Marketers are now operating in a world where data is more important than ever and those that can access and use it effectively can gain a significant advantage over their competitors. With the right data Marketers can more effectively and efficiently target leads and communicate with existing customers consistently across all of the channels that they use to increase sales and reduce churn. This makes the selection of the data platform (or platforms) to use as its source of truth one of the most important MarTech decisions facing Marketers these days.
Unfortunately there is not a single right answer to the question of “which data platform should I use for my marketing program?” aside from “it depends”. This is not only due to the plethora of data platforms available on the market or already present within an organisation but on the nature of the organisation itself. The rest of this post attempts to outline the data platforms most commonly used in modern marketing organisations and in which environments they are best suited.
Data Platform: CRM
Most organisations will have some form of Customer Relationship Management (CRM) solution in place. It is where all the details about an organisation’s customers are stored and managed. They are usually integrated with an organisation’s billing, phone, booking, fulfilment and email systems so that the status of each customer and the details of all of their interactions with the organisation can be kept up to date. These systems are usually used by customer service and sales staff but because they are usually an organisation’s system of record for their customer data, marketing often want to be able to use it as a source of their campaign targeting. Examples of common CRMs are Microsoft’s Dynamics, SAP CRM and Salesforce CRM.
- Excellent at recording and managing customer relationships and interactions.
- Highly flexible and extensible.
- A vital tool for Sales and Customer Service.
- Can be very hard to extract data from them for Marketing purposes as they tend to have complex data models and “dead” columns and tables that get left behind as the business evolves, which makes them hard to navigate.
- They are not optimised for querying or processing large amounts of data, especially not in real time.
Best Suited To:
- Should only be used as marketing’s single customer view where the number of customers is low (<100,000), all of the data required can be stored in the CRM and there aren’t multiple other data sources marketing requires.
- Also avoid if real or near real time data access is required.
Data Platform: DMP
Data Management Platforms (DMP) have a much narrower purpose than their name suggests. They are designed to store details of the visitors to an organisation’s online properties. The details that they store are usually derived from online behaviour so that it becomes much faster to determine which visitors have performed a particular action on a site. Their visitor profiles can be augmented with offline data, usually from a CRM or CDP, but as they are usually SaaS platforms they seldom let Personally Identifiable Information (PII) to be stored in them. Their main use is as a source of targeting data for paid media and remarketing data and for website personalisation. Examples of common DMPs are Adobe Audience Manager and Oracle BlueKai.
- Optimised for real time data insertion and query of online visitor profile data.
- Can be good at deriving new profile data from online and offline behavior.
- Should not be used to store Personally Identifiable Information (PII).
- Only a small proportion of the profiles it stores will be customers so most of the data may be irrelevant for direct communication programs.
- Inflexible data model so it can be hard to bring other data sources into it.
- They integrate primarily with paid media and web personalisation platforms, so may not be able to integrate with email and other direct communication platforms.
Best Suited To:
- Businesses that only transact online and so have no offline data, can use a DMP as a single customer view as long as they don’t do direct communication marketing, unless the email platform contains the PII and integrates well with the DMP.
- Not for regulated industries as DMPs tend to be globally distributed SaaS for optimal performance.
Data Platform: Analytics
In this context an Analytics platform is one that is used to collate website visitor events and metrics into a single place and that provides a User Interface that allows insights and reports on the performance and effectiveness of monitored websites can be generated and shared. It is usually possible to augment the data captured by these platforms with offline data, usually from a CRM or CDP, but as they are usually SaaS platforms they seldom let Personally Identifiable Information (PII) to be stored in them. The most common analytics platforms in use today are Google Analytics and Adobe Analytics.
- Perfect platforms for tracking online user behaviour and measuring the performance of online properties and digital marketing initiatives.
- Excellent real time data insertion capabilities.
- Should not be used to store Personally Identifiable Information.
- Designed for reporting and visualisation so it can be hard to use the data it stores in other marketing platforms, especially at scale.
- Very inflexible data model so it can be hard to bring other data sources into it. There are usually limits on how custom data can be stored in them.
- Optimised for real time data ingestion over API, which means other data sources are usually harder to ingest.
- Free platforms have data limits after which sampling occurs, lowering data resolution.
Best Suited To:
- Should never be used as marketing’s single customer view unless the business is very small (<10,000 customers) and only operates online.
- Not for regulated industries as they tend to be globally distributed SaaS for optimal performance.
Data Platform: CDP
Customer Data Platforms (CDPs) are a relatively new entrant to the world of Marketing data systems but have quickly become something that most organisations are either deploying or are actively looking for. CDPs collate all of the customer data that an organisation has into a single place where it can be easily segmented and shared with marketing platforms to drive their activity. They ensure consistency of targeting and communication with customers across all channels and removes the need for manual data manipulation and extraction to be performed for each campaign and channel. Their single customer view can be easily extended and augmented with online and second and third party data to create a much richer data set to enable a wider range of targeting to be performed. An example of a pure CDPon the market today is the n3 Hub CDP.
- Purpose built for the creation of a single customer view for marketing from multiple data sources for sharing with other marketing platforms.
- Support a wide range of data ingestion and sharing methods.
- Selecting the right platform for an environment can be a time intensive and confusing process given the number of platforms available.
- Emerging segment so not all vendors will survive.Local support may be hard to find for platforms, which may make the platform hard to generate benefit from.
Best Suited To:
- Appropriate for all types of organisations due to the range of available platforms.
- Non SaaS CDPs are ideal for regulated industries.
Data Platform: Data Lake/ Data Mart
A Data Lake or Mart is usually a general-purpose data platform where that an organisation puts all of its data so that it is accessible in one place to reduce the need for data analysts to have to interrogate a number of different systems to get the data that they require. These are usually not “platforms” but off the shelf databases/platforms coupled with some form of integration or ETL software to feed it with data. They are usually characterised by the storage of the source data in the same format and structure that it was stored in within its source system. Examples of common data lake technologies are Apache Hadoop, Teradata and GreenPlum.
Each of these platforms has its own strengths and weaknesses that makes each suited to a particular set of circumstances. The following table provides a broad guide of where each different type of platform is best suited to provide marketing its single customer view and what its strengths and weaknesses are:
- Highly flexible and customisable.
- Can be tuned to match a company’s requirements exactly.
- Generally hard to extract data from as it usually requires a detailed understanding of the source system data models.
- Can be very expensive to build, maintain and operate as they usually require highly skilled operators and administrators.
- Are only extended when required, so new features can take a long time to add.
- Generally is a shared platform that is used for more than just marketing, so marketing may not always get the level of access that they would like.
Best Suited To:
- In environments where development capability is readily available and marketing has dedicated data analysts to extract the data when required.
In summary, using the table above it should be easy to select the right type of platform for storing your marketing single customer view and from there the exact platform of the ideal type can be found. But overall in most cases the best choice for this purpose is a Customer Data Platform (CDP) as they are specifically created for this purpose. There are only a narrow set of situations where an alternative type of platform could be used and even then it would be hard to justify not using a CDP in the absence of a very specific requirement or requirements.