Dispelling the 80 Percent Myth of Declining Churches

Buckle your seat belts.

Over the next several posts, I will be sharing with you the results of an incredible research project on 1,000 churches. At the risk of overstatement, I think this data may point us to some exciting and positive opportunities. Indeed, I hope to share a plan for the evangelistic renewal and growth of our churches in the weeks ahead.

But I am getting ahead of myself.

Allow me to share, at the risk of boredom, the basis of this research:

  • Our program randomly selected 1,000 churches with available data for 2013 and 2016.
  • The strength of the study is its accuracy. The margin of error at the 95% percent confidence level is +/- 3.1%. If you’re not a numbers nerd, that means this data is incredibly accurate.
  • The possible weakness of this study is that it only includes churches of my denomination. We took this path because we have a gold mine of data. I do believe, however, this data can be a good approximation of evangelical churches, and a rough approximation of all Protestant churches in North America.

The Research Says 80 Percent Is Not Correct

Have you ever heard, “80 percent of churches are either plateaued or declining”?

I have. It’s wrong.

Here are the results of our research. We used average worship attendance as our metric rather than church membership. Unfortunately, church membership is fast becoming a meaningless metric.

  • 56 percent of churches are declining.
  • 9 percent of churches are plateaued.
  • 35 percent of churches are growing.

So here is the new and correct statement of reality: 65 percent of churches are declining or plateaued. There is a huge statistical difference between 80 percent, the myth, and 65 percent, the reality.

So What?

I loathe research projects that ultimately offer only statistics and not solutions. Over the next several weeks, I will be sharing with you some incredible and eye-opening research. But, ultimately, I will offer some solutions based on what God is doing in these growing and evangelistic churches compared to the declining churches.

Here are some areas I will cover in upcoming posts:

  • The danger line in worship attendance that becomes a predictor for church death.
  • The relationship between the growth of the community and the growth of a church.
  • How some smaller churches are thriving in the shadow of megachurches.
  • What the most effective evangelistic churches are doing differently.
  • The relationship between small groups/Sunday school and the growth of a church.

My goal is ultimately to provide a clear path for evangelistic growth and renewal in our churches. We are learning so much from these churches and their leaders. I can’t wait to share more of our findings in future posts.

In the meantime, remember this basic fact: 65 percent of churches are plateaued or declining, not 80 percent.

And 35 percent of churches are growing.

It is my prayer that we will help you understand how your church can be in that latter group of churches.

Posted on June 28, 2017

With nearly 40 years of ministry experience, Thom Rainer has spent a lifetime committed to the growth and health of local churches across North America.
More from Thom

Leave a Reply

Your email address will not be published. Required fields are marked *


  • Thom,

    Thank you for your passion for the church and your continual work.

    Please help me to clarify. As someone who’s utilized the 80-85% data, often in speaking and teaching, and viewed not only Barna’s data, but multiple others which conclude the 80-85% decline and nearly 12-20% church attendance on any given Sunday; am I to deduce by your assessment that the only difference is to neglect church membership because it’s irrelevant? And by this subtraction of data, we can rest easier, because it’s not as bad? But in reality, church membership is still in decline 80–85%, correct?

    • Matt –

      We studied worship attendance instead of membership because of the wide variances in church membership. For example, we had churches with a consistent membership of around 500 while attendance declined from 200 to 150. So we did not use membership data at all. And, as you will see in my subsequent posts, we certainly have no reason to rest easy.

      • Thank you for your prompt response, it’s greatly appreciated and I respect your work.
        If I may, doesn’t this mean when have an even bigger problem if churches with consistent membership of 500 have 200 attending? By the way, I agree. I’ve spoken to countless pastors and have come up with an approximate 35% activity percentage compared to memberships. Even wiping the “rolls” clean will not fix it, as the 35% seems to be revolving between separate members.
        Looking forward to reading your work!

  • Leslie Puryear says on


    I agree with your change in measurement criterion from members to worship attendance. From a practical perspective, average worship attendance is the most accurate indicator we currently have.

    One other thing I would like addressed at some point is this idea of declining baptisms equating to a reduction in evangelism. While baptisms are certainly a biblical goal, I don;t think it reflects an accurate picture of evangelism in the church. For example, during last year’s VBS, 17 children made professions of faith in Christ. Of these 17, 5 attended our church. The remaining 12 kids were from other denominations. I sent a letter to their parents explaining what decision they made and why they should be baptized. So, in this case, our church would show 5 baptisms when we have brought 17 to Christ.

    Some people might say, that’s only children and they don’t know what they’re doing. Our church also reaches people through personal visits and events. Many professions of faith are made, but they end up going to a family church or delay being baptized. My church may have led 30-40 people to Christ but according to the baptism count, we have only had 5. If I had 6 baptisms the year before, it may appear that my church was less effective when we were more effective.

    I believe the baptism count is necessary and we should baptize everyone we lead to Christ, but the reality is that doesn’t always happen. Personally, I would like to see a one question added to ACP that asks how many people received Christ, or something to that affect. Our churches are used by Christ to reach many more people than we baptize or ever join our church. In the case of my church, we are not declining in reaching people for Christ, no matter what the number of baptisms are.

  • David Hartman says on


    I’m excited about this new study. Thanks for keeping us on the cutting edge of ministry and evangelism. I have devoured most of your research-based books over the years. I got a DMin degree back in 2007 on matching the Witnessing Styles with the Faith Stages for maximum evangelistic impact. Your work in the Unchurched Next Door was invaluable to my project. I am currently writing a practical witnessing book on the same subject, and wonder if you have any updated info on the characteristics of the five Faith Stages (the U5-U1). I assume some of those characteristics have changed in the past 15-16 years since you conducted the original study (2001-2001). For instance, it seems like the Millennials and Z Generation and the Nones are leaning more U-5 in our ever-increasing secularized and Post-Christian culture. Any help would be appreciated. I want to modify these characteristics in my book and of course cite your work. Please use my email for any response. God bless you!

  • Ron Sheveland says on

    Since your research only includes SBC churches, which may be more biblically healthy than some other tribes, is it possible that across the wider spectrum the number might be closer to 80% than 65%?

  • I find this quite interesting. I only looked back 2 years, but the SBC membership overall has decreased by 200,000 or more, each of the last couple of years. With 35% of the churches growing, then those which remained must have lost a bunch of people.

    Baptisms seem to have been 295,000 last year, indicating that the declining churches lost a a serious number of members. I think that’d be worthy of a lot of attention.

    On THEIR part. You seem to have been addressing it.

  • Greg Noble says on


    I am looking forward to seeing your results. I have read some of your books and tried to apply some of the concepts and inform our leadership team.

    Will your data break down church growth and whys among different sized congregations. It seems like their are different growth plateaus that churches struggle to break. (I.e. 300, 500, 1000, etc.)


    • Thom S Rainer says on

      Absolutely, Greg. We will have some sobering information about the implications of the sizes of the church in my Monday post.

  • Todd Wilson says on


    Looking forward to your follow up posts. We use the 80% number at Exponential and believe it to be true as an aggregate across all denominations for several reasons. I suspect your tribe and / or the specific sample does skew some of it. But in the context for which most people use the stat, why do you feel there is significance in the difference between 65 and 85%. At Exponential, we believe the far bigger problem is that only 4% of churches are reproducing. In real life, just like people, 100% of churches are in the process of dying as we speak. Some are fortunate to grow, but they are still dying. If they don’t reproduce…that’s where the trouble comes.

    Just out of curiosity, when you “randomly selected 1,000 churches with available data”, does that mean a percentage of churches you contacted would not (or could not) give you the data? If so, what %? We have over 1500 churches (cross denominations) in a growth and reproduction assessment tool. A large % of declining and plateaued churches will not participate in surveys. How did you factor that in?

    • Todd –

      We randomly selected from a data base with complete information, so there was no non-participation. The statistical difference of 15 percentage points is deemed highly significant from a sampling perspective. The simple way to resolve the discrepancy you see is to compare our two sets of data. If you would like to send me your data, our researchers will be able to quantify the reasons behind the differences rather quickly. Since you have my contact information, feel free to send your data directly to me. We will give it a quick turnaround and give full credit to Exponential for your portion of the research project. Look forward to getting the data.

      • Todd Wilson says on

        So the database had 100% of SBC churches and all 100% had data in your database? If not, an adjustment is needed

      • Thom S Rainer says on

        Todd –

        As I indicated in the post, the research came from SBC churches. There are 47,272 churches as of September 30, 2016. Churches submit their data annually. No, not 100% of them report annually, but the compliance rate is higher than any body of data with which I have familiarity.

        I am not clear about your statement of making adjustments nor would I know how to do so from a objective statistical base.

        If you are concerned that the data is skewed toward healthy churches, I think you will see otherwise in my follow-up posts.

        Again, I am happy to compare our two sets of data to discern any discrepancy. That would be a good exercise for the Kingdom.

    • Jeremy Miller says on

      Todd’s just mad because the 80% number helps him sell more books and conference registrations. You’ve started meddling with his business by correcting his bad data. Because if 80% was correct, churches would be in freefall. But again, that scare factor really helps to sell Todd’s books and fills conference.

      • Todd Wilson says on

        Wow Thom. Surprised you let such derogatory comments through. I’ll jump back to USA Today’s forum that’s more friendly. Thanks for answering the question about the input. I was trying to understand if the database, not the sample, was voluntary. In our experience any voluntary sample database has to be corrected when trying to compare declining to growing churches. If the data going in is voluntary participation with ability to not participate, then that alone could explain the 15% difference. Was simply trying to understand. For the record, Exponential is a non-profit that gives most all its resources including books away for free. A 65% number is better for our cause than 80% so I’d love to use that better number. But unfortunately I’m certain it’s not correct for the broader church body

      • Thom S Rainer says on

        Todd –

        I just saw the critical comment of you a few minutes ago. While I let all comments in, especially those that are critical of me, I will delete that comment.

      • Thom S Rainer says on

        Todd –

        I was wrong. I deleted the negative comment and it deleted the entire thread, including all of your comments. So when I restored the thread, it brought back the negative comment. I don’t want to delete the concerns your expressed, so I am now at the mercy of the software. I have to delete all the comments in that thread or none of the comments. I am sorry.

      • Marguerite Colson says on

        Thom –

        I have been a reader of your blog for over five years, and a listener to the podcast since its inception. I have read most of your books, and some of them I have read twice. You are likely the most influential voice among local churches today.

        One of the reasons for your influence is the gold mine of content you provide. Another reason is your humility and willingness to hear other perspectives.

        For the latter reason, I am writing my first critical comment of you. I am terribly disappointed you even tried to delete the critical comment of Todd Wilson. You have let much worse criticism of yourself stand as is. Why did you attempt to capitulate this time?

        One of the many reasons this blog has grown in readership and influence is your listening ear. For whatever reason, this one person caused you to be inconsistent with your fairness through the years.

        I am disappointed.

        Rant is over. I am still a fan.

      • Malcolm Dodd says on

        Mr. Rainer –

        I am a senior statistician for a biomedical company. Statistics are my life, career, and education. I always appreciate your objectivity and candor. I have tried to understand what Mr. Wilson is asking from a statistical point of view. I am just unclear what he would like you to do with the clear and unbiased data you have. If political pollsters survey 1,000 people, we don’t cry foul because 10,000 refused to respond. Your data seems quite robust, and your conclusions are certainly sound and reasonable. Your offer to compare his data with yours was quite gracious

        So congratulations on some great research. I am looking forward to more of your posts on the topic.

      • Todd Wilson says on

        No worries Thom. It’s ok. My question was seeking to genuinely understand

  • I follow this blog and have read your books. 80% of the time I get a gold nugget and use it in leadership discussions where I am a pastor. Thank you. Thank you. Thank you.

    About the origins of the myth, see the following article for some leads to follow up on…


    I don’t think that sample groups are the same in any study and to compare results across sampling groups could lead to the formation of erroneous conclusions. A longitudinal study of the same polling group would be most accurate. But since the research hasn’t been found yet, a longitudinal goal can not be attempted and… it may not need to be remeasured. Just start fresh and keep with the new sample group over a 20 year period.

    I would say the populist view that “80% of churches are in decline” is not in operation in the SBC. In the SBC, 65% of its constituent churches are in decline. So, looking at the 35% of inclining churches, what are the observable biblical catalysts for vitality?

    • Thom S Rainer says on

      Great question, Paul. We are in the process now of getting that information from the growing churches. Stay tuned . . .

  • Praise God,
    We are a growing church, past year going from about 40 attendance to averaging 100 now. Contribute it to a pastor that is a Evangelistic speaker and motivator . Works diligently with follow up with people. Two great things I love to hear from the people is that they love our pastor and that our pastor says how blessed he is to be here. We’ve also been blessed with new people that are letting the Lord use them with what they can do, and it’s remarkable what God is doing through them. I just give God the Glory and am so thankful to him for what he is doing.

  • Interesting data Tom. Found that in certain States i.e. California the 80% tends to be accurate. SNC culture and location may sque the results
    Interesting to follow the whole matrix

  • Thanks for your faithfulness. Will you be looking at the Canadian church as well?

  • Rev. larry hurley says on

    The Barna research seems to have researched a different group of Christians.

    Their research and your research appear contradictory on many issues.

    • Thom Rainer says on

      Larry –

      Could you be more specific? I am familiar with a lot of the Barna research, and I am not aware of the points of contradictions.