×ChurchOps.AICross-church benchmark
×ChurchOps.AIYour workbook already cuts this archive by size and by age. What it has not had is a single respondent-level read across the complete 528: scores and written comments together, in one place.
Every score on this page was recomputed from the respondent-level rows in Master Analysis of Team Survey Responses, independently and in a second tool as a check. It reproduces the 20 Questions - Master Analysis tab exactly, on all forty values. Your analysis is sound.
Four things worth knowing before you quote anything here:
Across every size of church, every age band, and single-site against multisite, Culture moves less than any other category. The differences between churches live overwhelmingly in Structure, People and Systems.
Two honest caveats, because this is the finding you will want to quote. The size cut is the most dramatic one. By age, Culture still moves least, but the margin is narrower (0.16 against 0.37 for Systems). And a 0.02 spread on a 528-response sample is smaller than the sampling error on any one of those three groups, so the right claim is “Culture moves least” rather than “Culture is identical.”
The direction holds on every cut. The precise size of the gap does not, and it is better to say so than to have someone else say it for you.
| Category | Small | Medium | Large | Spread |
|---|---|---|---|---|
| Culture | 4.55 | 4.54 | 4.53 | 0.02 |
| Structure | 4.13 | 3.99 | 4.21 | 0.22 |
| People | 4.02 | 3.96 | 4.19 | 0.23 |
| Systems | 4.03 | 3.85 | 4.14 | 0.29 |
| Category | 0–5 yrs | 6–10 yrs | 11+ yrs | Spread |
|---|---|---|---|---|
| Culture | 4.62 | 4.46 | 4.56 | 0.16 |
| Structure | 4.27 | 3.97 | 4.17 | 0.30 |
| People | 4.11 | 3.84 | 4.16 | 0.32 |
| Systems | 4.20 | 3.83 | 4.08 | 0.37 |
Your framework already anticipates this. Discover Culture is the ground every church is standing on before you arrive, and the data says they are largely standing on it well. Design Structure, Develop People and Deploy Systems are where churches actually diverge from one another, and therefore where an engagement moves the number.
Said plainly: the archive shows three quarters of your framework is aimed exactly where the variation lives. That is a claim you have been making from experience. It is now a claim you can make from 528 responses.
Three dimensions clear the 4.50 mark that your own tiering calls a Strength.
I understand the mission and vision of my church.
I know how my role fits into the mission of the church.
I am inspired to do my best work.
“I see myself still serving at this organization in two years” averages 4.4981. Displayed to two decimals it reads 4.50 and looks like a Strength; measured against the threshold it is Healthy by two thousandths of a point.
It is a small thing, and it is exactly the kind of call a hand-built deck settles by eye and a platform settles by rule. Worth deciding once, deliberately, and then never thinking about again.
Every dimension where your people report needing more than they are currently receiving, ranked, in your four categories.
| # | Dimension | Category | Performance | Importance | Gap | Tier |
|---|---|---|---|---|---|---|
| 1 | Mission & vision understood | Culture | 4.84 | 4.88 | −0.04 | Strength |
| 2 | Values guide us daily | Culture | 4.44 | 4.82 | −0.38 | Healthy |
| 3 | Role fits the mission | Culture | 4.61 | 4.70 | −0.10 | Strength |
| 4 | Inspired to do best work | Culture | 4.57 | 4.36 | 0.21 | Strength |
| 5 | People we serve feel our values | Culture | 4.25 | 4.85 | −0.60 | Healthy |
| 6 | Clarity on role expectations | Structure | 4.18 | 4.73 | −0.55 | Healthy |
| 7 | Trained to do my role well | Structure | 3.92 | 4.44 | −0.52 | Developing |
| 8 | My team works collaboratively | Structure | 4.41 | 4.78 | −0.37 | Healthy |
| 9 | Teams coordinate across depts | Structure | 4.15 | 4.76 | −0.61 | Healthy |
| 10 | I have what I need to do my role | Structure | 3.98 | 4.67 | −0.69 | Developing |
| 11 | We develop our people | People | 3.77 | 4.77 | −1.00 | Developing |
| 12 | Clear priorities for my role | People | 4.22 | 4.73 | −0.51 | Healthy |
| 13 | Encouraged to bring ideas | People | 4.16 | 4.32 | −0.16 | Healthy |
| 14 | I get regular feedback | People | 3.81 | 4.25 | −0.44 | Developing |
| 15 | Accountable for progress | People | 4.33 | 4.24 | 0.09 | Healthy |
| 16 | Operating priorities are clear | Systems | 3.99 | 4.61 | −0.62 | Developing |
| 17 | Consistent communication | Systems | 3.84 | 4.46 | −0.62 | Developing |
| 18 | Time & energy spent well | Systems | 3.91 | 4.81 | −0.90 | Developing |
| 19 | We use our systems | Systems | 3.90 | 4.50 | −0.60 | Developing |
| 20 | I see myself here in 2 years | Systems | 4.50 | 4.31 | 0.19 | Healthy |
“We regularly develop our team members for growth.”
It is the only dimension in the set where the distance between what people need and what they receive reaches a full point. The written comments point the same way: 100 of the 1,663 written answers raise development directly, putting it fourth behind communication, leadership and clarity.
Three of the five sit in Deploy Systems. Addressing this cluster would give a church back the thing every one of these dimensions is really asking for: the confidence that the work is organised well enough to be worth their effort.
The archive is large enough to say something specific about which kinds of church have the most room to gain, and it is not the ones most people would guess.
Systems 3.85 and Structure 3.99, the widest gaps of any size band, while Culture sits at 4.54, level with everyone else. One reading, and it matches what these churches say in their comments: large enough that informal coordination has stopped carrying the load, not yet large enough to have built what replaces it.
Structure 3.97, People 3.84, Systems 3.83 are the widest gaps of any age band in every category, Culture included, though Culture moves least even here. Respondents in this window describe running on the energy that started the church and looking for something repeatable to hand to the next layer of leaders.
Multisite churches score higher than single-site on Structure (4.16 against 4.10), People (4.14 against 4.00) and Systems (4.08 against 3.99). Going multisite appears to force the structural work that a single campus can defer, which is a genuinely useful thing to be able to tell a church considering it.
We built one, looked at it, and took it out. Several states in the archive are represented by a single church. One state has twenty-three responses, all from one survey window, all from one demographic profile. Ranking states would have published that church's score under its own name in everything but the label.
State-level comparison becomes safe at roughly two or more distinct engagements per bucket, which is exactly what the region field is for and exactly why filling it in matters. Until then the honest cut is size, age and structure, which is what is above.
The written answers and the recommendation score point the same direction, and it is a hopeful one.
| Response | Count | Share |
|---|---|---|
| Strongly agree | 332 | 62.9% |
| Somewhat agree | 117 | 22.2% |
| Neutral | 51 | 9.7% |
| Somewhat disagree | 26 | 4.9% |
| Strongly disagree | 2 | 0.4% |
| Theme | Answers mentioning it | Share of answers |
|---|---|---|
| Communication | 267 | 16.1% |
| Leadership | 216 | 13.0% |
| Clarity | 168 | 10.1% |
| Development | 100 | 6.0% |
| Volunteers | 97 | 5.8% |
| Vision | 84 | 5.1% |
| Training | 66 | 4.0% |
| Systems | 65 | 3.9% |
| Processes | 61 | 3.7% |
| Culture | 54 | 3.2% |
85% of everyone who has taken this survey agrees they would recommend their church as a place to work or serve. “I see myself still serving here in two years” is one of the highest-scoring dimensions in the set.
And the theme raised most often across 1,663 written answers is communication, ahead of leadership, clarity and development. Not commitment. Not belief in the mission. Not whether people want to be there.
That is a considerably easier conversation to walk into than the alternative, and it is the same conclusion your own reports reach church by church, now with the whole archive standing behind it.
Each of these is a small, bounded fix, and each one unlocks a cut you have already described wanting.
All 528 rows carry a State, and 15 states are represented. Every one of those rows has an empty Region. The “how do we compare to churches in our part of the country” cut, the one you named as the differentiator, cannot currently run. State is populated, so region is derivable; this is a lookup table, not a research project.
The export has no column identifying which church a response belongs to, which means every published average is weighted by staff count rather than by church. A church that returned 75 responses currently counts more than eight times as much as one that returned 9. The scores move modestly when that is corrected, moving Systems from 4.03 toward 3.99 on the closest proxy available, but the principle matters more than the movement, because it is the difference between “the average staff member” and “the average church.” Those are two different benchmarks and both are worth publishing.
“I would recommend this organization as a good place to work or serve” is answered by every respondent and appears on at least one report slide, but it lives outside the twenty-item instrument. It is arguably the single most quotable number you have. It should be a first-class part of the model.
The current twenty-dimension set and a legacy fifteen-dimension set are both present in the archive, and one report in it works from the fifteen. These are not the same questions renamed. Several are substantively different, so a full mapping is not possible for every item. What is possible, and what year-over-year needs, is an agreed decision from you about which dimensions are genuinely comparable and which should simply be shown as new.
Not two. We checked all twenty. Between the respondent sheet and the summary tab:
| # | Respondent sheet | Summary tab |
|---|---|---|
| Q7 | training to do my role well | training to do my job well |
| Q10 | everything to do my role well | everything to do my job well |
| Q12 | clear priorities for my role | clear priorities for my work |
| Q15 | progress in my role | progress in my work |
| Q16 | the operating priorities are clear | the work priorities are clear |
| Q20 | still serving at this organization | still working at this organization |
Harmless in a spreadsheet. Not harmless once it is software, and question 20 is the one that matters most: serving and working mean different things to the volunteer leaders who make up a fifth of your respondents.
The archive is not fully de-identified. Several reports still carry church names in the slide copy, one includes a place name alongside a founding date, and one carries an attendance-and-staffing profile specific enough to identify the church to anyone who knows the region. Program names appear in a couple of others that narrow the field considerably.
Nothing has gone wrong. This is simply what happens when de-identification is done by hand under time pressure. It is also a good argument for making the scrub a step the system performs rather than a step a person remembers, which is straightforward to build and worth building before this archive becomes the engine behind anything.
Right now this benchmark is a thing we built by hand, once, from an archive. The whole point of the platform you have already started is that it stops being a thing anyone builds by hand.
When a church finishes its cycle, its results join this picture automatically. The church sees where it sits against churches its size, its age, its region and its stage. You see the whole field move. And the sentence that opens every report, here is how you compare to twenty-five churches like yours, becomes a sentence nobody else in your space can say.
That is not a new product. It is the product you have already built, with its data layer working and its reports in your brand.