Ed Tech Trends
Does pre-arrival community engagement actually improve enrolment? What the evidence shows
.png)
At Uni-Life we build pre-arrival communities for universities. The question we are asked more than any other is whether it works. Does this actually move enrolment?
It is a fair question, and it deserves a straight answer. The short version: there is good evidence that pre-arrival communities do things worth doing, and there is no evidence that anyone in this market can isolate their causal effect on the enrolment decision itself. Those are two different claims, and the gap between them is where most of the confusion in this category lives.
Here is how we read the evidence, and what we think you should measure instead.
Sit through four demos and you will see roughly the same chart four times. Students who joined the community enrolled at 84%, or 88%, against a general population at some much lower number.
Those numbers are real. What they are not is proof that the platform caused the difference.
They compare students who chose to join a community against everyone who did not. Students keen enough to download an app, make a profile, and start talking to future classmates in March are already, on average, more likely to turn up in September. They were more likely before they ever encountered the platform.
This is the oldest problem in social science, and it has a name: self-selection. It runs underneath essentially every outcome statistic in this market, ours included. We are not pointing at anyone here. We are describing how the whole category measures, and we measure the same way. Our own figures come from students who opted in, without a control group, and we think it is better to tell you that than to let the number do work it cannot support.
That does not make the numbers useless. It gives them a description of an engaged cohort rather than a measure of impact. Read them as the former and they are quite informative.
The most rigorous work here is worth knowing about.
Bird, Castleman and colleagues (NBER working paper 26158, 2019) ran two randomized controlled trials of light touch engagement, nudges to admitted students designed to reduce summer melt, across more than 800,000students. Randomized, so no self-selection. Control groups, so a real comparison.
They found no impact on enrolment, overall, or in any subgroup.
A text message nudge is not a community platform, and we would not pretend it is. But the lesson generalizes: once selection bias is removed, diffuse effects tend to shrink, and what survives testing is usually acute and deadline shaped. Anyone working on belonging should take that seriously rather than argue with it.
Choosing where to study is one of the biggest decisions a young person makes. Money, family, visas, rankings, a partner, a city, or a scholarship that came through or did not. A pre-arrival community is one input among dozens.
Isolating any single input causal contribution to that decision is close to impossible, and not only for us. It is hard for a university, hard for an independent research body, and the literature agrees.
So, we do not build our story on it. Enrolment is distal: it happens months later, after a hundred other things, and it is not ours to claim. That is a deliberate position rather than a gap in our data.
The useful half is what sits closer.
Universities are not only trying to increase enrolment. They are also trying to reduce uncertainty, answer repetitive questions more efficiently, identify emerging issues before arrival, and help students feel connected long before Welcome Week. A pre-arrival community supports all of these objectives, even if its individual contribution to enrolment cannot be isolated.
What happens in the community is proximal, real and verifiable from your own dashboard:
None of these are enrolment. All of them are honest, and all of them matter.
And here the evidence is genuinely encouraging. The link between early social connection, belonging, and student outcomes is one of the better-established findings in higher education research. A student who arrives already knowing three people has a different first term from one who arrives knowing nobody: for their wellbeing, their confidence, and their odds of staying on the course.
We believethat helps them turn up too. We think the mechanism is sound, and theproximal data points the right way. We just will not dress a belief up as ameasurement.
Ask what any number is comparing. For any figure a vendor shows you, ours included, ask about the control group. If the comparison is "students who joined versus students who did not," read it as a portrait of an engaged cohort, not as impact.
Check which product earned it. Some of the most quoted statistics in this category were produced by a different product in the vendor's range than the one being sold to you. Ask plainly which one.
Do not hold out for causal proof. It is not coming, and it would probably not decide it for you anyway. Only around 7% of edtech suppliers use randomised trials, and buyers overwhelmingly decide on pilots and peer references. Causal proof and peer proof are different things, and itis peer proof that moves committees.
Run a small pilot and watch the proximal signals. One intake, one faculty. Adoption, real interactions, what students say, what it costs your team in hours. That is evidence you own, about your students, needing no trust in any vendor.
The honest state of this category is a plausible mechanism, encouraging proximal data, strong research on why belonging matters, and no causal proof of the number everyone puts on the slide.
We would rather be clear about which of those we can stand behind. It makes the things we do claim worth more.
If it would help, we are happy to walk through what the data looks like for a university of your size, including where our claims stop.
