How to Design a Sustainable Data Strategy

Posted by in Data and Analytics, Privacy and Security, Sustainability tagged with ,

Graphic illustration showing the metamorphosis of a caterpillar into a butterfly.
Metamorphosis: transform your organization’s approach to data using these nine steps.

In this guide, we walk you through nine essential steps to designing a sustainable data strategy for your organization.

If you want to become a data-driven organization, read on to learn our nine recommended steps to improve capabilities across your data value chain. Otherwise, feel free to jump to any individual section via the links below:

  1. Define a sustainable data strategy
  2. Data collection
  3. Data management
  4. Data tools
  5. Data skills-building
  6. Metrics and KPIs
  7. Data storytelling
  8. Building a data culture
  9. A data-driven organization

Data Strategy, Data Confusion

Much of Mightybytes’ work for clients involves pushing and pulling data from one place to another, often in the name of improving business or marketing goals. For example:

  • Data captured by a website or digital product must be migrated to a CRM, email marketing, purchasing, or other system.
  • Data from a third-party system must be pulled into a digital product or service to help users make better interaction decisions.
  • Data resides in several places and must be pulled into a unified system for better analysis and organizational decision-making. 

This may seem pretty straightforward. However, data confusion often arises during early discovery workshops we host with clients. People are unsure about who owns what data, where it lives, or even why it’s important. Usually, they’re just frustrated that the data systems they employ don’t talk to one another. 

During these conversations, data questions are also often project-specific, perhaps part of a website redesign or email marketing strategy, for instance. Rarely does the organization have clear, holistic, or strategic goals on how data helps them achieve overall success.

Undermining Data Success

This is especially true at nonprofits, associations, startups, and other organizations where resources are scarce. It is daunting in a resource-constrained environment to create consensus on your long-term data strategy. An organization might have a ‘data champion’ in its midst. However, to successfully leverage data, you need more than one person prioritizing good data practices.

A lack of data-specific resources leads to an environment where data is deprioritized, which is particularly problematic considering that many organizations are required by laws such as GDPR and CCPA to prioritize data privacy and security. Fines for breaking these laws, or not respecting informed consent, can be stiff. 

With all this in mind, let’s explore how you can design and execute a sustainable data strategy for your own organization. 

Graphic illustration of eggs on a leaf.

1. Define a Sustainable Data Strategy

Becoming data-oriented is like changing how you eat: if you want the effects to be sustainable, it’s a lifestyle change, not a one-off.

— John Challice, SVP, Business Development, Hum

First, take a good, hard look at where data stands in your organization: 

  1. Does data play a key role in organizational decision making? 
  2. Is your entire team—including staff and leadership—up to speed on data strategy, collection, and analysis? 
  3. Do you regularly train new hires on data science practices?
  4. How often do you revisit data strategy practices to identify areas for improvement, better understand resource allocation, and so on?

If you haven’t defined a purpose for data collection and analysis within your organization and there’s no analytics experience among team members, you’re data-impaired. Understanding where you stand currently—your baseline—will help you figure out where you need to go. 

The Power of Roadmapping

To set a baseline, define the problem and opportunities that data presents. This will help you build consensus on desired outcomes—what success looks like. 

At Mightybytes, we start this process with workshops grounded in human-centered design practices. Stakeholder mapping and problem-framing workshops help everyone better understand specific data-related problems while roadmapping engagements create a shared vision for the organization. 

During these sessions, we assess the organization’s data-maturity level and analytical capabilities, then map out opportunities to make better decisions and achieve desired outcomes. The roadmapping process allows us to create a phased development plan (roadmap) that balances quick wins with long-term investments to use data as a key strategic asset.

What Makes a Data Strategy ‘Sustainable’?

There’s a growing recognition of humans’ digital footprint on the environment.

— Wunderman Thompson Intelligence

Any roadmap will fail if it isn’t translated into working processes that bring the organization long-term success. Plus, if those processes are inefficient or not tended to over time, they create waste. In fact, Wunderman Thompson Intelligence listed Digital Sustainability as the #13 trend in their report, The Future 100: Trends and Change to Watch in 2021. Data waste contributes to this. Reducing it is a key part of an effective climate strategy.  

Over time, digital waste builds up just like physical waste, contributing to climate change and making it harder to glean impactful insights that will propel your organization forward. This is especially challenging if your organization is already resource-constrained. Thus, effectively—and sustainably—managing organizational data should be a priority. 

Sustainable Data Tactics

So, how can you make your data strategy more sustainable? How can you use data to regenerate systems while also reducing waste? Consider the following:

  1. Map Your Stakeholders: Clearly understand all stakeholder needsincluding the environment—when devising a long-term data plan for your organization. Break down organizational silos by including stakeholders from different areas of your organization and impressing upon them the importance of a clear data strategy. 
  2. Dedicate Resources: No strategy survives without resources—human, financial, and otherwise. Include data strategy requirements as part of ongoing budgeting, staffing, and training conversations, then resource accordingly. 
  3. Efficiently Manage Data: Focus on data efficiency across the organization and in customer journeys. Do you really need to collect all the data? Also, consider web hosts that power their data centers with renewable energy (which may not be as easy as it seems).
  4. Data Disposal: Devise and adopt a clear process to dispose of unused data. Establish how you will respect customers’ ‘right to be forgotten’, which is increasingly becoming law in many countries. Data disposal is both a sustainability issue and a privacy/security issue. 
Graphic illustration of a caterpillar coming out of an egg.

2. Robust Data Collection

Only 6% of nonprofit leaders believe they are making good use of the data they gather.

— Kathleen Kelly Janus, The State of Data in the Nonprofit Sector

Collecting data is easy. However, most organizations do it wrong. Software products, digital platforms, and third-party services are often designed with data collection as a prime directive. Given this, there are probably untold data collection opportunities that you’re not taking advantage of. Conversely, you might be collecting far more data than you actually need. 

Your organization’s data strategy should help you drive smarter decisions regarding which data to collect and what to do with that data once it’s collected. First, understand the types of data you want to collect and why that data is important for your organization’s strategic priorities. 

Understanding Data Types

Robust data collection requires that you clearly understand data types available to you. What’s the difference between zero-party, first-party, and third-party data? Let’s take a look:

  1. Zero-party data is typically given to an organization and can include information like name, email address, company, personal preferences, and so on. This usually represents less than 10% of your organization’s overall data.
  2. First-party data is observed or predicted based on existing relationships you have with stakeholders. It includes interests, motivations, intent, urgency, and so on. First-party data can represent as much as 90% of an organization’s overall data. 
  3. Third-party data, the most common type, is that which is typically purchased from, you guessed it, a third-party. This often includes email addresses, social media data, purchase history, and so on.  

Of the three data types above, zero- and first-party data are likely any organization’s most strategic differentiators. This data has been voluntarily provided to you by organizational stakeholders, often with the intent that it will be used to create a better, more relevant customer or user experience. 

Ethical Data Collection

Next, when collecting data, make sure to follow ethical practices. This is a deep topic, but in short, consider the following:

  • Make sure you have permission to use data you collect for a clear and specific purpose. 
  • Don’t sell that data to third parties without explicit permission from users that it is okay to do so. 
  • Identify and follow clear processes to remove user data when requested. 

For more detailed guidelines, we have a data privacy checklist that you can use to maintain ethical data practices. Make sure your entire team clearly understands these guidelines. 

Data Collection Questions to Ask

Finally, consider the following questions when assessing your organization’s data collection capacity:

  • How does current data map to desired outcomes identified in our strategy?
  • Are there new data sources we’re not taking advantage of? 
  • If so, what will be required to collect and use this data?
Graphic illustration of a caterpillar.

3. Good Data Management

For many organizations, data is often strewn across multiple platforms or in a combination of internal and external systems. These systems don’t usually talk to one another. Compounding this, most organizations are also typically quite siloed. For example, finance or operations don’t effectively communicate with marketing or IT, and so on. Because of these and other issues, stakeholders don’t often understand the organization’s bigger data picture. This can undermine your ability to effectively manage data. 

Improving Data Management Processes

To support better data management practices within your organization, consider the following:

  1. Evaluate data storage, data governance, quality assurance/control, and data sharing practices as they relate to your organization’s long-term data strategy. 
  2. Identify gaps in processes or resources and diagnose opportunities to improve data management practices. 
  3. Evaluate new tools or platforms for better data management versus whether or not it’s better to create your own.
  4. Co-create revised or new standard operating procedures for the data management practices mentioned in point one above.
  5. Schedule regular check-ins and revisit each of the above points as needed to make ongoing progress.
Graphic illustration of a caterpillar eating a leaf.

4. Appropriate Data Tools

If you’re still managing your organization’s data with spreadsheets, it might be time to consider a new approach. Spreadsheets are powerful tools. However, they shouldn’t be the only thing in your data toolkit. Depending on your needs, there are many SaaS tools to help you better manage data and be more strategic in data decision-making. 

Create a Data Toolkit

If your data isn’t portable or if accessing and analyzing data is time-consuming and riddled with workarounds, it is probably time to look for new tools. Consider the following:

  1. Data Assessment: Consider the utility and value of current data tools. Do they meet your needs? Will they set you up for long-term success as your data efforts grow? What are you missing?
  2. Tool Evaluation: Evaluate new tool options to better meet current and planned data needs. What features do they offer at a price point you can afford? Are there upgrade options? 
  3. Build Capacity: If budget doesn’t exist, think about new revenue opportunities or the ability to down-scale your existing tech stack to build capacity for a new data toolkit. You’ll also need to think about improving your team’s data literacy skills (see next point below).

Questions to Ask About Data Tools

  • What data capabilities are we missing that a specific data tool or approach might help with?
  • Which data tools will provide the best features at a price we can afford?
  • Can we find the resources necessary to fund these tools and build team capacity in using them?
Graphic illustration of a caterpillar attaching itself to a tree.

5. Analytics Skills-Building

Building your team’s data capabilities and analytical skills is critical to getting the results you need and managing a more sustainable data strategy over time. Embrace good skills-building practices within your organization and earmark resources to improve team data literacy skills. 

Skills-Building Steps to Take

Consider the following steps to improve your team’s knowledge and capacity: 

  1. Skills Assessment: Identify the existing analytics and technical skills of your staff. Is there a ‘data champion’ among them? Will that person be willing or interested in training other team members?
  2. Skills Gap Identification: Given your organization’s new, overarching data strategy, identify analytics skill gaps in your team. Be sure to distinguish between opportunities for skills-building and the need to hire. 
  3. Skills Building: Prioritize organizational resources to build analytical skills, then offer training to all or select staff.

Data Training Questions to Ask

  • Do we have a data champion in our midst? 
  • What data skills are we missing? 
  • Do we need to hire someone or can we build skills within our existing team? Is there room for consultants to temporarily fill gaps?
  • What is an appropriate amount of training to expect before we might see results? How long might it take? 
  • How often should we revisit training options to keep our team up to date?
Graphic illustration of a cocoon.

6. The Right Metrics and KPIs

You might already have a set of key performance indicators (KPIs) that help your organization define and reach success. When you review these metrics through the lens of your new data strategy, how do they stack up? 

Be sure to define the right metrics for your organization. Again, you don’t need to report on everything, just the important stuff. Analysis paralysis can set in if you’re tracking too many things, which can dilute progress. While emerging AI data tools can help with this, it is still important to stay focused on the data that matters most to your organization. 

KPI Tasks to Consider

Consider doing the following:

  1. KPI Assessment: Define how effectively current metrics help you achieve organizational goals. Be sure to include what’s not being measured alongside what is.
  2. KPI Revisions: If current KPIs and supporting metrics aren’t aligned to desired outcomes, revise them accordingly. This can often be done effectively in a workshop setting. 
  3. Stakeholder Alignment: Create a communications and implementation plan to align staff and other stakeholders on the new approach. Incentivize your team accordingly to stay on track. 

Data Measurement Questions to Ask

  • What KPIs and metrics are we currently measuring? How do they stack up against our organization’s overarching goals?
  • How well do they contribute to desired outcomes?
  • What effect do KPIs/metrics have on incentivizing staff behavior?
  • What is currently desired but not being measured?
Graphic illustration of a butterfly coming out of a cocoon.

7. Data Storytelling

Data reporting can get pretty dry. However, good data practices also provide fertile ground for telling your organization’s most compelling stories. Dig into the data to identify and share stories that will resonate with your stakeholders. Then deliver those stories in ways that are meaningful to them. 

  1. Reporting Assessment: Audit how internal and external reporting currently performs. How can processes and deliverables be improved?
  2. Storytelling Opportunities: Evaluate possible storytelling opportunities based on your new data strategy.
  3. Data Sharing: Share your data stories in formats that are relevant to your audience’s needs and goals. Also, be sure to make them accessible.  

Data Storytelling Questions to Ask

  • What data is used by our organization, and how is it used?
  • How well is progress towards desired outcomes measured?
  • Which stakeholders do we need to engage in data storytelling efforts?
  • How effectively do external reports currently communicate progress to key stakeholders? How might we improve this?
  • Do opportunities exist to improve our data storytelling efforts? If so, what are they?
Graphic illustration of a butterfly getting ready to fly.

8. Instill a Data Culture

To build a mature, data-informed organization, you need to develop and nurture a data-friendly culture. 

Data Maturity Steps to Consider

  1. Maturity Roadmap: We discussed the power of roadmapping above. Successful roadmaps are revisited over time. Develop and revisit a maturity roadmap to achieve the desired analytic capabilities. Ensure that your organization is planning for outcomes, people (skills, leadership, managing analytical talent), processes, and technology.
  2. Define Data Culture Elements: Based on roadmap objectives, overlay corresponding analytic culture elements onto overall organizational values, such as:
    • Searching for the truth
    • Continuous testing and using analytics to drive decisions
    • Seeking data not just anecdotes to analyze an issue
    • Valuing negative results 
    • Being pragmatic about tradeoffs
    • Never resting on your laurels

Data Culture Questions to Ask

These questions should help you gauge your team’s data maturity levels:

  • Do we have analytical skills in place across the entire organization? If not, what do we need to do to get there?
  • Does leadership think in data-driven terms? How does this filter throughout the organization?
  • How do we balance data and anecdotal evidence? What is considered a ‘fair’ tradeoff?
  • What methods might we employ to always strive for better data integrity and more compelling data-driven stories?
Graphic illustration of a butterfly flying.

9. Better Decisions with Data

As your organization builds skills and grows its capacity, data-driven decision-making should eventually become second nature to everyone. While team leadership may not need as many hands-on data and analytics skills as those working with data every day, if they’re lagging behind, this could impede your organization’s progress. 

  1. Improve Data Leadership: Assess current leadership practices around using data to make decisions
  2. Leadership Skills Gaps: Identify analytical skill gaps among leadership and create a learning and development plan to help them
  3. Leadership Learning: Offer training to individuals or the entire leadership team on analytical thinking, data-guided decision-making, and analytics skills/tools

Data Strategy: Final Word

However elusive a sustainable data strategy might seem, it is possible to make data an essential organizational asset with leadership buy-in, an annual budget, data scientists on staff, team training, and so on. The key is to follow a process like the one outlined above and address the challenges one step at a time. We hope you found this useful. If you have any questions, feel free to reach out.

Improve Your Data Privacy & Security Practices

Download our free Data Privacy Checklist

Download the Checklist
Tim Frick founded Mightybytes in 1998 to help mission-driven organizations solve problems, amplify their impact, and meet business and marketing goals. He is the author of four books, including Designing for Sustainability: A Guide to Building Greener Digital Products and Services from O'Reilly Media. Connect with Tim on LinkedIn.