How to Audit Your Church Website Like a First-Time Visitor (The 10-Question Test)

On a slow Tuesday afternoon, the office administrator at a mid-sized church opens a private window on her phone and tries to find her own church.

She has worked there nine years. She orders the bulletin paper. She knows which key opens the side door and which one only looks like it does. For the next twenty minutes she is going to pretend she has never heard of the place.

It goes worse than she expected. Google shows the church with a service time that changed in the spring. The homepage still has a banner for Vacation Bible School, which ended in July. She finds the real times in the footer, in small gray type. The "Plan Your Visit" button opens a form asking for her name, her email, and how many people are in her party, and nothing else. Nowhere on the site does it say whether the nursery is open at the early service, which is a little funny, because she is the one who schedules the nursery volunteers for the early service.

She writes it all down on the back of an envelope. It is the most useful twenty minutes anyone has spent on that website in years.

This post is that exercise, written out so anyone on your staff or welcome team can do it in an afternoon. You need a phone, a private browser window, a timer, and a willingness to be slightly embarrassed. At the end you will have a short list of fixes sorted by how much they matter, and a much clearer picture of what a stranger goes through before they ever meet you.

Why you cannot just look at your own website

The problem with reviewing your own church's website is that you already know every answer on it.

In 1990, a Stanford graduate student named Elizabeth Newton ran a small experiment that has since become famous. She asked one group of people to tap out the rhythm of well-known songs on a table, things like "Happy Birthday," and asked a second group to name the song from the tapping alone. Before they started, the tappers guessed their listeners would get about half of them. Out of 120 songs tapped, the listeners got three.¹

The tappers were not bad at tapping. They just could not stop hearing the melody in their heads, so they could not imagine what it was like to hear only knocking.

That is what it is like to look at your own church's website. You hear the song. You know the 9:30 is the quieter service. You know "the Commons" means the big room downstairs. You know the front doors stay locked on Sunday and everybody comes in from the parking lot side. None of that is on the page, and you cannot see that it is missing, because to you it is not missing. It is just known.

A first-time visitor hears only knocking, and the website is usually the first thing they hear. Thom Rainer, who has spent years sending "secret guests" into churches, says about seven in ten first-time guests now tell his team they went to the church's website before deciding where to visit.² Pew found the habit is strongest among younger adults: of people under 30 who had looked for a new congregation, 59 percent searched online.³

And they are judging more than the information. When Stanford researchers asked 2,684 people to explain whether they trusted a website, the most common thing their comments mentioned was how the site looked, in 46 percent of comments, ahead of how the information was organized or what it actually said.⁴ A visitor who cannot find your service time does not just lack a service time. They have formed an opinion about you.

Churches have been doing this in person for years

None of this is a new idea. It is just a new door.

Since 1998 the British Christian site Ship of Fools has sent volunteers it calls Mystery Worshippers into churches where nobody knows them, to report back on what it was like to walk in as a stranger. The questions are wonderfully plain. How warm was the welcome? How long was the sermon? What happened when you hung around afterward looking lost? How cold was the coffee?⁵ Thom Rainer's Church Answers runs a "secret guest" program built on the same instinct, and its survey starts before the visit, with the website: "Does it provide the information you need to get to the church on time? What conclusions do you reach about the church based on its website?"⁶

Plenty of churches have done a version of this on a Sunday morning, with a friend from across town walking in the front door and taking notes. Very few have done it on the door most visitors actually come through first.

What you need

• A phone. Not your office computer. Most people find a church on a phone, often in a parking lot or late at night. If you can borrow someone else's phone, even better. Yours remembers too much. (More on this in mobile-first church websites.)
• A private browser window. In Safari it is called Private; in Chrome it is Incognito. It does not sign you into anything and does not use your history, so your browser will not helpfully autocomplete your church's address the way it does every Monday. It is not a perfect stranger's view. Google still estimates roughly where you are,⁷ which is fine, because your visitors live nearby too.
• Cellular data, not the church wifi.
• A timer and something to write on. The back of an envelope is traditional.
• About three hours, though not all of it on the website. Part of the test is waiting for a reply.

Pick who you are pretending to be

The same website can work fine for one stranger and fail completely for another, so decide who you are before you start. Write it on a sticky note. A few to choose from:

• A mom with a two-year-old and a six-year-old who grew up going to church with her grandmother and has not been inside one in twelve years.
• A seventy-four-year-old widower who just moved to town to be near his daughter. He uses a cane and does not hear well in big rooms.
• A college sophomore who has never been to a church service in her life. A friend invited her and she is curious, and a little nervous.
• A couple who left their last church after something painful and are quietly looking for a new one. They know exactly what they are looking for, and what they are not.

Each of them searches differently. The mom types "church with nursery near me." The widower types "Methodist church" and the name of your town. The couple types your denomination and "near me" and reads the reviews. A Catholic coming back after years away types "Mass times near me." Type what your person would type.

The rules

1. Start at the search, never at your homepage. Do not type your web address. Your visitor does not know it.
2. Sixty seconds per question. If you cannot find the answer in a minute, it counts as not found, even if it turns out to be sitting on page four. A real visitor does not keep looking for four minutes. They tap the next church in the results.
3. Count your taps. Anything that takes more than three is worth writing down.
4. Think out loud, and have someone take notes. Every "hm," every "wait, where is," every time you scroll back up. The usability researcher Jakob Nielsen has called thinking aloud "the single most valuable usability engineering method,"⁸ and it is the one that costs nothing.
5. Do not fix anything yet. The urge to stop and correct the typo is strong. Resist it. The moment you log into the website editor, you stop being a stranger.

The ten questions

These are the ten things a first-time visitor is trying to find out, roughly in the order they try. Most of them will look obvious written down. Do the test anyway. Obvious is exactly what we stop being able to see.

1. Is this the right church, and is it still meeting?

Before you tap anything, look at what Google shows you. Most visitors meet your church first in the search results and on the map, not on your homepage, and the business profile panel there shows hours, photos, reviews, and a link that often nobody at the church has looked at in years.

Passing looks like: the right name and address, service times that match reality, a link that goes to your current site, and photos from this decade.

Where churches usually fail: the hours listed are the office hours, Monday through Friday, so to a stranger it looks like you are closed on Sunday. A parish that merged or renamed still shows up under the old name. A campus that closed still has a pin. Google is plain that those hours are what people see in Search and Maps, and it suggests confirming special hours around holidays even when nothing changes.⁹ The full walkthrough is in how to show up when people search "churches near me".

2. When is the next service I could actually come to?

Not "what are your service times." The next one, with a date. If today is Thursday, can you tell from the site what happens this Sunday, and whether this Sunday is a normal one?

Passing looks like: you find it in under a minute, and the time is the same everywhere it appears: Google, the homepage, the footer, the Facebook page, the plan your visit page.

Where churches usually fail: the summer schedule is in a PDF. The Christmas Eve times go up on December 22. The Saturday vigil Mass or the Wednesday night service is only on the calendar page. The Spanish service is only on the Spanish page. And the worst one, two different times in two different places, because now the visitor does not trust either. Rainer tells the story of two guests who arrived late to their first service because the church's website listed the wrong time.¹⁰ They did the work. The website let them down.

3. Where do I go when I get there?

Tap the address. Does the pin land on your building, or on the church office across town, or in the middle of a field? Then the harder questions. Where do I park? Which door?

Passing looks like: the address opens a map with the right pin, and the site tells you where to park and which door to use. A photo of that door is even better.

Where churches usually fail: the photo of the building is the one taken from the street, of the beautiful front doors nobody uses. Members know the front doors are locked on Sunday morning. Visitors walk up to them, pull, and stand there while people they do not know walk past to the side entrance.

4. How long is it, and what actually happens?

How long does it last? Do people stand, sit, kneel? Are the words on a screen or in a book? Is there a point where people go forward? What happens at the passing of the peace? None of this is obvious to someone who has never been, or has not been in a long time, and the not knowing is what makes people nervous.

Passing looks like: a plain description of the service, with how long it lasts.

Where churches usually fail: "Join us as we worship," and nothing else. Liturgical churches especially tend to assume people know the shape of the hour. A sentence like "Everything we say together is printed in the worship aid, and it is fine to just stand when others stand" does more for a nervous visitor than a paragraph about your heart for worship. There is a whole rewrite of this exact page in church website copy that actually sounds human.

5. What do I do with my kids?

For the mom on your sticky note, this is the whole visit.

Passing looks like: a parent could walk in knowing where to take each child, how check-in works, which services have childcare, whether kids start in the service and leave partway through for a children's sermon or Children's Liturgy of the Word, and what happens if a child needs them.

Where churches usually fail: "We have an amazing children's ministry!" with a photo of a bouncy castle from last year's fall festival. Rainer lists missing children's and student information among the most common website mistakes his teams find.²

6. What kind of church is this?

Can a stranger tell your tradition, and roughly what you believe, without a theology degree? A lot of church names say nothing about denomination. Sometimes that is a deliberate choice. Either way, the visitor wants to know, and a church that makes them dig reads as a church that is hiding something.

Then there is the question almost nobody asks out loud: what am I allowed to do here? Communion is the clearest example, because the answer really does differ from church to church. In a Catholic parish, other Christians are ordinarily not admitted to Communion and are invited to pray for unity instead.¹¹ In an Orthodox parish only Orthodox Christians receive, though visitors are often offered blessed bread at the end of the liturgy.¹² Missouri Synod Lutherans practice close communion.¹³ United Methodists invite all the baptized to the table, members or not.¹⁴ The Episcopal Church welcomes all baptized Christians to receive.¹⁵

Whatever your practice is, a visitor should be able to learn it on your website, calmly, at home. Not when the plate is coming down the row or the line is moving toward the front and they have about eight seconds to decide what to do with their hands.

Passing looks like: your tradition named plainly, a short statement of beliefs in everyday words, and a few sentences on what guests are and are not expected to do.

Where churches usually fail: a beliefs page that is a scanned confession, or no beliefs page at all. We wrote more about this in why church AI has to know your denomination, and the same is true of pages.

7. Will I stick out?

What do people wear? Who goes here, families or retirees or college students or everyone? Will I be asked to stand up, fill out a card, give money?

Passing looks like: dress described in one honest sentence, recent photos of the people who actually attend, and a line that says guests are not expected to give.

Where churches usually fail: stock photos, and "come as you are" with nothing after it. Every church says come as you are. A visitor learns nothing from it.

8. Can I get in the building, and can I take part?

More than one in four adults in the United States lives with some kind of disability.¹⁶ That is the widower with the cane, the mom whose son uses a wheelchair, the man who cannot hear the sermon without a hearing loop, the woman with celiac disease wondering about Communion.

Passing looks like: even three lines. Is there a step-free entrance, and which one? Accessible parking? An elevator to the kids' rooms? Assistive listening devices or a loop? Large-print bulletins? A gluten-free option? A quiet room?

Where churches usually fail: saying nothing at all. Here is the uncomfortable part. In a 2025 Lifeway survey, 98 percent of Protestant pastors said a person with a disability would feel welcomed and included at their church.¹⁷ The website is usually where that confidence quietly goes unproven, because the person who needs to know cannot find out before they come, and so does not come.

9. Is anyone home?

Strangers read stale pages as a sign about the church, not just the website. They are not wrong to.

Passing looks like: nothing on the first two screens has already happened. The latest sermon or news item is from this month. The staff page matches the staff.

Where churches usually fail: the Easter banner still up in October, which Rainer mentions so often it is practically a genre.² A staff page with a youth pastor who left two years ago. A "latest sermon" from 2023. A copyright line in the footer that says 2019.

10. If I have a question, will a real person answer?

This one has two parts, and the second is the one nobody does.

First, find a phone number and an email in under a minute, ideally with a person's name attached. Second, actually use them. From a personal email the church office will not recognize, send the question your sticky-note person would really ask. "Is there somewhere for my two-year-old at the 9:30?" Write down the time you sent it. Then call the office at a time a visitor might, say 4:30 on a weekday, and listen to what happens.

Passing looks like: a reply from a human within one business day, and a voicemail greeting that says who you have reached and when someone will call back.

Where churches usually fail: no reply for three days. A mailbox that is full. A greeting recorded by the previous pastor. A contact form quietly delivering to an inbox nobody has opened since spring. If this is where you lose points, the 48-hour rule is the next thing to read.

The scorecard

Copy this into a notes app or print it. Score each question 2 if you found the answer in under a minute, 1 if you found it eventually, and 0 if you did not find it.

#	The visitor's question	What passing looks like	Score
1	Is this the right church, and is it still meeting?	Google shows the right name, address, Sunday hours, and current site
2	When is the next service I could come to?	Date and time in under a minute, identical everywhere
3	Where do I go when I get there?	Correct map pin, where to park, which door
4	How long is it, and what happens?	Plain description of the service and its length
5	What do I do with my kids?	Ages, check-in, which services, what happens if they need you
6	What kind of church is this?	Tradition named, plain beliefs, what guests can and cannot do
7	Will I stick out?	Dress, real recent photos, no pressure to give
8	Can I get in, and can I take part?	Entrance, parking, hearing, large print, gluten-free
9	Is anyone home?	Nothing expired on the first two screens
10	Will a real person answer?	Human reply within one business day

The total matters less than the zeros. Each zero is a question that a real person asked about your church this week and did not get answered.

How to spend the afternoon

1:00 to 1:15. Set up. Borrow the phone, open the private window, turn off wifi, pick your person, write it on the sticky note.

1:15 to 2:00. Run the ten questions. Out loud, with a colleague taking notes. Send the email and make the phone call while you are on question ten.

2:00 to 2:45. Hand it to an actual stranger. This is the part that makes the whole afternoon worth it, and it gets its own section below.

2:45 to 3:30. Sort what you found. Three piles, explained further down.

Before you leave. Put the next one on the calendar. Once a quarter is plenty, plus a quick pass a few weeks before Christmas and Easter, when the most strangers are looking.

Then hand it to someone who really is a stranger

Pretending gets you most of the way. An actual outsider gets you the rest, and you need fewer than you think. Nielsen's well-known rule of thumb is that testing with about five people turns up roughly 85 percent of a website's usability problems.¹⁸ Even one finds a lot.

Ask a neighbor. The barista who knows your order. A parent from your kid's soccer team. A teenager, who will be merciless and fast. Not a member, and ideally not someone who has ever been to your church. Hand them the phone with the search page open, give them the ten questions, and then do the hardest thing in this whole post. Keep your mouth shut.

Do not help. Do not explain. Do not say "oh, it's under the About tab." Just watch their thumb, and write down every place it hesitates. The places it hesitates are your list.

Then go find the people who were strangers most recently: the families who started coming this year, the folks in your newcomers' class, your membership class, your OCIA group, your confirmation class. Ask them one question. "What couldn't you find on our website before you came the first time?" They remember. Most of them have just been too polite to say.

What to fix first

Sort everything you wrote down into three piles.

Fix today. Anything wrong is worse than anything missing. Wrong service times, a wrong map pin, an expired banner, a broken link, a voicemail box that is full. Most of these take ten minutes.

Fix this month. Anything missing from questions 1 through 5. Those decide whether anyone comes at all. A kids page, a plan your visit page that answers real questions (here is the anatomy of a good one), parking and door instructions, the next service with a date.

Needs a decision. Some fixes are not really website fixes. Writing down your Communion practice for guests, deciding which entrance to unlock on Sunday, committing to answer email within a day. Take those to whoever decides, your pastor or your council or your vestry or your elders, with your notes in hand.

The man reading alone on the road

There is a story in Acts that church people of nearly every tradition know. An official from Ethiopia is riding home in his chariot, reading the prophet Isaiah out loud. Philip runs up alongside and asks him, "Do you understand what you are reading?" And the man says, "How can I, unless someone guides me?" Then he invites Philip up to sit with him.¹⁹

He had the text. He had come a long way, and he was clearly sincere. What he did not have was someone sitting beside him who already knew what it meant.

A church website is often a stranger reading alone on the road. Everything they need might technically be in there somewhere, written by people who love the place and know it by heart. This afternoon is you climbing up into the chariot for a little while and reading it next to them, slowly, the way they have to.

The question your audit will not find

One more honest thing. An afternoon with a pretend stranger finds the questions you thought to ask. Real visitors bring the ones you did not. Is it all right to come by myself? My son has autism; will he be okay in the kids' room? I'm divorced; am I still welcome to receive? Can my mom come if she uses an oxygen tank?

You cannot write a page for every one of those, and you should not try. That gap is the reason we built Greetyr, a digital greeter that reads your church's own website and answers a visitor's questions in your words, at whatever hour they ask. It also keeps a list of the questions it could not answer well from what your site says, grouped by theme, so your staff can see them. In a way it is this audit, running every week, with real strangers instead of a pretend one.

But do the afternoon first. It is free, it is humbling, and it will tell you more about your front door than anything we could sell you.

Frequently asked questions

How do I audit my church website?

Open a private browser window on a phone, using cellular data, and pretend you have never heard of your church. Start from a Google search rather than your own web address, and give yourself sixty seconds to answer each of ten questions a first-time visitor asks: whether the church is still meeting, when the next service is, where to go, what happens, what to do with kids, what kind of church it is, whether they will stick out, whether the building is accessible, whether the site is current, and whether a real person answers. Write down every place you hesitate, then repeat the test with someone who has never been to your church.

What should a church website include for first-time visitors?

At minimum, the next service with its date and time, the address with parking and entrance instructions, how long the service lasts and what happens during it, childcare and kids' check-in details, the church's tradition and a plain statement of beliefs, what guests are and are not expected to do, accessibility information, and a named person to contact with a phone number and email. Service times and address matter most. They are the information guests most consistently say they look for and cannot find.

How often should a church audit its website?

Do the full ten-question test once a quarter, plus a quick check a few weeks before Christmas and Easter, when the most new people are searching. Service times, staff, and events should also get a short monthly check, because outdated information does more damage than missing information. A visitor who finds one wrong detail starts to doubt everything else on the site.

What is a church secret shopper or mystery worshipper?

It is someone who visits a church anonymously as a first-time guest and reports back on the experience. The idea goes back years in church circles. Ship of Fools has published Mystery Worshipper reports since 1998, and Church Answers runs a secret guest program. Most of these programs focus on the Sunday visit itself, but the best ones start with the website, because that is where most guests start.

Why use an incognito or private window to review a website?

A private window is not signed into your accounts and does not use your browsing history, so your browser will not autocomplete your church's address or show you the pages you visit every week. That puts you closer to a stranger's view. It is not perfect. Google still estimates your general location, and you still know the church. But it removes the shortcuts you do not notice you are taking.

Who should do a church website audit?

Ideally two kinds of people: someone on staff or the welcome team who can act on what they find, and at least one real outsider who has never been to the church. The staff member catches what is wrong. The outsider catches what is missing, which is the part insiders cannot see. Recent newcomers are also worth asking, since they remember what they could not find before their first visit.

What is the most common church website mistake?

Hard-to-find service times and address. In secret guest research, it is the complaint that comes up again and again, and it is the simplest one to fix. Put the next service, the address, and where to park on the homepage, and make sure they match what Google shows.

---

Greetyr is a digital greeter for church websites. It answers visitors' real questions in your church's own words, day or night, and hands anything pastoral to the people in your building who should be holding it. If you are working through your site after this audit, start with the church website mistakes costing you visitors and the questions every church visitor is already asking.

Footnotes

1. Newton, E. L. (1990). The Rocky Road from Actions to Intentions (Doctoral dissertation, Stanford University). Tappers predicted listeners would identify about 50% of the songs; listeners identified 3 of 120 (2.5%). Popularized in Chip Heath and Dan Heath, "The Curse of Knowledge," Harvard Business Review (December 2006). See: hbr.org. ↩

2. Thom S. Rainer, "The Front Door Churches Often Forget," Lifeway Research (March 6, 2018), excerpted from Becoming a Welcoming Church (B&H, 2018): "Seven out of ten guests will go to a church website as a determinative factor in where they will choose to visit." Rainer describes this as objective and anecdotal research from asking guests, not a controlled study. The same piece lists hard-to-find address and service times, outdated information, and missing children's information among the most common mistakes. See: research.lifeway.com. ↩ ↩² ↩³

3. Pew Research Center, "Choosing a New Church or House of Worship" (August 23, 2016). Among all adults who had looked for a new congregation, 37% searched for information online; among those under 30, 59% did. The report also notes that some of the factors people value most "are difficult to assess over the phone or on a website." See: pewresearch.org. ↩

4. Fogg, B. J., Soohoo, C., Danielson, D., Marable, L., Stanford, J., & Tauber, E. R. (2002). How Do People Evaluate a Web Site's Credibility? Results from a Large Study. Stanford Persuasive Technology Lab and Consumer WebWatch. 2,684 participants; "design look" appeared in 46.1% of the 2,440 comments, ahead of information design (28.5%) and information focus (25.1%). Published in peer-reviewed form at DUX 2003. See: consumerreports.org (PDF). ↩

5. Ship of Fools, "The Mystery Worshipper." Launched in 1998, with more than 3,000 published reports. Its own description: "How long was the sermon? How hard was the pew? How warm was the welcome? How cold was the coffee?" See: shipoffools.com. ↩

6. Church Answers, Secret Guest Survey. Section A, completed before the visit, asks whether the church's website "provide[s] the information you need to get to the church on time" and "What conclusions do you reach about the church based on its website?" See: churchanswers.com (PDF). ↩

7. Google Chrome Help, "Browse in Incognito mode": Chrome "doesn't automatically sign you into your Google Account or other websites," though Incognito "doesn't make you invisible." Google Search Help notes that "Google will always estimate the general area that you're searching from." See: support.google.com/chrome and support.google.com/websearch. ↩

8. Nielsen, J. (2012). "Thinking Aloud: The #1 Usability Tool." Nielsen Norman Group, quoting his 1993 book Usability Engineering. See: nngroup.com. ↩

9. Google Business Profile Help. A profile lets you "provide your business hours, website, phone number, and location so they're visible to customers on Google," and Google advises: "It's a good idea to confirm your hours for official holidays, even if they're the same as your regular hours." See: support.google.com/business. ↩

10. Thom S. Rainer, "The Top Ten Worst Church Guest Experiences," Church Answers (August 11, 2025). See: churchanswers.com. ↩

11. United States Conference of Catholic Bishops, "Guidelines for the Reception of Communion." Members of other churches "are ordinarily not admitted to Holy Communion," and those not receiving "are encouraged to express in their hearts a prayerful desire for unity." See: usccb.org. ↩

12. Orthodox Church in America, Guidelines for Clergy (2023), on the Eucharist being received by Orthodox Christians. Practice around the blessed bread (antidoron) varies by jurisdiction and parish; many offer it to all visitors. See: oca.org (PDF). ↩

13. The Lutheran Church, Missouri Synod, "Frequently Asked Questions: Denominations." "The LCMS practices 'close communion.'" See: lcms.org. ↩

14. Joe Iovino, "An open table: How United Methodists understand communion," United Methodist Church. "The baptized present are all invited, even if they belong to a different church." See: umc.org. ↩

15. The Episcopal Church, Canon I.17.7, and diocesan guidance such as the Diocese of Virginia: "All baptized persons are invited to share in Communion. If you are not baptized, you are welcome to come forward for a blessing." See: episcopalvirginia.org. ↩

16. Centers for Disease Control and Prevention, Disability and Health Data System. Based on 2022 Behavioral Risk Factor Surveillance System data, 28.7% of U.S. adults report a functional disability, more than 1 in 4. See: cdc.gov. ↩

17. Aaron Earls, "Churches Believe They're Welcoming to Those With Disabilities," Lifeway Research (August 11, 2026). Phone survey of 1,003 U.S. Protestant pastors, September 2025. See: research.lifeway.com. ↩

18. Nielsen, J. (2000). "Why You Only Need to Test with 5 Users." Nielsen Norman Group. Based on the model in Nielsen & Landauer (1993), "A mathematical model of the finding of usability problems," INTERCHI '93. See: nngroup.com. ↩

19. Acts 8:30-31 (ESV). ↩