In Conversation with Andy Krackov: The Urgency and Impact of Data Storytelling
Andy Krackov, Founder of Hillcrest Advisory
As federal investments in data collection and infrastructure face mounting cuts, the ability of state systems to make informed, equitable decisions in education (and across every governmental sector impacting day-to-day life in the United States) stands at a critical crossroads. So we were thrilled this summer for the opportunity to connect with Andy Krackov, founder of the data storytelling consultancy, Hillcrest Advisory, for a timely conversation on the power of data insights to drive meaningful change.
In this special Q&A, Krackov offers an example of how data is being used to inform and mobilize communities to action, and discusses the unique impact and potential of data storytelling. And as he explores what is at stake if cuts to federal data collection capacity and systems are not reversed and persist, Krackov also reflects on what could be some opportunities in this moment.
Advance Illinois: Hillcrest Advisory has done considerable work around child well-being, which is actually a priority area for Advance Illinois. Would you talk to us about a recent time the consultancy worked to use data in that realm to share key information and inspire action? What was the problem, and what was the solution?
Andy Krackov: For roughly the past five years, Hillcrest Advisory has been working on a project in California to address adverse childhood experiences (ACEs). The state of California has committed significant resources to training clinicians across the state to screen for ACEs – in both children and adults – so that we can help individuals with high ACE scores recognize and heal from their trauma. Through the statewide ACEs Aware initiative, we’ve been publishing an array of county-level breakdowns about trainings and screenings.
One of my favorite things to do is to travel to communities across the state, and share these data at local meetings. I’ve done that, for example, in rural Tehama County in far Northern California, and it’s so exciting to see how a community like that can collaboratively review its data, then generate ideas for how to improve service offerings – in this case to ensure that local clinicians are trained to screen for and address ACEs among their adolescent population.
What is it about quantitative data that can drive action, drive change in ways anecdotes and longer, straightforward storytelling cannot?
I like to think of what’s needed for effective communication as data + story. Sure, they can both exist on their own, but it’s best when we can integrate them. A well-told, poignant story draws us in, capturing our attention, while the data can help convey prevalence and degree – that is, the data signal to us that it’s not just an issue with this one person whose story we’re telling. The data help us see that this is a community-wide problem that we need to come together to address.
I give a lot of data storytelling trainings, and there’s a quote that I often turn to to help explain why we need data + story. It’s from a 2021 New Yorker article, “What Data Can’t Do,” by a mathematician, Hannah Fry:
“Numbers are a poor substitute for the richness and color of the real world. It might seem odd that a professional mathematician…would work to convince you of this fact. But to recognize the limitations of a data-driven view of reality is not to downplay its might. It’s possible for two things to be true: for numbers to come up short before the nuances of reality, while also being the most powerful instrument we have when it comes to understanding that reality.”
“A well-told, poignant story draws us in, capturing our attention, while the data can help convey prevalence and degree – that is, the data signal to us that it’s not just an issue with this one person whose story we’re telling. The data help us see that this is a community-wide problem that we need to come together to address.”
In your recent blog on Nightingale, you mentioned speaking with someone who described what’s happening to the U.S. data infrastructure right now as a ‘slow data rot’. What is meant by this, and are you and your team beginning to see it or residually hear about it in your work with clients? What kinds of things are at stake if, to continue the metaphor, termites continue to eat at the house?
A lot of what’s happening now with our federal data infrastructure isn’t necessarily headline-grabbing. Let’s take an example. The Education Department’s National Center for Education Statistics (NCES), whose tagline is “Empowering Education with Accurate, Timely, and Nonpartisan Statistical Products” but may be hamstrung from doing just that in the coming years. Staff have been laid off and funding has been cut, and all of this invariably will impact the quality of NCES’ data collection and reporting, which communities across the country rely on to support education improvements.
But we may not necessarily see these changes immediately – it will be a slow rot. Let’s take another example. The Census Bureau provides data on local-level poverty – a fundamentally important measure – through its annual American Community Survey (ACS). The ACS data we now use likely won’t be impacted by staff and program cuts. These data are from surveys conducted a few years back, given that it takes a year or two for the Census Bureau to process, analyze, and publish the data. For example, we’re now using the 2023 ACS poverty data.
But what about the 2025 data that would be released in roughly 2027? It’s conceivable that staff and budget cuts could delay the release of those data and potentially impact how much data are available for release if budget cuts mean the sample size is reduced. As a result, rural communities may not be able to get poverty data. In addition, however, the Census’ federal advisory committees, which provide important external input on Census data offerings, including the ACS, are now disbanded, and this calls into question the trust and accuracy of future surveys. Can we trust the ACS data on poverty that will be released in a few years? The interplay that happens when a federal agency collaborates with advisory committees helps us know we can trust the data, so without these partnerships, our faith in federal data may diminish in the coming years.
This is all part of that slow rot – we don’t see it now, but the damage that’s been done could be exposed a few years from now, when it’s harder to simply patch up the foundation.
What you described sounds deeply troubling – no, we won’t see effects immediately but when they do begin to emerge in efforts like the next Census, to your point, the implications of a reduced or less-supported infrastructure would be broad and to a lasting effect. From a glass half-full perspective, however—and not at all to diminish the gravity and urgency of this moment we are in where it is critical that cuts to federal data collecting efforts and infrastructure are reversed—is there a second lens through which you’re simultaneously challenging yourself to see this moment, where there is possibly opportunity for helpful innovation? What comes up for you here?
I’m glad you highlighted opportunities. There is a sliver of hope regarding what’s to come with data access, and maybe that will grow into something more substantial as the realization of federal data cuts begins to hit home. I’ve already talked to a number of people who see opportunity with these cuts to revamp measures and to look for more efficient ways to collect data for core indicators such as poverty or reading proficiency. I launched an informal group of individuals in California – I live near San Francisco and do work across the state – who want to respond to the federal data threats by making sure that local communities across the state still have access to needed data. In the coming weeks, this group will begin to explore new ways to capture data we use for local decision-making.
I know there are efforts underway in other states, too – I’ve heard from Advance Illinois about the work being done there to address federal data threats. We’re going to need to find ways to come together across states to share ideas – and perhaps even agree on similar standards so that we can compare data across states. In other words, if the federal government won’t make it easy for us to compare, say, health insurance rates in communities across states, could we build some kind of federated, decentralized model to do just that? Building this may not seem realistic now, but, as they say, necessity is the mother of invention.
This has been a couple years in the making, actually preceding what's happening at the federal level, but the Illinois State Board of Education (ISBE) is actually set to release its Children’s Adversity Index this month. It is our hope that this tool will help local, district, and state leaders make more targeted arguments for funding, support, and policy change that fosters healthier community environments and supports greater child resilience.
With that in mind, what are some actions everyday people can take (particularly when it comes to child well-being) to highlight the importance of good data and strong data infrastructure?
As a first step, use the data that we have now to help you understand how your community is faring. Data tools like Census Reporter, County Health Rankings, and the Child Opportunity Index make it easy for you to access and interpret data for where you live. I bet the Children’s Adversity Index will fill a similar role for people doing meaningful work across Illinois.
I work on one such data tool, AARP’s Livability Index (it’s about overall livability, not just livability for seniors), so I know first-hand what goes into creating tools like this and how valuable they can be to everyday individuals – for example, when making decisions about where to purchase or rent a home.
What’s important to know about these tools is that federal data helps fuel their offerings, so without such data, city planners, school boards, city councils and county supervisors won’t have the data they need to improve our communities. That’s a problem that our elected officials in Congress should know about. So what else can you do? Tell your Senators and House members that you’re concerned about threats to federal data.
One last question, a bonus. If you had to narrow your must-have considerations/mindsets/elements for making a data storytelling project effective to just three things, what would they be?
I’d say it’s vital to know who you want to reach – that is, what specific audience will most benefit from the information you need to share – as well as what action you want that audience to take. In my data storytelling workshops, I remind participants that understanding our audiences and their information needs help us get out of our own mindset. After all, we already know the data, so building a data story that’s meaningful to us isn’t necessary. What we need to do is find ways to present the findings that are relatable to others, so think about your audiences’ needs.
I’ll cheat, and call ‘audience and action’ one consideration. A second consideration is weaving in story elements – a narrative about an individual, quotes, photos/videos – that add needed context to the numbers you’re sharing. When I’m working on a data storytelling project with clients, I not only ask for spreadsheets to see the data they want to convey, but also for story elements so that we can integrate the so-vital human experience. As noted above, we need the heart and head for effective communication about policy matters.
Finally, I think it’s important to recognize that there are an array of freely available data storytelling tools to help you communicate effectively with data. My go-to tools are Infogram, Flourish, and Datawrapper, all of which have pretty robust free offerings. Give them a try. They’ll definitely help up your data storytelling game.
Andy Krackov runs the data storytelling consultancy, Hillcrest Advisory, that works with universities, government agencies, nonprofits, and others in the social sector to help them communicate data findings. Krackov last month wrote a blog post for the Data Visualization Society on how threats to federal data may impact communities across the country.