Jacqueline Davis made an online purchase. Jackie Davis has used your phone ordering system and picked up products curbside. J. Davis entered a contest in one of your stores.
Your customer database has separate customer records for Jacqueline Davis, Jackie Davis and J. Davis. Their names are similar; their phone numbers are identical. So, you have three different records for one customer, which means she is receiving three marketing messages from you for each of your campaigns.
How much better would it be if your customer database had only one entry for each customer, and that entry was under the customer’s preferred name?
Making matters more complicated, you’ve recently acquired one of your competitors. Their database also has separate records for Jacqueline Davis, Jackie Davis and J. Davis, but are they all the same customer as the one in your database? Resolving this confusion may not be easy, because the two customer databases are formatted differently.
You can only wonder how many more of the 250,000+ records in your database share similar problems.
This is where the Single Customer View comes into play.
(SCV) is a systematic procedure for making all data available to engage customers across touchpoints. For today’s marketers dealing with customer databases with anywhere from 5,000 to millions of records, it is an absolute necessary. Here’s why.
We’ve all heard the expression, “Garbage in, garbage out.” The simple fact is that in them, including:
It’s natural for customers to enter different versions of their name at different times. When using a credit card, they must use their name as it appears on the card. On other occasions, they feel comfortable using the nickname they are most often called.
Plus, circumstances change. People frequently change the information some databases use to identify them:
According to Marketing Sherpa, 25-30% of data becomes inaccurate every year, leading to less accurate campaigns.
And in some cases, they might dash off a quick and easy version of their name, such as their first initial and last name.
A well-designed Single Customer View makes sure you are using the most accurate data.
With legacy data systems, multiple points of origin often serve several distinct databases. Examples of multiple points of origin include:
And the collected data can be siloed in any of a number of internal platforms, such as:
The combination of Incorrect data, missing data, duplicate data and data siloed in multiple locations is not a data solution, it’s a mess.
As customer expectations rise and competition becomes more intense, a Single Customer View that yields one source of truth for each and every customer and informs all marketing interactions is not just a preference, it’s a necessity.
The procedural stages of a Single Customer View are to clean, integrate, unify, segment and activate the data. Performing these tasks manually is a labor-intensive and time-consuming task. The more data you have, the harder it is.
A Single Customer View cannot be created without clean data. That means you need to:
Here you merge your first-party personally-identifiable data with third-party data from different systems into one database. The types of data you integrate might include:
In this stage, also known as identity matching, you consolidate individual profiles and standardize the data. Once accomplished you will have only one record for each customer, and it will be accessible to every part of your marketing organization.
You also connect identities with attributes, completing a fuller and more well-rounded picture of each customer. Not only do you know who your customers are, you also know what they prefer.
At this point you have created a clean and high-quality database, but there’s more to it.
Here you merge your first-party personally-identifiable data with third-party data from different systems into one database. The types of data you integrate might include.
Set the data to trigger personalized marketing events ranging from birthday greetings to fully customized versions of your website under specific circumstances.
As should be obvious, the segment and activate stages cannot be accomplished manually. Artificial Intelligence is needed for both.
Each of the five stages described above comes with options, lots of options. The need to make these decisions explains why the Single Customer View is a systematic procedure, not a product, and why every seller’s Single Customer View is different.
Here’s a simple example to help clarify that statement.
One element of the created in the Unify stage is the customer’s demographic profile.
That profile can contain a wealth of information, including:
For one company the socioeconomic status might be the most important information about each customer. For another company, life events might be most important, and so on.
And within the Consolidated Profile there are still two more elements: transactional behavior and behavioral patterns. Each is made up of a number of data points, and each data point calls for decisions to be made.
The Single Customer View is fully customized for each seller, and decisions have to be made about the relative importance of each type of data within a given element. All told, there are hundreds of decisions to be made for every Single Customer View. Fortunately, the Customer Data Platform is a powerful tool for managing every stage of the SCV.
The (CDP) equips data engineers and marketers with the tools they need to customize the five individual stages of SCV from one centralized platform.
As an example of capabilities, the allows you to:
CDP is where . CDP streamlines the decision-making process for humans and incorporates AI to enable advanced segmentation, automated personalization and more.
But if you are sending catalogs at $3 apiece, you tighten the match confidence to eliminate waste.
Here’s an example of AI + HI at work on the Customer Data Platform. When matching records with AI, each match receives a score that reflects your confidence in the match. Your selection of that rank might vary, depending on how the record is used. When running digital ads, the cost and impact of waste is minimal, so you might loosen up the match confidence.
Once you have a database of clean and correct names, you can use it as “training records” to train the AI algorithm. Then, as more customer interactions are added to the master database, AI will automatically correct the data to match the training record.
There are numerous Customer Data Platforms available, but they were all developed with creation of a Single Customer View in mind.
As previously noted, the Single Customer View is a systematic procedure. The goal of that procedure is the creation of a Golden Record, the gold standard of customer data.
The Golden Record is shared across business functions, but with one very big caveat. Different marketing units might have different requirements for the data it uses.
New data is ingested by the Customer Data Platform — the hub of the Single Customer View — on an ongoing basis, much of it in real time. AI matches the new data to the training record and validates it before introducing it into the Golden Record.
For example, Ecommerce, email, digital advertising and direct mail will likely have different data record-matching priorities. Those priorities are addressed in the Customer Data Platform, which then produces and distributes as many unique Golden Records as are needed.
Accurate records that are segmented and activated by AI + HI enable marketers to reach the right people at the right time with the right offers. Here are some of the numerous benefits you can expect to enjoy with a Golden Record created from one clean, consistent, unified master database.
Better Decision Making
With a golden record you can really get to know your customers for the first time. Instead of wondering how much each customer is worth to your business and what they are likely to buy, you will know.
Enhanced Customer Experiences
In today’s marketing environment, where customer experience is identified as the single-greatest differentiator of sellers, providing a more personalized, friction-free customer experience than the competition is a path to acquiring and retaining more customers.
Higher Response Rates
A clean database produces better results because all the garbage records have been scrubbed from it. Also, thanks to your Golden Record, accurate personalization presents customers the offers in which they will be most interested.
Lower New Customer Acquisition Costs
Thanks to highly granular persona group segments created by AI, you’re able to eliminate customers who are unlikely to buy particular products from direct marketing campaigns offering those products.
Improved Customer Retention
With one single source of truth for your customers in which identities and attributes are connected, you can build stronger relationships with your customers. Personalization means relevant offers which lead to a larger share of wallet.
Automated Regulatory Compliance
Meeting individual customer privacy preferences is nearly impossible when your data includes duplicate records from multiple sources that is stored in several silos. Keeping data private is also a growing concern, especially with the EU’s General Data Protection Regulation (GDPR) taking effect in 2018. Your Golden Record simplifies and automates privacy preferences and data protection, so you don’t have to worry about them anymore.