The state of
post-16 admissions2026
Of every 100 applicants who start a post-16 application, only 19 walk through the door in September. This report uses live data from 903 post-16 settings to show exactly where the other 81 go, and what the settings losing fewer of them do differently.
What this report found, in three minutes
903 post-16 settings, 517,000 applications, one admissions cycle. If you read nothing else, read this page.
The headline
Applicaa is the UK's largest post-16 admissions platform: 1 in 4 schools and colleges run applications, offers, enrolment, events and marketing through one system. This report is drawn from real usage data across that network.
Our figures show that of every 100 applicants who start a post-16 application, only 19 walk through the door in September. The other 81 don't disappear all at once. They leak out at every stage, and the biggest leaks are ones most settings can actually close: how fast you reply, whether you nudge accepters before September, and whether you run a taster day at all.
Three things worth doing
Don't let response time drift past three weeks. Not "be instant": the data doesn't support that. Just guarantee a reply inside three weeks.
Keep closing the melt gap. "Melt" is the term for a student who says yes to a place and then never actually turns up in September. It's moved before (48.7%→47.9%), and a proactive nudge between "yes" and September is the most direct lever.
Run a taster day, and make reference turnaround visible. A first taster day is linked to a 7.6-point rise in acceptance, and most settings still haven't run one. Reference turnaround (how long a feeder school takes to send a reference) is invisible to most admissions teams today; the slowest quarter of settings wait ten weeks for one to come back.
Full detail, methodology and the sector context behind all of this follows below. Scroll on, or use the section list opposite.
What 903 settings' live data tells us
This report uses real records from Applicaa's post-16 network, not a survey: every application, offer, acceptance and enrolment across the 2025/26 admissions cycle (1 September 2025 to 31 August 2026), for students starting post-16 education in September 2026. Where a figure is measured directly from the network it's tagged network; where it's a published external statistic it's tagged sector, with a source. A prior-year comparison appears wherever the data supports one.
How we define "post-16"
A post-16 setting is one running a live post-16 application form (state-funded or college/FE). Independent post-16 admissions work differently: families often apply to several at once and complete one, so they're shown as a separate panel in Section 02, not blended into the headline network figures.What's covered, section by section
- 0104The landscape903 post-16 settings, 517k applications, and the capacity squeeze behind them.
- 0205The conversion funnelOf every 100 who start an application, only 19 enrol in September. Here's where the other 81 go.
- 0306Enrolment: the summer meltRoughly 1 in 2 accepters never enrol, even after they've said yes.
- 0407How fast schools say yesNot a straight line: a three-week cliff, not a gradient.
- 0508CommunicationsEmail volume keeps climbing, so SMS is worth a second look as the channel that still gets noticed.
- 0610Open days, visits and taster days1,694 events run this cycle, and the one event type with real proof it works.
- 0712References and predicted gradesA small, slow, easily-missed bottleneck, properly measured.
- 0813The sector under pressureDemographics, funding, NEET, and the demographic cliff ahead.
- 0915What's nextThree priorities, grounded in what the data actually shows.
- 1016Self-assessmentThree real network benchmarks, and four things worth asking yourself.
The network grew 11.6% in a year.
Applications grew faster still.
903 post-16 settings ran live admissions on Applicaa through the 2025/26 cycle, up from 809 the year before. Applications outgrew the network: up 15.4%, against 11.6% more settings, so the average setting is still doing more with the same admissions team. That 903 is deliberately the more conservative count: 957 settings have post-16 admissions switched on, but we only count the 903 that actually had an applicant this cycle. The gap between the two isn't necessarily inactive settings: some of it may be settings that didn't use the enrolment feature at all this cycle, which would mean their data doesn't reflect a genuine full cycle and could otherwise have skewed the figures below.
Applications per setting, 2025/26 network
How many applications a post-16 setting handles in a cycle varies hugely. Ranking all 903 settings from quietest to busiest: the quietest 25% of settings received 259 applications or fewer, the typical (median) setting received 418, the busiest 25% received 673 or more, and the busiest 10% received 1,082 or more. This is the whole network together, without a further split by state-funded vs college/FE; independent settings are already excluded here and covered on their own in Section 02.
Applications received per setting in the 2025/26 cycle, ranked quietest to busiest
The capacity gap, still open
DfE workforce data shows admin headcount in state schools has stayed essentially flat over the same years post-16 application volumes have climbed. This cycle, applications grew 15.4% against 11.6% more settings joining the network. The typical admissions team is still doing more work with the same people.
19 out of every 100 now enrol in September
Up from 18 the year before. Growth wasn't just more applicants at the top of the funnel: every stage converted a little better than it did a year ago.
Toggle between cycles · figures are per 100 starts, that cycle
"Started" means a family opened the application form, whether or not they finished it. "Completed" means exactly that: the form was fully submitted with every required field filled in.
The genuine finding here
Completion rate held roughly flat (66%→65%), but the offer rate climbed: from 73% of completed applications getting an offer last cycle to 77% this cycle. In absolute terms, offers grew 19.5% against completions growing only 13.9%, meaning settings are converting a finished application into an offer more often than they did a year ago. That compounds all the way down to enrolment, which grew fastest of all (+21.2%).Independent post-16: still a different shape
Families apply to several independent settings and complete one; offers routinely exceed completions. We don't show an enrolled figure for this segment. Most independent settings don't track enrolment status the same way in-platform.
| Stage | 2024/25 | 2025/26 |
|---|---|---|
| Started | 100 | 100 |
| Completed | 31 (30.6%) | 47 (46.7%) |
| Offers made | 61 (199% of complete) | 60 (129% of complete) |
| Accepted | 33 (54% of offers) | 33 (55% of offers) |
Worth watching
Independent completion rate leapt from 30.6% to 46.7%. Nearly half of families who start an independent post-16 application now finish it, up from under a third the year before. Offer-per-completion eased back toward a more typical ratio (199%→129%) as a result.How many students hold more than one accepted offer?
Within a single school, almost never: only 339 of 184,919 accepters (about 0.2%) held more than one accepted offer at the same setting. Across different Applicaa schools, it's far more common. Matching students by email address across the whole network, 22,634 of the 153,618 students who accepted an offer somewhere (14.7%) also held an accepted offer at a different Applicaa school in the same cycle. That doesn't fully explain the melt in Section 03, but it's a real part of it: some students who "melt" at one setting have simply enrolled at another Applicaa school, not vanished from education altogether. Email-matching across schools isn't perfect (shared family addresses can create false matches), so treat this as directionally right rather than exact.Still roughly 1 in 2 accepters never enrol
Accepting an offer isn't the finish line. "Melt" is the term for a student who says yes to a place and then never actually turns up in September. They change their mind, move away, choose somewhere else, or simply drift. Section 02 has a genuinely relevant caveat here: some of this "melt" is students who enrolled at a different Applicaa school, not students lost altogether.
What it costs an average setting
At current volumes, roughly 101 accepted students per average post-16 setting never enrol, against 96 a year ago. That's more in raw count because the network is bigger, but a slightly smaller share of accepters. At the ESFA's national 16-19 funding rate of £5,026 per full-time student for 2025/26,1 that's roughly £508,000 of annual income sitting in the gap between "yes" and September, for the same applicant pool the setting already won.
Not a straight line: a three-week cliff
It's tempting to assume "the faster you reply, the more you convert." Grouping settings by how fast they typically reply, then comparing how well each group converts, shows something more specific, and more useful.
Offer-to-accept rate, by how fast the setting typically responds
Three similar bars, then a drop: a cliff, not a gradient
| Speed band | Settings | Offer → accept |
|---|---|---|
| Under 24 hours | 96 | 73.3% |
| 1-7 days (best band) | 323 | 79.8% |
| 1-3 weeks | 140 | 78.0% |
| Over 3 weeks (the cliff) | 83 | 58.8% |
What the data actually says
It's a cliff, not a gradient. Everything inside three weeks converts alike (73-80%); cross three weeks and acceptance drops 19-21 points. The sub-24-hour band converts worse than the week-long band, so instant isn't the goal. The right instruction is "never let it drift past three weeks," not "be as fast as possible."Speed buys the acceptance. It does nothing for the melt.
Running the same bands against accept-to-enrol, the rate sits between 46% and 52% in every band, with no direction. A setting replying in an hour and one replying in three weeks lose students to the summer melt at about the same rate: fixing response time wins the acceptance, not the melt (Section 03).
Email volume keeps climbing. SMS is still almost untouched.
Schools across the Applicaa network sent 13.8 million emails to families this cycle, up 35% on the year before. The more email a family gets, the easier any single message is to miss, including the ones that matter most, like an offer or a decision deadline. Text messages don't have that problem yet, simply because almost nobody sends them: just 32,408 texts went out across the whole network this cycle, against those 13.8 million emails.
The channel almost nobody uses
As inboxes get more crowded, every individual email has to work harder to get noticed. A text message doesn't have that problem: because so few schools send them, a text is far more likely to actually be read. That gap between how much email families get and how little SMS they get is exactly why SMS is worth trying, even at small volumes.
Reading the shift
More settings are trying SMS (+25%) than are sending meaningfully more of it (+12% more texts). That's light, occasional use, not a core channel yet. Families increasingly expect a text alongside an email, so this gap is worth closing even at a small scale.
The one event type with proof it moves the needle
Open days and evenings are the backbone of the recruitment calendar, alongside smaller trial and private visits. Settings choose which type of event they're running when they set it up, so the split below reflects how each setting itself categorised the event.
| Event type | Events run, 2025/26 |
|---|---|
| Open day | 774 |
| Open evening | 638 |
| Trial visit | 189 |
| Private visit | 93 |
A trial visit is typically a short taste of one course or subject, ahead of a full taster programme (below). A private visit is a one-to-one tour for a family, usually outside the main open day season.
Taster days: the one move with real proof behind it
A taster day gives an applicant a real day in your setting: lessons, subject taster sessions, a feel for the place, before they commit. It's the closest thing in this report to a controlled experiment: 60 settings that ran their first-ever taster day saw offer-to-accept rise 7.6 percentage points in that same cycle, same staff, same applicant pool, nothing else changed.
The case is already made: this is about who's still not doing it
The within-setting evidence (+7.6pp accept, +13.9pp enrol, from a same-staff same-pool before/after study) is about as clean as this kind of proof gets in admissions data. Adoption is growing, but slowly: 633 settings still haven't run one. Section 09 covers what to do about it.
The real number is far smaller, and far slower
References and predicted grades are the slowest step in the entire journey, and the easiest to underestimate.
Easy to conflate with the wrong number
The network-wide reference count, spanning every admissions phase Applicaa runs, is 1.29 million records: two to three orders of magnitude bigger than the post-16-specific figure above, and easy to reach for by mistake. It also understates the real problem: post-16 references take almost twice as long to come back as the network average (34.2 days vs 17.5 days). Post-16 is where this bottleneck genuinely bites hardest, which the blended network number hides rather than reveals.Three forces pressing in on every admissions office
These figures come from DfE, ESFA and ONS publications, not the Applicaa warehouse. Pick any one and the case is uncomfortable; together they compound.
This cycle's growth (903 settings, +15.4% applications) is the dividend the DfE's own projections say has about four more years left to run. Settings investing in application experience now are building the operational strength that retains pupils once it ends.
Three priorities, updated against what actually moved
Don't let response time drift past three weeks
Not "be instant": the data doesn't support that. Everything inside three weeks converts alike (73-80%); past it, acceptance drops 19-21 points. Automate enough to guarantee a reply inside three weeks, then stop optimising this lever.
Keep closing the melt gap
"Melt" is the term for a student who says yes to a place and then never actually turns up in September. It improved slightly this year (48.7%→47.9%), proof it's movable, not proof it's fixed. Response speed alone won't shift it (Section 04); a proactive nudge between "yes" and September is the more direct lever.
Get post-16 reference turnaround visible, then under three weeks
Now properly measured for the first time: 34.2 days median, nearly double the network-wide average. 633 settings still haven't run a taster programme either. Both are proven levers with most of the network not yet using them.
Three real benchmarks. Four things worth asking.
| Question | 2025/26 benchmark |
|---|---|
| Most families accept your offer within 3 days of receiving it? | Post-16 median 3.2 days (unchanged) |
| Application data flows into your MIS without re-typing? | 676 settings on a live SIMS sync alone in 2025/26; more again across Arbor, Bromcom and iSAMS |
| You know what % of your offers turned into enrolments? | Network median: 52.1% |
What else should you be asking?
-
Do your offer letters generate and send themselves?Save hours of manual admin. Here's how it works in Applicaa: offer letters auto-generate and send the moment an offer is approved, pulled from your own template, no manual drafting per student.
-
Do you send at least one update between offer and decision deadline?Keep accepters engaged automatically. Here's how it works in Applicaa: reminder emails and texts can go out ahead of the decision deadline without a staff member tracking the date.
-
Can your leadership team see live application numbers?Give leadership real-time visibility. Here's how it works in Applicaa: the dashboard shows a live view of applications, offers and enrolments, no manual report needed.
-
Have you completed your own application form on a phone in the last 12 months?Meet families where they are. Here's how it works in Applicaa: application forms are fully mobile-responsive, so applying from a phone just works.
How we worked these numbers out
Every figure in this report comes from real records, not a survey.
- Post-16 settings are state-funded schools and colleges running sixth-form admissions. Independent post-16 admissions work differently (families apply to several and complete one), so they're shown separately in Section 02 rather than blended into the headline figures.
- The cycle runs 1 September to 31 August. This report covers 2025/26, compared against 2024/25 wherever both are available.
- References (Section 07) are linked to the specific post-16 application they support, not just counted network-wide across every type of admissions Applicaa runs.
- Response time (Section 04) groups settings by their own typical reply speed and compares how each group converts, rather than assuming one straight-line relationship.
Melt and response-time figures cover state-funded and college/FE settings only. Figures from outside the network (DfE, ESFA, ONS, IFS) are cited and tagged sector.
Data refreshed nightly. Snapshot used here: 7 September 2026, the first working day after the 2025/26 cycle closed.