The average church announcement script scores 45 out of 100. We scored 23,518 of them to find out.
Since 2017, churches have sent Pro Video Announcements the written announcements they want turned into a weekly video. Most send one a week for a season, some for years. That is 424 churches and about 55 scripts each. In October 2026 we scored every one of those scripts against the 100-point rubric we use to coach churches, then had a second, stricter model re-read the top of the pile and a random slice of the whole. The aggregates behind every number here are public.
Two things to say before the numbers. First, we sell the fix. PVA produces the videos and coaches the writing, so read this knowing that; the method is at the end so you can check our work. Second, the score measures the writing against a rubric, not what happened in the room on Sunday. The rubric is built on the research we know of about what makes a listener act, and story and the feeling it creates carry a quarter of the points for a reason we come back to. We have no attendance or sign-up data to set beside the scores, and we say so.
Here is what the scores say. Nine in ten scripts fall short of the line the rubric calls Effective. The thing almost every script is missing is a story. That is a default, repeated in hundreds of churches, and in 2025 it began to change.
The score is 100 points across six categories and ten measures. Each measure is a question the scorer answers about the script, with a one-sentence reason written before the number, and four bands from none to strong; the table gives the lowest and the highest. The weights are the argument: story and feeling 25 points, the next step 25, simplicity and guest access 20, the balance of heart to logistics 15, attention 10, the church's own voice 5.
| Category (points) | Measure (points) | The question the scorer asks | Average script |
|---|---|---|---|
| Story and emotional transport (25) | Story presence (15) | Was there a specific person, moment or outcome I can remember? A testimony, a "here is what happened last time", a name and a face, a quote. 0 to 3 is no story, pure logistics; 12 to 15 is a story the viewer can picture. | 3.2 of 15 |
| Emotional transport (10) | Did I feel something, or just hear something? Joy, anticipation, gratitude, compassion. 0 to 2 is flat; 9 to 10 means the viewer can recall the feeling afterward. | 3.5 of 10 | |
| Next steps and action clarity (25) | Next step clarity (15) | Could I explain the next step to someone in one sentence? 0 to 4 is no clear step or a buried one; 13 to 15 is a step that is plain, easy to remember and feels worth taking now. | 9.5 of 15 |
| Next step singularity (10) | Does each announcement point to one next step, and do they all point to the same place? 0 to 3 is competing asks inside one announcement; 9 to 10 is one ask each, all to the same place. | 4.8 of 10 | |
| Clarity, simplicity and accessibility (20) | Simplicity and focus (10) | Could a viewer sum up each announcement in one sentence? 0 to 2 is several ideas crammed together; 9 to 10 is one idea, no fat. | 6.7 of 10 |
| Guest accessibility (10) | Could someone who has never been to this church understand every word? 0 to 2 is insider language with no explanation ("connect card", "life group", campus shorthand); 9 to 10 needs no prior knowledge. | 5.5 of 10 | |
| Content balance (15) | Celebration, inspiration, information (15) | Does this feel like a church celebrating what God is doing, or a list of events? The target is roughly 40% celebration, 40% why it matters, 20% what, when and where. 0 to 3 reads like a bulletin; 12 to 15 has momentum and heart with the logistics woven in. | 5.1 of 15 |
| Engagement and attention (10) | The hook (5) | Would I stop scrolling for this? 0 to 1 opens with "just a reminder"; 4 to 5 opens with story, surprise, emotion or energy. | 1.9 of 5 |
| Pacing and energy (5) | Does the writing shift energy and tone, or run flat? 0 to 1 is static and monotone; 4 to 5 has intentional shifts that match the content. | 2.5 of 5 | |
| Brand, voice and authenticity (5) | Brand and voice (5) | Could this announcement belong to any church? Does the presenter sound like a real person? 0 to 1 is generic; 4 to 5 is a voice that is unmistakably this church. | 2.9 of 5 |
Read the right-hand column top to bottom and the shape of the problem is already there. The average script earns about two thirds of the points for naming a clear next step and for keeping each announcement simple, so churches are good at saying what, when and where. It earns a fifth of the story points and a third of the balance points. The heart is what is missing, and the rest of this piece is about that gap.
The median script scores 45. Half of all scripts land between 39 and 52, and eight in ten between 33 and 58. About 8% reach 60, the rubric's Effective line. Under 2% reach 70. On the stricter second model's scale almost none do; see How we measured. The share above any line is the softest number in this piece.
Part of that tightness may belong to the scorer. A model grading on a rubric tends to pull toward the middle of its scale, so we do not rest the claim on the spread. We rest it on what the scripts contain. 94% of them are mostly or purely informational: the scorer found a list of events with little celebration of what happened or why it matters. 71% have no story at all. Put together, that is one approach: open with hello, list the events, give the date and the place, point to several sign-up links, say goodbye. It is the announcement most of us grew up hearing, and scripts from every year between 2017 and 2024 score it within a point of 45.
Of the ten things the rubric measures, story is the weakest by a distance. Scripts capture 21% of the 15 points available for it, and fewer than one in two hundred score 12 or more, the band the rubric calls a compelling story.
Story and the emotion it carries are 25 of the 100 points, 15 for the story itself, and that weighting is deliberate. The rubric draws on Paul Zak's research on narrative and action, on Chip and Dan Heath's Made to Stick, on Phil Bowdle's Rethink Communication, and on Brady Shearer's one story plus one next step rule, learned from producing thousands of church announcements. In that work a specific story is the most reliable way to pull a listener out of passive listening, and emotional narrative moves people to act where information alone does not. Some of Zak's findings have been contested in replication, and Shearer's rule is a practitioner's rule rather than a study, so take the weighting as a considered bet. If you accept it, story has to weigh a lot. If you do not, the rest of this piece still stands on the other 75 points.
A story in this rubric is not a sermon illustration. It is ten to twenty seconds about a specific person, moment or outcome: a name, what happened, what changed. The men's breakfast announcement that says forty men came last month and three of them had their first real conversation about faith in years is a story. The one that says Saturday at 8am in the fellowship hall is not. That example is ours, not a church's; we do not publish scripts from the dataset.
The neighboring measures say the same thing. Content balance, which asks whether the script celebrates what happened or only lists what is coming, puts about 1% of scripts in its top band. Emotional transport, which asks whether the viewer felt anything, finds 47% flat. The hook, which asks whether the first line earns attention, averages 1.9 of 5. Information comes first. Story, if it comes at all, comes last.
Churches do ask. Clarity of the next step is one of the strongest measures in the data: the average script scores 9.5 of 15, and a next step is almost always there. The trouble is how many. 24% of scripts have competing asks inside a single announcement. 62% ask for one thing per announcement but send each announcement somewhere different: a website here, a text keyword there, a table in the lobby. Only 14% keep the whole video pointed at one place.
The announcement count tells the same story.
Scripts with two announcements average 49, and 13% of them reach 60. By five announcements the average is 43 and 3.5% reach 60. Scripts with a single announcement sit at 45, below two and three; most are a one-off event promo. One next step per video is the goal, not one announcement. Part of the drop is built into the rubric, which rewards a single ask. But story, simplicity and content balance all fall as the count rises too, and nothing in the rubric forces that. More items may mean less room for the one story that would make any of them land. Two is best. Three is the most the data forgives. Everything else goes in the bulletin, the app, or next week.
Length is not the problem. The average script runs about 400 words, and across all 23,518 the correlation between word count and score is 0.05, which is to say none. The longest scripts, 600 words and up, score a little higher, and they are the band with the most story in it. Trimming words will not move the score. What the words are spent on will.
For eight years the number did not move. From 2017 to 2024 the yearly average stayed between 44 and 46, and story stayed between 3.0 and 3.3 of 15. Then it moved. Scripts submitted in 2025 average 50, scripts from the first nine months of 2026 average 49.5, and story is at 4.1.
The first objection is that the churches changed. The data says mostly not. 27 churches sent at least ten scripts in both 2022 to 2024 and the period from January 2025 to February 2026, all through the same email form, all scored the same way. Their average went from 45 to 50, and 23 of the 27 improved. Churches that stayed with us through 2025 wrote differently in 2025. We cannot say the same about churches that left.
The second objection is harder, and we cannot rule it out: churches started drafting with AI tools in 2025, and an AI scorer may read AI-assisted writing kindly. What we can say is where the gain landed. Half of it is in the three measures that ask for story, feeling and celebration; the measures for the next step barely moved. Whatever the cause, the gain is in content, not polish: more story and celebration, not tidier logistics. A drafting tool could supply that too, so we hold this as an open question.
One thing that does not explain it is our own dashboard. Since March 2026 churches have submitted scripts through the PVA client dashboard, where every script is scored on this rubric, and those scripts average 49, in line with the 50 of the 2025 email scripts. The scores in this piece are for what the church wrote, never for any rewrite. Weekly scoring and Next Steps Engine coaching began in February 2026, after the 2025 rise, so they cannot be what moved it.
The average hides a wide spread. Among the 252 churches with 20 or more scored scripts, the typical church averages 46. The bottom tenth, as a group, average 36. The top tenth average 58. The lowest church averages 29 and the highest 65.
57 of those 252 churches average 50 or more. Six average 60 or more. Those six matter more than the average does. Effective is not a line we drew above everyone's reach. It is where a handful of churches already live, on the same rubric, with the same production as everyone else.
A model's scores are consistent but not infallible, so we had a second, stricter model re-score every script the first one put at 60 or above, 1,787 in all, on the full rubric, and counted a script as confirmed if it also scored 60 or above. It confirmed 70. It was harsher across the board: it agreed with nearly half of the scripts the first scorer put at 80 or above, and with almost none of those it put at 60 to 64. Scripts the first scorer put below 60 were not re-read in that pass, so these 70 are the scripts two readers agree are Effective, not a claim that no others are. A random 500 scripts re-scored by the stricter model showed it ranks scripts much as the first model does, on a lower scale; the numbers are in How we measured.
The 70 look alike. Story averages 10.6 of 15 against 3.2 for the average script. Content balance averages 11.5 of 15 against 5.1. Emotional transport averages 7.1 of 10 against 3.5. They carry three announcements, about the average, and run about 460 words, a little longer than the average, so they are not shorter or sparer than the rest. They spend the words on a person and an outcome, then ask for one thing. And 52 of the 70 were written in 2025 or 2026.
The gap between a 45 and a 60 is not production. Every script here got the same video treatment. The gap is a few decisions made before anyone hits record.
Forty-five is the score of the default, not a verdict on your church. The scripts at 60 had the same production as the 45s. They told one story and asked for one thing. That is what the 60s have in common.
Where does your church land?
Paste your last four weeks of announcements into the free audit and get them scored on the same rubric.
Score My Announcements Free Start My Two Free Sundays