Choosing Models
Making Your Data Work
In 2002, I made a critical career decision that changed the course of my life forever. I had been working as a product developer at Kimberly-Clark for seven years. I had made a successful start in a research career, inventing new products that would serve people around the world. But I was also deeply interested in global Christian missions. An opportunity came up for me and my new family to move to Budapest, Hungary, to serve as a research coordinator for a network of church-planting ministries in Eastern Europe and Central Asia. How could I not say “Yes!”?
I was eager to bring my experience and expertise from the business world into this new realm. I sat down with other Christian product developers I knew at Kimberly-Clark, and we brainstormed how the kind of world-class research we were using at Kimberly-Clark could be used in the realm of Christian missions. We imagined disciplined, data-driven insight guiding ministry decisions in the same way it guided product innovation. I was ready to make my mark on the world of Christian missions research.
A Disappointing Reality
But when I arrived on the field, I discovered that many of the existing researchers had already been taught a different way of doing research—a way that seemed reasonable and intelligent, but that was absolutely useless for the mission at hand. Now, I do not want to disparage these amazingly talented men and women. Many of them had also left careers to be missionaries. Because of their comfort with numbers or proclivity for technology, they took on the mantle of “researcher.” They genuinely wanted to bring their best gifts and talents into global missions; it was beautiful and noble. Unfortunately, the work that they were doing did little to advance the mission on the field.
My role in Budapest was to coordinate research across numerous churches, denominations, and agencies working across 27 different nations. One day, I was meeting with a researcher from Slovakia, and they were sharing their latest research data with me. They had tracked church attendance and baptisms from every church in their nation over the last three years—quite an impressive feat in data collection. They then decided to take this data and, using a simple compound growth formula (the type used to calculate bank interest), project church growth into the future. It was something that other Christian researchers had taught them to do. Unfortunately, this was a fundamentally flawed approach.
I noted that the number of Christians in the nation was growing at a rate of about 2–3% annually. At the same time, the population of Slovakia was shrinking by about 1–2% each year. I then showed him that, over time, the number of Christians in Slovakia would surpass the number of people. He was a little embarrassed and soon saw the same issue that I was seeing. The problem wasn’t the math—the problem was the model.
A New Kind of Model
A model is a simple description of part of the real world written in math so we can analyze, predict, or explain what might happen. In other words, it represents some real situation (like a population, a budget, or traffic) using variables and equations instead of words or pictures. Essentially, it is a deliberate simplification of reality: you keep a few important factors, ignore the rest, and connect them with mathematical relationships. The point is to get a usable tool for your work: you should be able to enter your data, and the model should serve as a kind of prediction engine based on your data and model inputs.
The core issue with using a compound interest formula for church growth is that it is not at all grounded in the realities of how churches grow or even how people become Christians. Christians are not like money, such that if you store them in a “bank” they will miraculously multiply by some spiritual interest rate. So what is the spread of Christianity like? I would posit that it is most like a disease.
Now, I am not making this proposition to disparage the Christian faith. I am a committed follower of Jesus, and I am passionate about others coming to share my faith. But the Christian faith spreads, and churches are planted, much in the same way a disease would spread through a population. Just like a disease, a number of people in a given population are Christian (“infected” by belief in Jesus) and others are not. Also, like a disease, people become “infected” by Christianity as a function of a few key factors: how often they come in contact with an infected person, their susceptibility to becoming infected, and the contagiousness of the infected person. Also important is the incubation period—how long it takes for a newly “infected” person to become contagious enough to “infect” someone else.
These factors easily translate into strategic behaviors that an organization can encourage or practice.
The number of Christians that need to be regularly sharing their faith with others.
How frequently Christians need to share their faith with non-Christians.
How well trained (or contagious) these Christians are who are sharing their faith.
How long it takes before these faith-sharing Christians are capable of effectively sharing their faith.
How likely a person is to become a Christian—data that could be gathered from existing ministry data.
With this data and these factors, a ministry could not only predict when an area would be “saturated” with Christians; they could also predict how rapidly they should expect Christian growth, when the amount of growth may slow as saturation approaches, and when to expect full saturation of a population. This also helps regional ministries know when to pivot a strategy from local mission to cross-regional or global mission.
The Accountability of a Good Model
The model I built, the key inputs, and the outcomes all showed clear utility. It provided more than a prediction—it provided strategic guidance. Because, in addition to data, you are also supplying key assumptions to a model. You can use your data to show how adjusting other factors you control will affect your outcomes and ultimately your impact.
I debuted this model at a church-planting conference in Kyiv in 2004. Some received the model enthusiastically. They were excited to see how the numbers provided a healthy challenge and a clear exhortation: they needed more people actively meeting with non-Christians. They needed to accelerate their training programs so that newly minted Christians would be out effectively sharing their faith in the population sooner rather than later.
But other leaders seemed threatened and pushed back. They were now accountable for action to see results. They had to do more than pray and hope—they needed to commit to a strategy of finding and evangelizing people outside of the church. They needed to consider accelerating their discipleship and church-planting training to meet their vision and goals. This demanded investment of resources in specific areas and required risky and uncomfortable activity.
Models are both frightening and wonderful. They are frightening in that they hold your organization accountable to make an impact. You can no longer avoid the hard work and humility required to develop and commit to realistic strategies for impact. But this is the wonderful part: models make your research work. They provide guidance and a roadmap rooted in your data that points your organization to opportunities to maximize its impact on the people you serve.
Scott Friderich is the founder of Clarity Research, a market research and impact measurement consultancy helping leaders discover and measure what their customers are experiencing. He facilitates focus groups and qualitative research across multiple industries and publishes regularly at scottgainsclarity.substack.com.