Does Your Brand Really Need A Customer Data Platform (CDP)

 Does Your Brand Really Need A Customer Data Platform (CDP)


My experience as a CDP Administrator for a large US retailer


If you are in Marketing then the acronym CDP will sound familiar to you because CDP is the new buzz word and synonymous with Personalization. 


Now that data is ubiquitous and a lot easier to access, personalization is the new trend in marketing and therefore a lot of companies are trying to reinvent themselves by offering their customers a personalized experience. This personalized experience is delivered either with the help of a SaaS Platform like Tealium AudienceStream or custom-built based on the complexity of the business. 


Customer Data Platform (CDP) technically creates unique customer profiles in JSON format and  stores various types of customer-related data such as your purchase history, Call center interaction, online browsing behavior, demographic data, channel interactions, etc. and all this data is updated realtime or near-realtime, for example, if your brand has brick and mortar store then the customer purchase transaction is upload by ETL process on a set cadence into the CDP. But, Online transactions and browsing data are updated in real-time. 


If you are in the market looking for a Customer Data Platform (CDP) then think about the number of potential data sources you currently tap into for your marketing campaigns and I’m sure there is no one place currently. WIth CDP you can definitely have all of that data in one place and your go-to-market plan will be expedited. However, you need to understand the 4Vs of Big Data really matter here - I’m not so sure about the Veracity. The volume of data for a large retailer is really huge and therefore you need to know:


  1. If the CDP that you are considering is able to process large amounts of data and if so how fast can it process?  

  2. WIll the Data is being processed, does it impact other processes in the Customer Data Platform?

  3. What is the average time required to onboard a new Data source?

  4. On a granular level, how much processing is required before loading the data?

  5. What are the potential impacts of adding a new object or Javascript on your website?

  6. Will your website slowdown if you add new tags or extensions on your website?

  7. How much time does it take to onboard a new channel? 

  8. Average time for building a new Use Case, QA, Test, and finally make it live in production?

  9. Does the vendor offer an easy way to move the Use Case from Test to Production or you will have to buy two instances - test and production?


Once you buy a CDP, you’d think that your go-to-market will be a lot faster, accurate and of course Personalized. BUT, the biggest hurdle is when you realize that you will have to build and test in a sandbox and build that same use case allover in the production environment. So if you buy a CDP that does not offer the capability to switch the use case from test to production realtime i.e without re-building in production environment all over again then you are still looking at a 4-6 week go live at the minimum.  


However, there are workarounds like appending the URL and inserting keywords test in your offline data feeds to separate and distinguish test profiles from actual customer profiles. But, that will mean that your CDP will have test data as well as actual customer data in the same database, which is not a best practice and definitely not advised as development and testing are never done in production. 

Comments

  1. Thank you for sharing such a useful article. I had a great time. This article was fantastic to read. Continue to publish more articles on

    Data Engineering Services 

    AI Services

    Data Analytics Solutions

    ReplyDelete

Post a Comment