Choosing a CDP: do you need one and which fits your business

The best customer data platform is the one that unifies the data you actually have, activates it in the channels you actually use and handles personal data in line with GDPR. You need one when your customer data is spread across systems that don’t talk to each other, and that gap is holding back personalisation, reporting, or service. Size matters less than you might think; the number of data sources and channels matters more.

This guide helps you decide whether you need a CDP, which type fits and how to evaluate vendors without being dazzled by demos. For what a CDP is and how it differs from a CRM and DMP, start with our guide to customer data platforms.

Do you need a customer data platform?

Probably, if several of these sound familiar:

• The same customer looks different in your shop, your email platform and your CRM.
• Building a segment means exporting lists from several systems and matching them by hand.
• You cannot act on data from one channel in another, such as using in-store purchases in email.
• Personalisation stops at a first name because behavioural data is out of reach.
• Nobody can say with confidence how many customers you have, or how valuable they are.
• Consent and preferences are stored in more than one place and do not always match.

You probably do not need one yet if you run one webshop and one email platform that are well connected, and your list is small enough to manage there. Our article on what stops companies from using a CDP covers the common doubts.

Which businesses benefit most from a CDP?

Our articles on getting the most value out of ecommerce data and WhatsApp and CDP in retail show what this looks like in practice.

Can a small business use a CDP?

Yes, if the problem a CDP solves is one you actually have. Small businesses with a webshop, a point-of-sale system and an email platform often have exactly the fragmented data a CDP fixes. Modern CDPs are priced and packaged for smaller organisations too, often by number of profiles or data sources.

Start small. Connect the two or three most important sources and build a handful of segments that drive revenue, such as repeat buyers and lapsed customers. Expand once they prove their worth. If your email platform’s own database already gives you what you need, invest there first.

What types of CDP are there?

• All-in-one CDPs collect and unify data and also activate it, for example by sending email or personalising the website. They suit teams that want fewer tools.
• Data-focused CDPs collect and unify data, then send segments to other tools for activation. They suit organisations with established channel tools they want to keep.
• Warehouse-native or composable CDPs work on top of your own data warehouse. They suit organisations with a data team and a mature warehouse.
• CDP features within marketing suites offer unified profiles inside a broader platform. They suit teams that mainly activate through that suite.

None is better in general. The right type depends on your existing tools, your data maturity and who will run it day to day.

Which CDP features matter most?

• Data collection: ready-made connectors for your shop, CRM, email, website, point of sale and service tools, plus a way to send your own events.
• Identity resolution: how the platform matches records to one person, for example by email address, customer ID or login, and how it handles anonymous visitors who later identify themselves.
• Segmentation: building segments on behaviour, transactions and profile data without writing code, with segments that update themselves.
• Activation: getting segments and profiles into email, ads, website, WhatsApp and CRM, natively or through reliable exports.
• Prediction: optional, but useful: predicted lifetime value, purchase likelihood or churn risk.
• Consent and privacy: storing consent per purpose and channel, honouring opt-outs everywhere and supporting data subject requests.
• Data quality: deduplication, validation and visibility of where each field came from.
• Security and hosting: where data is stored, which certifications apply and who can access it.

Real-time processing is valuable for websites and high-volume ecommerce. For many B2B and smaller organisations, hourly or daily updates are enough.

How do you evaluate CDP vendors?

1. Define three use cases first. For example: win back lapsed customers, personalise the homepage by product interest, give service agents a full purchase history. Judge every vendor on those.
2. Map your data sources. List each system, what it holds, how often it changes and who owns it.
3. Shortlist on non-negotiables. Connectors for your key systems, hosting and data protection requirements, and the channels you need to activate.
4. Run a proof of concept with your own data. Connect two real sources and build one real segment. Demos with sample data hide the hard parts.
5. Ask the uncomfortable questions. How is pricing calculated as profiles and events grow? Which connectors are native and which need custom work? Who maintains them when a source system changes? Can you export all data and segments if you leave?

Can a CDP work with your existing marketing stack?

It should. A CDP is meant to sit between your systems, not replace them. It takes data in from your shop, CRM, website and service tools, and sends unified profiles and segments out to email, ads, website personalisation and analytics.

Usually you add a CDP rather than replace anything. The exception is when your current tools overlap heavily with the CDP’s activation features and you want fewer platforms to manage. Our article on starting with a customer data platform walks through the first integrations. Our guide on building a personalisation strategy with a CDP shows what to do with the data once it is connected.

Why do B2B companies use a CDP?

Because B2B buying involves several people over a long period, and their signals rarely end up in one place. Website visits sit in analytics, email engagement in the marketing platform, deals in the CRM and product usage in the product itself. A CDP brings them together per person and per account.

That gives marketing better segments, such as accounts showing renewed interest, and gives sales a fuller picture before a call. It also helps with consent: business contacts are still people under GDPR, and a CDP can keep their preferences consistent across every tool.

How long does implementation take, and what does it cost?

Both depend on the number of data sources, the quality of your data and how many channels you activate. A focused first phase with two or three sources and a few use cases can go live far sooner than a project that connects everything at once.

Cost usually depends on the number of profiles or events, the number of sources and destinations, any add-ons such as prediction, and implementation support. Ask vendors to price your expected volumes in two years’ time, not just today’s.

Which mistakes should you avoid?

• Buying features instead of outcomes. Start from use cases, not from a feature checklist.
• No data owner. Someone needs to own data quality and definitions, or the unified profile slowly decays.
• Importing messy data unchanged. A CDP unifies data; it does not make bad data good. Our guide to data management covers the groundwork.
• Forgetting consent. Decide how consent and preferences flow before you activate. Our article on turning GDPR into an opportunity with a CDP explains the approach.
• Connecting everything at once. Phased projects show value sooner and learn as they go.

Frequently asked questions

Spotler Activate: a CDP built for marketing

Spotler Activate is our customer data platform. It brings customer data from your shop, website, email and other systems into one profile, builds segments that update themselves and makes them available for your campaigns. Its predictive AI add-on groups customers by predicted lifetime value. Book a demo and bring your three use cases.