How to Develop a Healthy Data Culture

Why The Conversion Wall Matters to Marketers and Business Executives

Marketers and business leaders should care significantly about The Conversion Wall, and the challenges it presents.  The Conversion Wall essentially blocks an organization from utilizing data correctly in decision making.  The true issue at hand is what I like to call “data culture,” which is a company’s approach towards data in general.  This culture is developed within and permeated across an organization, and it’s often a disease which is very difficult to cure.

Not sure what The Conversion Wall is? Read my previous post.

Throughout my decade-long career in digital marketing, I’ve experienced the inner workings of more than 400 organizations. I’ve seen very few companies with a “healthy” data culture.

Every organization needs data to illustrate results, make decisions, remain competitive, and grow their business.  Simply put, it’s crucial that you develop a healthy data culture, otherwise, you risk any number of the following challenges:

1. Communication about data is incomplete and leads to misunderstandings about what is driving results.
2. Data does not correlate to or support the revenue and profit goals of an organization.
3. Decisions are made based on an incomplete understanding of a situation due to missing data, or poorly structured data.
4. The organization settles for a low level of data-quality or integrity, thus operating in a less-than-ideal setting.

Healthy vs. Unhealthy Data Culture

Let’s define some characteristics of a healthy, and unhealthy data culture so that you can start to understand where your organization falls.

Healthy data cultures tend to meet these descriptions:

• The organization develops data governance, including centralized definitions and policies regarding data source, granularity, cadence, and integrity.
• The organization supports and incentivizes constant data improvement, seeking to collect and utilize more data in decision making.
• Individuals in the organization are trained to be skeptical of data, seeking out a personal understanding of a data’s source and methodology before accepting it as truth and using it for decisions.
• Individuals are trained in data governance and seek to understand the true meaning of data prior to passing judgment.
• Individuals utilizing data are trained in basic statistics and understand that making decisions based on data, when done incorrectly, can involve assumptions that lead to costly mistakes.

Believe it or not, healthy data culture is something that touches every role in an organization – from the lowest ranking positions up to the Board of Directors.  I’ve experienced, first hand, major organizational shakeups triggered as a result of the Board of Directors not understanding the source and meaning of data presented to them.

The worst thing you can do with data is to pass judgment based on false assumptions.  Conversion data from marketing channels (often times meaningless!) falls prey to this outcome easily.

If you’re communicating “conversion” data with no clear and granular connection to revenue, you’ll likely be the next victim.

An unhealthy data culture will contain many of the following characteristics:

• The organization doesn’t invest in data governance, definitions, education or improvement.
• There is very little structure or documentation around how the organization should collect and align data with its business objectives.
• You’ll often times encounter very disparate segmentation methods. For example, no one in the organization understands which initiatives belong to what business unit, and how data flows to the P&L.
• Outside partners, such as CRM, analytics, and marketing agencies present “standard” data and have little understanding of how to align their dimensions to the organization’s needs.
• Marketing reports often end with and focus on, conversions, leads, form submissions, phone calls.
• The organization does not openly promote data-quality improvement, but rather operates in a passive state when it comes to data structure, integrity, definition, and education.
• Individuals consistently accept data at face value, with only a few select leaders often questioning the sources but not digging deep enough for validation.

I would posit that one of the largest organizational challenges facing companies over the next decade will relate directly to data culture.

Living in a marketing & sales world with no clear and granular connection between “conversions” and revenue opportunities is one sign that you’re facing this challenge as well.

The Structure of Healthy Data Culture

Healthy data culture may come in many forms, but I tend to believe it can be defined by a few principles: definitions, integrity, granularity & improvement, and accessibility.

Data Definitions: It’s hard to communicate without common definitions. We often take this for granted within our everyday lives because so much of what we communicate is governed by language definitions that have been ingrained in our culture for centuries.

Take a moment and imagine how difficult it would be to communicate if we all used the same words but didn’t share the same understanding of how those words are defined. 

Data is no different. From the organizational level to the individual employee, knowing the definition of specific data is of utmost importance to understanding it.  This seems quite simple but think about how many times you’ve sat in a meeting and realized everyone doesn’t understand, and sometimes doesn’t care to understand, the true definition of the data – I’ve seen it hundreds of times, at all levels: from the passive executive who is too busy and ultimately settles on assumption to the low-level employee who feels it is not part of his or her job.

In addition to understanding specific definitions around data, organizations must develop macro-level definitions to guide data structure. Take, for example, an organization with multiple business units, service lines, and product groups. Everyone in these organizations, agency partners included, should operate from a centralized understanding of these dimensions, and all data should be structured around them where possible. Again, this sounds very elementary, though I have yet to work with an organization that does this well without first experiencing major cultural shifts.

Data Integrity: People often assume that integrity and quality go hand in hand.  While an organization should constantly be improving data quality (see the next section), integrity is a completely different concept.

Integrity is a word that people generally understand, but often don’t apply correctly. Let’s review the exact meaning of the word:

The quality of being honest and having strong moral principles.

Data integrity is not about having a perfect data source or the highest quality, rather it’s the characteristic of being honest about, and honestly understanding your data.

Let’s repeat that, it’s very important: Data integrity is being honest about, and honestly understanding your data.

Let’s keep this very simple and call a spade a spade: if you are reporting on Conversions, that is NOT the same thing as leads – The Conversion Wall prohibits this.  Conversions represent actions relative to a website session and there is never a one-to-one correlation between conversions and lead records.

If you believe that there is a one-to-one correlation, or that there should be, you need to revisit your Data Definitions – and you need to get your team on the same page!

Data Granularity & Improvement: Once you’ve established definitions (and a culture that constantly seeks definitions) and treat your data with honesty & integrity, you’re ready to focus on data quality. Organizations with healthy data culture develop cross-functional initiatives to improve the data they collect and report on.

Leaders in the organization should be constantly asking for and enacting structured data quality improvement. When it comes to sales & marketing data, this often requires collaboration between

• Marketing
• Sales
• IT
• Outside Agencies

Stop settling for the data you already have and begin to define the data you need more of.

A great place to start is to focus on the areas in your reporting and analysis that result in assumptions.

For example, assuming that each “Chat” or “Inbound Phone Call” equates to a “Lead” is a huge assumption. And while these actions might create lead records in your CRM, you need to understand the quality of those leads in relation to the contact medium.

Data Accessibility: The last element of a healthy data culture is accessibility. Data should be generated systematically and reliably either in real time or on a schedule. Reports and analysis should be compiled centrally and access to the data should be appropriate for the situation. Most importantly, these things should be defined, and the right stakeholders should be involved in laying out those definitions.

For example, I’ve often seen organizations settle for reporting provided by an outside agency, rather than integrating across organizations to create a more robust, accessible reporting mechanism. Weekly PDF reports from your agency that show “Conversions” leaves a lot to be desired – I hope you are starting to understand why!

The Conversion Wall and Poor Data Culture

Throughout this article, I’ve discussed data culture in general; however, I’m sure you’ve picked up on the underlying focus around marketing data and “conversions.”  The Conversion Wall goes hand in hand with unhealthy data culture, especially in a marketing organization.

In the first part of this article we learned that The Conversion Wall is the invisible, but extremely powerful blocking force between conversion count and business revenue.  

Most marketing organizations focus on driving inbound activity, resulting in prospective customer’s engaging with the sales team through the website (and via multiple contact mediums).

This is where things typically break down: agencies and marketers focus on generating conversions, and very few of them take a concerted effort to trace granular detail through to revenue (in a scalable and automated fashion). In most situations, marketers report on conversions, and then they report on MELs or MQLs, but they lack granularity about each one’s source and ignore the gap between conversions and MEL.

This conflicts with The Marketer’s Directive:

Generate revenue for the company while collecting data about what works and what doesn’t work so efficiency and growth can be achieved.

If you are a marketer or run a marketing organization, it’s time to focus on improving your personal data culture as well as the data culture within your company.