Three ways for Churches to count COVID-19 online worship attendance

For churches part of organized denominations that collect reports on yearly statistics, the first year of COVID-19 might be a tough one to figure out. Patrick Scriven, communicator at the Pacific Northwest Annual Conference of The United Methodist Church, lamented on a recent blog post the problems with counting this year. But sadly, they are required by denominational authorities so pastors and church administrators are doing them without a guide. This blog post will help.

Many denominational authorities are not providing “how-to” but instead are just naming what NOT to do: don’t count total views as attendance, etc. While your regional or national denomination may have specific guidelines, in their absence, here’s three ways to count your attendance on end-of-year reports that are credible and offer faithful best guesses in a varied year. 

Step One: Congregational Survey on Attendance

All of these formulas and proposals have one requirement: knowing, on average, how many people are on each account (screen or view) in your congregation. If there are 10 accounts on a worship service, then you might register 10 in attendance. But if 3 of them are couples and 2 have combined 5 kids that are participating, then your count of “accounts” is inaccurate (18 people are worshipping over 10 accounts). While, yes, you can count visually on some platforms (like Zoom), some accounts don’t have cameras and in larger congregations, it can be tedious. 

The solution is to poll the congregation and ask how many people usually worship regularly in each household (equivalent to an account, screen, or view). That gives you a simple way to determine how many people are represented by each account, screen, or view. Take the number of actual people, divide by the number of of households, and you get your multiplier. Here’s an example church that we’ll be using throughout this blog post.

EXAMPLE: First Church polled their congregation with the above question. 100 households that responded to a congregational survey indicated a total of 170 people were regularly worshipping in that group. 170 people divided by 100 households = 1.7. Therefore, each account, screen, or view could logically represent 1.7 people. 

We’ll use this example throughout this blog post. And keep the decimal accurate not rounded up or down: it matters significantly when you get into larger numbers! 

Make sense? Good, because each method uses this multiplier. So figure yours out by congregational survey (not by assumption!) 

Option 1: Counting Attendance by logged-in accounts on Zoom

This seems easy: if you have a worship service on zoom, just count the number of people you see on screen, and there’s your attendance. Done, right?

Usually, yes. But not always, for the reasons in the previous example. So if you are consistently having trouble knowing how many people are logged in, or you hold Zoom worship in Webinar mode and can’t see faces, then you use the above multiplier.

EXAMPLE: First Church holds a worship service over zoom. They can see 100 accounts are logged into the Zoom meeting. Their worship leaders can go through and tediously count faces or assumptions. Or they can apply the multiplier found previously: 1.7. Therefore, 100 accounts x 1.7 multiplier = 170 in attendance. 

The next week, there’s 50 accounts logged in on the worship service. So 50 accounts x 1.7 multiplier = 85 in attendance. 

While counting faces on Zoom is the most solid way to report attendance, you can use a multiplier (the same one, consistently, or update it every 6 months) to calculate probable attendance for a particular Sunday or overall. 

Option 2: Counting Attendance on Live-streamed Worship

If you use a streaming platform like Facebook, YouTube, or Vimeo, you won’t be able to count faces because that is a broadcast platform. Each gives a bewildering number of statistics back, so you might be overwhelmed by the options.

Luckily there are only two numbers you need: Concurrent Attendees and Total Minutes Viewed. Here’s how to use each one (some multistream services have different terms, check with them to find out the equivalent). 

Counting Attendance by peak concurrent attendees 

If you think about it, calculating worship attendance in person is the number of people who attend during the hour of worship at the same time. If you want to recreate that number most effectively, then you would want to count the number of people simultaneously in attendance during a livestream or premier on a streaming platform. 

My local church has chosen to count attendance by peak viewers, which means the moment when the most people were worshipping together. You can find this on YouTube or Facebook in their statistics section, or you can simply register the highest number of people you see logged in live during worship (On Facebook, it’s at the top of the livestream; on YouTube, it’s underneath the livestream). Then you use your multiplier and—boom—there’s your attendance.

EXAMPLE: First Church holds a worship service live-streamed over YouTube. During the worship service, a worship volunteer sees the highest number of people watching at one time was 100. 100 people x 1.7 multiplier = 170 people in attendance.

The next week, the worship volunteer forgot to check during the worship service. They login to the account and click on the YouTube Studio section, go to “Engagement” on that week’s video and see the number of “peak concurrent viewers” was 50 people. 50 people x 1.7 multiplier = 85 people in attendance. 

Thankfully, the number of “peak viewers” is retained in the statistics, so you would be able to go back and check previous weeks (especially if you understandably have a gap in reporting during in 2020 rollercoaster of changes). My church has noted that some weeks don’t report peak viewers on Facebook for some reason, so use an average for those weeks. 

Counting attendance by minutes

Conversely, online worship means that people may not be logging in at the same time, and may be worshipping at different hours of day and night. We no longer need to be shackled to only counting those present at a place, but who logs in to an experience. But how do you count during the week? 

The Rev. Melissa Cooper, Minister of Worship and Arts at St. Luke’s United Methodist Church in Orlando, shared with me how her congregation calculates their attendance, which they readily admit is “non-scientific but likely pretty accurate.” They use the “total minutes watched” statistic that is calculated in the streaming platforms backend and divide that by the number of minutes in a worship service (to find the number of full worship services that were viewed), then apply the congregational multiplier. 

EXAMPLE: First Church has a 60 minute worship service on Sunday. At the end of the week, the statistics indicate that a total of 6000 minutes were viewed on YouTube. 6000 divided by 60 minute service = 100 accounts x 1.7 mulitiplier = 170 in attendance. 

The next week for a 60 minute service, they check the statistics on Saturday night and 3000 minutes were viewed. 3000 divided by 60 minutes = 50 accounts x 1.7 multiplier = 85 in attendance. 

This makes sense to me because it does reflect the reality that while there are tremendous benefits to logging in and worshipping together, including live comments and reactions on some platforms, that’s not the way how people are worshipping anymore. People join in when they can and when they’ve carved out a time for reflection. Counting viewed worship services during the week reflects more accurately the number of people being inspired each week. 

Comparing Viewers and Minutes

These numbers can vary widely and it’s up to your congregation to pick one and be consistent.

Here’s a real-life example of comparing one Sunday by these numbers in a UM church that sent me their statistics, so you can see they vary widely. This church also had a 1.7x multiplier, though yours may be higher or lower. 

• PEAK ATTENDANCE: Sunday worship had 90 peak concurrent viewers. Using a 1.7x multiplier, that would yield 153 attendees.
• MINUTES VIEWED: The same Sunday worship was 57 minutes and had a watch time of 88.6 hours (6516 minutes). 6516 minutes divided by 57 minutes = 114 accounts. Using a 1.7x multiplier, that equals 194 people. 

That’s quite the variance, so only you and your ministry context will know which method is better for your context. Whatever it is, use it consistently, even though other numbers may be better or worse. 

Final advice is to update your multiplier by polling the congregation or viewers every 6 months or so to see if the numbers have changed, if you are reaching more or fewer people. 

Your Turn

How do you count online attendance? What cautions or celebrations can you share?

Thanks for reading, commenting, subscribing, and sharing on social media.