3 Simple Steps to Build a Successful Customer 360 View

The Customer 360, Explained

A 360-degree view of anything simply means taking a full, comprehensive view of the subject area in question. For example, a manager may ask her employee to complete an “employee 360 review” which would encompass feedback via the employee’s staff from below, the employee’s horizontal peers, and perhaps a superior or two from above.

In similar fashion, obtaining a 360 degree view of your customer means surveying the full range of data points across the customer lifecycle; including marketing, sales, and support activites. At the end of the day, the rationale behind building a customer 360 is quite simple: the more you know about your customer, the better suited you’ll be to acquiring and retaining that customer.

Customer relationship management (CRM) systems have been around for years, but they tend to pump piles of customer data into operational data silos. While no organization is in shortage of customer data to peruse, the quest for actionable knowledge remains more relevant than ever. Fortunately, recent advances in big data and machine learning have provided a means for unlocking key streams of customer insights which would otherwise remain hidden in tumultuous torrents of raw data. With the right balance of relevant data, technology, and operational goals, a customer 360 view can be a powerful means of finding these digital diamonds in the rough.

Separating Value from Vaporware

While customer 360 views have their upsides, such efforts are not without perils and pitfalls. Customer 360 projects can go sideways when business objectives are overly broad. Likewise, technology solutions must also be scrutinized to ensure they hit near-term goals, but also eventually scale to meet longer-term objectives for a broader business audience. Both of these areas are covered in deeper detail in the following sections.

Marketing Use Cases

From a marketing perspective a customer 360 aggregates and correlates data from web sites, CRM systems, and advertising servers to ultimately “connects the dots” across the user journey; thus helping marketers conduct quantitative measurement across the various stages of user activity. Enumerating the steps from anonymous user interest to qualified lead to conversion to sale is the holy grail of marketing management, and it’s practically impossible to achieve without an intermediate system that “joins” entities among dissimilar systems.

Sales Use Cases

When it comes to sales, perhaps the most venerable “go-to” use case for the customer 360 is that of the cross-sell / up-sell. Example: if you know a shopper has a Gillete shaving razor, then sell him some blades. For the lady who purchased a one ounce bottle of Coco Chanel perfume last year, urge her towards the two ounce bottle this year.

All of these selling decisions require past sales data, which is easily obtained. However more complex scenarios could include pulling cloud service usage data from the engineering team to see if user usage is declining, thus signaling potential churn. Likewise, data from a third-party solution could indicate that your customer (at least at the account level) is conducting trials with a competitor. Without quick intervention, that customer could be lost forever.

Support Use Cases

Support is no longer the department where customers go simply to open help tickets. Support centers are expected to be proactive, and in some cases, even conduct direct sales when they have customers on the phone. On the proactive front, a customer 360 view can help by monitoring social media channels and scout out things such as a Tweet about a delayed flight or a Facebook rant about less-than-professional service in one of your retail stores. Likewise, a user’s device can “phone home” and proactively open a support ticket all before the user is even aware of an issue. Finally, a user who is about to hit a quota on a metered cloud solution may get a call from a support technician (a real human or robotic) who’s ready to execute an upsell opportunity within as few steps as possible.

Finding Your Use Case

In all the examples above, the entities found across these systems aren’t intuitively matched. For example, your Zendesk support tool may know of customer Jane.Doe@example.com, but it may not realize that the Twitter handle @TheRealJaneD123 is actually owned by the very same person. Again, there are countless ways to connect these dots; from having your Zendesk solution simply ask for social media handles to creating gated content on your website that allows social media logins (thus passing handle and email to your internal systems). The key takeaway is to build a customer 360 view that has immediate value, yet can evolve and grow to meet future use cases.

Tying the examples above together, the marketing department may initially spearhead the customer 360 effort by building the first iteration of the customer view. Over time, sales would bolt on their use cases, and finally support would join the party as well. Building one solution for all needs may be overly optimistic, and thus not very realistic. However, what you certainly don’t want, are three different customer 360 platforms serving three different departments.

Customer 360 Technology Options

A discussion on specific technology solutions would not only be vast, it would likely be obsolete the very moment this article is published. Therefore it may make more sense to think about technology stacks as “implementation patterns,” or more specifically, general design approaches.

The Analytical Approach

The analytical stack is the easiest to understand, given it’s essentially an “out of band” data repository that centralizes all correlation intelligence without impact to operational systems. Common implementation examples include large-scale Hadoop data lakes, very large Amazon Redshift relational databases, semi-structured Elasticsearch or Splunk bunkers, and in some cases a mix of all of the above.

The analytical stack is a natural “first-step” towards building a customer 360 view given the fact it consolidates data, and aside from pulling data from source systems, no functional changes are required to those operational systems. The downside to the purely analytical stack is that all that synergistic goodness of customer insight remains locked away in an ivory tower data silo which only a select handful of data analyst access.

The Operational Approach

A full pendulum swing in the opposite direction of the analytical stack is the operational stack. The operational stack hard-wires data flows from one component to another and the value associated with this cross-system data pollination is available for both analytical and operational applications. The most common examples of operational stacks are found in “off the shelf” solutions such as Salesforce.com or Oracle CRM which can inherently store customer data across sales, marketing, and support functions.

Operational stacks are great, and represent a relatively easy way to get cross-functional views of user activity. But due to their monolithic nature, operational stacks fall short when other solutions need to participate in the customer data ecosystem. For instance, what if we wanted to blend website usage from Google Analytics into our Oracle CRM solution? Or how about if we needed to push social media feeds into Salesforce.com? Unless the vendor provides a simple plug-and-play application integration, you’re going to be faced with some heavy lifting in the data integration department.

The Hybrid Approach

It may feel as though the analytical and operational stacks are a bit extreme, and if you guessed that the hybrid stack is the compromise in-between, then you guessed correctly. At a high-level, the hybrid stack blends data from multiple systems in a manner that can be used for both operations and analytics. This can be done in multiple ways, but two basic approaches include: (1) pushing data into a centralized analytical system which then backpropogates enriched data into operational systems and/or (2) creating a “full-mesh” of data integrations among the operational systems.

The hybrid stack may represent the optimal balance of data sharing, but that doesn’t mean it’s the easiest approach to implement. The full-mesh approach is often implemented using native application integrations, such as a Marketo to Salesforce connection here or a Google Analytics to WordPress plugin there. Of course managing this architecture quickly becomes a nightmare given there are so many different application integration types with no overarching data management infrastructure to speak of.

The centralized backpropagation solution is a far better approach, but still requires intelligence residing in the hub of all these spokes; usually in the form of a master data management solution which in and of itself can be a behemoth to manage.

Keep in mind that just because the hybrid stack represents this Goldilocks-like “just right” zone between operations and analytics, it isn’t always the right place to start. Some enterprises may have better luck starting off with an analytical solution, then evolving into a hybrid stack. Alternatively, some firms start off with an operational-centric approach and stay relatively simple for years to come with no need to weave in foreign data systems. It really comes down to meeting the specific needs of a particular organization.

On Master Data Management

You may be wondering if a master data management solution is needed in order to build out a customer 360 view. As outlined above, the analytical, operational, and full-mesh variant of hybrid do not require a centralized “brain” to create master records.

Of course the hybrid model which centralizes data then backpropagates enriched data back into operational systems is the quintessential MDM use case. Moreover, if you ever want to branch out from customers and build additional 360 views of other business objects within your organization (think products, store locations, or employees) then one should strongly consider putting an MDM at the heart such data infrastructure in order to maintain any semblance of data governance.

Key Takeaways

When building out your customer 360 view, take these takeaways into consideration, regardless of the chosen approach.

1. Build for Today, Plan for Tomorrow

Perhaps the toughest challenge around building a customer 360 view is determining the right scope to begin with. By including many different departments and stakeholders, the project can easily bloat, thus creating a Swiss Army Knife of sorts that does a lot of things, but no one thing well.

Conversely, it’s not uncommon for business units to build their own organically-grown views of the customer; thus creating departmental point solutions and data silos. Even worse are system-centric customer 360 views which cannot be leveraged outside of the core operational system where they reside. A common example here is a Salesforce.com AppExchange bolt-on that won’t scale to meet additional sales department use cases; let alone marketing or support functions.

A commonsensical way to manage scope is simply to build a solution that satisfies immediate needs, yet shows promise for scaling into other domains of applicability.

2. Take an Iterative Approach to Deployments

Even before the initial customer 360 view is defined, it may be advantageous to manage your customer 360 with an iterative, product management-centric methodology. This ensures that the customer 360 features are prioritized based on well-defined/socialized criteria, and gives the stakeholders of the solution an idea of “what they can expect to get, and when they’ll get it.”

As a trivial example, the first minimum viable product (MVP) release may be a simple aggregation of customer attributes into an analytical model for marketing segmentation analysis. A well-groomed feature backlog would show that the sales department is next in line to receive their business request of detailed cloud product usage statistics, which will be delivered as a feature in release 1.1 in six weeks. Finally, the 1.2 release due in six weeks after that will give support a view into Twitter and Facebook sentiment analysis which they deem is the most important near-term customer analytical need.

3. Measure, Measure, and Measure Again

The whole point of a customer 360 is to create additional visibility around the customer lifecycle, and you should do the same with the customer 360 program itself! Strategically measure how the solution is impacting the bottom line while tactically measuring which features are being used. The customer 360 view will inevitably evolve, and using quantifiable data points will serve as guideposts for how to shape the solution over time.