Sojourn is a research product with real users. The question we are testing is whether AI can put people in front of scripture and then fade away. This page is where we keep score. It helps us be honest with our users and ourselves. It guides us towards new experiments to try and how to evolve existing features.
¶
What’s the premise?
Most AI products are built to hold your attention. We think that is exactly backwards for this one. Scripture is the thing worth your time, and a good guide walks you to the door and stops talking. So we measure Sojourn by one thing: how much of the time in the app is spent reading the Bible.
¶
How do we measure?
Sojourn does not require your name, email, location or any other personally identifiable information. We also never store any of the questions you ask or messages you send. What you read and the trail you leave stay on your phone. You can read more in our privacy policy.
¶
How we are doing?
We choose the following measurements based on the limited data we collect. We feel these measurements best capture how well Sojourn is tracking towards our goal of Redemptive AI.
Scriptural accuracy
100%
✓done
Making sure that every scripture shown to a user is 100% accurate was a non-negotiable before the app was ever published. Our design ensures that it's technically impossible for Sojourn to misquote cited scripture. All scripture, whether it's embedded in the chat response or in the chapter reader, comes directly from a World English Bible database that's bundled into the app.
Visits that reach scripture
30%
↑we want this up
Percent of app opens where one or more scriptures are read. Ultimately we want this to be 100%. An app open that does not reach scripture is not something we consider a successful engagement. It's not bad, per se, but it's not our goal.
Time spent in scripture
7.7%
↑we want this up
Percent of time spent in Sojourn reading scripture. This number goes up when people make it all the way into the chapter reader. It's the most immersed you can get into scripture in the app. We don't expect this to ever reach 100%, but our goal is for it to increase as much as possible.
Peeks that become real reading
40%
↑we want this up
Percent of peeks into scripture that result in the whole chapter being opened. The peek is meant to indicate tepid interest. Dismissing a peek without reading the full scripture is not considered bad. We want exploration to feel lightweight. But once you do find a verse which resonates we absolutely want you to read the surrounding text.
Turns before a verse opens
2turns
·context only
Median number of chat turns before scripture is read. We don't have a directional goal for this number. But we keep an eye on it because if it starts to increase then it indicates that people are becoming more engaged with the chat and we may want to look into it more closely.
Typical length of a visit
3.0min
·context only
Median session length when a user opens the app. We don't care about this metric in isolation. However, a lengthy session where someone is reading multiple scriptures is an indication that they are deep in the Word.
¶
Have questions?
We are learning. Reach out if you have any questions or ideas. We’re at hello@trysojourn.app.