retail marketer needs to understand

The core of a modern retail data stack is four platforms: a Customer Data Platform (CDP), a CRM, an ESP or marketing automation platform (MAP), and a data warehouse. Each does a fundamentally different job. The confusion, and the budget waste, happens when teams expect one platform to do what another was built for.

What is a Customer Data Platform (CDP)?

A Customer Data Platform ingests customer data from every source, including ecommerce, POS, loyalty, email, web, app, and resolves it into a single unified profile for each customer. That profile is always current, always available to marketing teams, and feeds every downstream activation channel.

The defining characteristic of a CDP is the single customer view: one record per person, regardless of how many systems they appear in. A customer who bought in-store last week and clicked an email yesterday is the same person in your CDP, even if your POS and ESP have no idea they're connected.

What a CDP is built for:

• Ingesting structured and unstructured data from any source
• Identity resolution and linking fragmented records into unified profiles
• Real-time segmentation across any data attribute
• Feeding segments to every activation channel: email, SMS, paid, in-store
• Enabling personalisation at the individual level

What a CDP is not:

• A replacement for your CRM (CDPs don't manage sales pipelines or service workflows)
• A replacement for your ESP (CDPs feed audiences to email tools; they don't send the emails)
• A data warehouse (CDPs are built for marketers, not data engineers)

The retail-specific distinction matters: CDPs built for retail understand the data model of physical and digital retail, including POS transactions, loyalty IDs, product catalogues, in-store events. Generic CDPs often require significant engineering to get to the same starting point. A retail CDP solution is built to handle omnichannel retail data without that overhead.

What is a CRM?

A CRM (Customer Relationship Management system) manages the history of interactions with known customers, including contact records, service tickets, sales pipeline stages, and account activity. It originated in B2B sales workflows and is designed for one-to-one relationship management.

What a CRM does well:

• Tracking individual customer service and sales interactions
• Managing pipelines for enterprise or high-value accounts
• Cross-channel contact history for service teams

Where a CRM falls short for retail marketers:

• Not designed for segment-level analysis across millions of customers
• Doesn't handle purchase history, web behaviour, or in-store data natively
• Limited real-time capabilities as data is often historical rather than live
• Not built to feed activation channels at scale

CDP vs CRM: what's the difference?

The clearest way to think about it: a CRM is a record of what you've said to customers. A CDP is a real-time intelligence layer across everything customers have done. A CDP consumes your CRM data and combines it with ecommerce, POS, loyalty, and behavioural data to create a richer view than either system has alone.

They're not alternatives and retail brands benefit from running both. The CRM handles service and high-value account management. The CDP enables segment-level marketing and real-time personalisation.

What is an ESP or marketing automation platform (MAP)?

An ESP (Email Service Provider) or marketing automation platform handles the delivery of communications to customers: email, SMS, push notifications. Platforms like Klaviyo, Braze, and Attentive sit in this category.

What an ESP/MAP does well:

• Building and sending email and SMS campaigns
• Automated flow triggers (welcome series, abandoned cart, post-purchase)
• Template management and deliverability

Where an ESP/MAP falls short:

• Its audience-building capability is limited to data within the platform
• It doesn't hold a unified customer record across channels
• It can't segment by in-store purchase behaviour, loyalty tier, or predictive attributes unless those are fed in from another system

Can a CDP replace Klaviyo or HubSpot?

No, and it's not designed to. A CDP feeds your ESP with better audiences and richer customer context. A customer segment built in a CDP, say, "high-LTV customers who haven't bought in 60 days", gets pushed to Klaviyo where the email is built and sent. The CDP handles the data intelligence. The ESP handles the communication.

The combination is significantly more powerful than either alone. The Lexer platform integrates natively with Klaviyo, Braze, Attentive, and most major ESPs, allowing you to activate your customer data across every channel your team already uses.

What is a data warehouse?

A data warehouse (Snowflake, BigQuery, Redshift) is a centralised repository for large volumes of structured data, designed for complex querying and reporting by data engineers and analysts. It's the technical infrastructure layer, not a marketing tool.

Do you need a data warehouse if you have a CDP?

They serve different purposes and are often used together. A data warehouse is built for historical analysis, complex SQL queries, and data engineering workflows. A CDP is built for real-time marketing activation, segmenting customers and sending audiences to channels. Most mid-market retailers find that a CDP handles their marketing data needs without requiring a warehouse, though enterprise retailers often run both.

The practical test: if your marketing team needs a data engineer to run a query before they can build a segment or send a campaign, you have a data warehouse problem. A CDP is designed to make that capability self-serve for marketers.

What does a modern retail data stack look like?

For a mid-market omnichannel retailer, the core stack typically looks like this:

CDP: the unified customer data layer, feeding every other system with accurate, real-time customer profiles

CRM: manages service and high-value customer relationships, with data flowing into the CDP for enrichment

ESP/MAP: handles email and SMS execution, receiving audience segments from the CDP

Analytics/reporting: either within the CDP or a dedicated BI tool, for campaign measurement and customer health reporting

POS and ecommerce: the transaction data sources that feed the CDP

The CDP sits in the middle of that stack because it's the integration point — it ingests from POS and ecommerce, enriches with third-party data, feeds segments to the ESP, and keeps customer profiles current across every system. That's the customer identity resolution capability that makes omnichannel personalisation operationally possible.

To see what this looks like in practice for a mid-market retailer, read how SHEIKE uses Lexer to power personalised marketing across channels.

What differentiates retail-focused CDPs from generic ones?

Retail has a specific data model that generic CDPs weren't designed for:

• Omnichannel identity resolution: linking in-store loyalty IDs to online behaviour to email addresses, without a loyalty programme or POS integration being an engineering project
• Transaction-level data at scale: millions of purchase records across multiple years, multiple stores, and multiple product categories
• Predictive models built for retail: CLV prediction, churn risk, next-best-category, and product affinity models require retail-specific training data, not generic ML models
• Store-level and staff-level attribution: in-store clienteling and associate-level marketing requires data granularity that general-purpose CDPs don't support natively

Generic CDPs are built for digital-first businesses with clean, structured web data. Retail, especially omnichannel retail with physical stores, is messier, richer, and requires a platform designed for that complexity.

‍

FAQs

What is a customer data platform (CDP)?

A CDP is software that collects customer data from every source, including purchases, browsing behaviour, email engagement, in-store transactions, and unifies it into a single profile per customer. That profile powers segmentation, personalisation, and marketing activation across all channels.

What is the difference between a CDP and a CRM?

A CRM manages service and sales interactions with individual customers. A CDP is an intelligence layer across all customer behaviour. It ingests CRM data alongside purchase history, web behaviour, and loyalty data to build a complete view. They complement each other rather than replacing each other.

Can a CDP replace Klaviyo or HubSpot?

No. A CDP feeds better audiences to your ESP (like Klaviyo) but doesn't replace it. The CDP handles data intelligence and segmentation; the ESP handles email and SMS delivery. The two work together and the CDP makes your ESP significantly more effective.

Do I need a data warehouse if I have a CDP?

Not necessarily. A CDP handles real-time segmentation and marketing activation, which covers most marketing team needs. A data warehouse handles historical analysis and complex engineering workflows. Many mid-market retailers find a CDP alone is sufficient; enterprise brands often run both for different use cases.