One controlled pilot, and what it does and doesn’t show.

The pilot

Between 21 October and 1 December 2025, Svenstrup School in Aalborg ran the method across its whole intake: 800 students in 40 classes, 40 teachers, grades 0–9. The design was stepped: one cohort of ten teachers ran it from week 0 with a matched comparison cohort that did not; a second pair of cohorts swapped in from week 4. Students answered validated five-item scales weekly (94% completion); teachers reported weekly (100%); absence came from school records.

The primary comparison is the first three weeks, intervention against control. Effect sizes were large (Cohen’s d of 1.42 for cohesion and 1.38 for engagement). The cohort that ran the full six weeks finished at 4.2 on cohesion (+35.5% from baseline) for the cohort that ran the full six weeks.

What it does not show

The study lists its own limitations before its conclusion, and they are the right ones: one school; six weeks; mostly self-report; classes not randomly assigned; participants knew they were studied; the principal investigator works for the company. Participants also knew they were being studied, so some of what was measured is the Hawthorne effect.

So the honest claim is small. One school, six weeks, eight hundred children: the relationships in the rooms that ran this changed measurably more than in the rooms that did not, and fewer children stayed away. That is a signal, not a proof. Larger studies with random assignment and longer follow-up are needed, they have not been done, and anybody telling you otherwise about any product in this category is selling you something.

Where this sits in ESSA’s tiers

US districts buy against four evidence tiers, and vendors are careless with the words, so here is ours in plain terms. The Svenstrup pilot is a quasi-experimental design with matched comparison cohorts, which is the design ESSA associates with Tier 2 (Moderate). It does not meet Tier 2, because Tier 2 also requires the study to run in at least two sites and this ran in one school. It is not Tier 1: nothing here was randomised.

What we can defend today is Tier 4 — “demonstrates a rationale”: a logic model grounded in published research, with an evaluation completed and further evaluation planned. That is the tier we will write on a form. If a district needs Tier 2 or above, we do not have it yet, and you should ask every other vendor in the room for their study, their sample size, and how many sites it covered.

One more thing an evaluator should weigh: this was a Danish school. Grade bands, contract hours and the school day are not American, and the children are not a US demographic. A result there is a reason to try it in your building, not evidence about your building. A three-week pilot gives you a number that is actually about your students.

What the method stands on

The mechanism is older than the product. Structured, reciprocal self-disclosure between strangers produces measurable closeness inside a single short session — the “fast friends” procedure of Aron and colleagues (1997), replicated many times since. Mutual gaze does related work. The method is a delivery mechanism for those two things, run by an ordinary teacher on a timetable.

The wider case for teaching these skills in school rests on Durlak, Weissberg and colleagues’ 2011 meta-analysis of 213 school-based programmes covering 270,034 students, which found an average gain of 11 percentile points in academic achievement alongside improved behaviour and reduced distress. That is evidence for the category, not for this product; it is why the category exists.

Belonging is one of the strongest predictors of attendance and effort in the OECD’s PISA data, and the 2023 U.S. Surgeon General’s advisory on loneliness recommends peer-led programmes in schools. None of that proves VEN Method works. It explains why we bothered to test it.

What we will not say

We will not quote a percentage without its control figure and its source. We will not call the method “proven”, “certified”, “evidence-based” without naming the tier, or “#1”. We will not describe a study as approved by the What Works Clearinghouse, which reviews studies and never products. We will not claim it prevents crises, detects trauma or predicts anything about a child — it does none of those things and is built so that it cannot. If you find a claim on this site that breaks these rules, it is a mistake: tell us.

Sources