Fill Your Cohort in 90 Days With a Small Two-Person Team

Ninety days out from census date with a cohort still short is not a problem two people solve by working longer hours — it's a problem they solve by refusing to work most of the list in front of them. A two-person admissions team has one genuine advantage over a bigger, better-resourced competitor: nobody has to sit through a committee meeting to decide what stops getting done. Below is the plan, broken into three 30-day phases, with the trade-offs a lean team has to make explicit rather than left to whoever answers the phone first.

Days 90-61: audit and triage before anyone makes a single outbound call

The first 30 days are not for outreach — they're for deciding which prospects in the CRM are worth the two people's limited hours, because working the wrong list burns time that can't be replaced later. Most CRMs are full of dormant enquiries, incomplete applications, and open day registrants who never showed. Sorting that list is the highest-leverage task in the whole 90 days: every hour spent chasing someone who was never going to enrol is an hour not spent on someone who might.

Start by mapping where the current cohort's applicants are actually dropping out of the funnel, not where the team assumes they are. Across a broad benchmark of prospect journeys, the pattern is consistent: 91% of website visitors never make first contact, 64% of those who do never submit an application, 42% of applicants never sign up for an open day, 35% of open day registrants don't show up, 28% of attendees never complete their file, and only 18% of completed files convert to a final enrolment — an overall visit-to-enrolment conversion of roughly 0.8% (source: content/zpd-bank.json#prospect-dropout-funnel). The biggest single leak is the first one: visitors who never get a reply. That's where a two-person team should look first, because it's the stage cheapest to fix and the one currently losing the most volume.

Pull three lists out of the CRM this week: dormant enquiries older than 30 days, applications started but never submitted, and open day registrants who never attended. Rank each by recency and by whether the prospect gave a phone number or only an email — both predict responsiveness better than anything else in a sparse CRM. Anyone who hasn't opened an email in 60+ days goes to the bottom of the pile. If the team has never looked at drop-off by stage rather than headline enquiry count, run a proper student recruitment funnel audit now — the next 60 days only work if this triage was accurate.

What to cut, not just what to keep

Cutting is the harder half of triage and the part most teams skip. If a segment converts below 2%, has no recent engagement, and needs manual outreach to revive, don't work it — not because it's worthless, but because reviving it costs more hours than the enrolments it's likely to produce in time. Park it for the next cycle. The goal isn't a smaller list for its own sake; it's a list two people can work end-to-end without burning out in week four.

Days 60-31: activation — automate the first reply, save the humans for high intent

The middle 30 days are for reactivating the highest-potential segments from the audit while making sure the two people never personally handle a first response again. Response speed is the single biggest lever here, and the one a two-person team cannot deliver manually at volume. Response time varies enormously by channel: an AI chatbot answers in roughly 3 seconds and is available 24/7, against phone at 3 minutes 20 seconds when someone actually answers — which happens only 34% of the time — human live chat at 8 minutes but business hours only, email averaging 47 hours, and a contact form averaging 72 hours (source: content/zpd-bank.json#response-time-by-channel). A dormant prospect re-engaged by a campaign and then left waiting two or three days for a human reply is functionally back to square one.

This is where a chatbot widget earns its place: it answers the reactivated prospect's first question the moment they click through, qualifies whether they're worth a human callback, and only then routes the high-intent conversation to one of the two team members. That's a division of labour, not a replacement — the chatbot handles volume and first response, the humans handle the judgement calls a borderline application or a family-specific fee question still needs.

Run exactly one tightly scoped outreach push in this window, not three campaigns a two-person team can't staff or measure properly. Choose the channel with the best signup economics rather than the most familiar one. Benchmarked across acquisition channels, open day signup rate runs 18.4% for a chatbot on the school website, against 6.2% for a contact form, 4.8% for an email campaign, 3.7% for paid social, and 2.1% for organic social (source: content/zpd-bank.json#jpo-registration-by-channel). For a team with no capacity to run a paid social campaign properly, a chatbot-driven push converting at roughly three times the contact form rate is the more decisive use of the window.

Line up an open day or information session that lands inside the final 30-day sprint, since attendance there feeds the conversion phase. If your program follows a change-of-preference cycle after ATAR results, time it to land just before or during that window, when prospects are re-ranking preferences across UAC, VTAC, QTAC, SATAC, or TISC and a fast, personal answer can decide between two similar offers. See the recruit more students pillar and the 12-month admission campaign timeline for how a single well-timed push fits a full-year plan.

Days 30-0: the conversion sprint — cut no-shows, apply yield management, don't burn the team out

The final 30 days convert registered interest into a signed enrolment, and the single most preventable loss in this window is the open day or advising session nobody shows up to. Follow-up method changes the no-show rate dramatically: no follow-up at all leaves a 52% no-show rate; email-only follow-up brings that to 38%; SMS-only to 31%; a personalised chatbot follow-up to 19%; a combined chatbot-plus-SMS sequence to 14%; and a personalised reminder naming the prospect's specific program to 11% (source: content/zpd-bank.json#jpo-no-show-rate). A two-person team can't personally call every registrant the week of the event — an automated sequence that fires a personalised reminder with the specific program name, not a generic template, is the difference between a half-empty open day and one that runs near capacity.

Once attendees are in the room or on the call, yield management decides how many convert to a firm enrolment: prioritise follow-up by genuine intent signal — fee, accommodation, or Commonwealth Supported Place questions are late-stage signals — over chronological order, and give the team a short, non-negotiable script for the handful of concessions it's allowed to offer, so nobody is improvising a decision that should have been agreed in week one. The yield management guide covers converting a confirmed offer into a paid, enrolled student without discounting your way there.

Protect the two people's hours as deliberately as the prospect list. A sprint that burns out the only two staff who can answer a complex query in week 12 defeats its own purpose — build in one full day off each per week even during the final push.

How two people should actually spend 90 days

The honest allocation across each phase — what the two humans do personally, and what should already be running on automation so their hours aren't spent on it.

The pattern is the same across all three phases: humans handle judgement and relationship, automation handles volume and repetition. A team that tries to do both loses the sprint by week six — not because the plan was wrong, but because the hours ran out before the list did.

What a two-person team should not attempt in 90 days

Be explicit about what's off the table, because an unstated scope is how a lean team quietly overcommits. Don't run more than one outreach channel at once — a parallel email, SMS, and paid social campaign each needs someone to build and monitor it, and that's capacity two people don't have. Don't redesign the application form mid-sprint; fix friction points after the cohort is full. Don't chase every dormant contact equally — the triage in the first 30 days exists to stop that. And don't promise a personal follow-up call to every open day registrant; automation is already doing that job so the humans can focus on prospects who need an actual conversation.

TEQSA expects accurate, timely information for prospective students, which a lean team can only sustain by automating the repeatable share and reserving personal attention for what genuinely needs it (TEQSA). If your program is FEE-HELP or Commonwealth Supported Place eligible, keep fee and loan information current before the sprint starts — exactly the kind of question a chatbot should answer instantly, since Study Assist expects providers to keep this current for every enrolling student (Study Assist). For international cohorts, check your institution's current allocation under the government's managed international education arrangements before an overseas-facing push, since 2026 places are capped per provider (Department of Education).

What this looks like in aggregate

No single tactic above is dramatic on its own. Combined, institutions running automated first response alongside this triage and follow-up discipline have measured qualified prospects rising from 120 to 195 a month, a 62% increase, alongside a 38% reduction in cost per qualified prospect and a 12-month ROI of 280% (source: content/zpd-bank.json#chatbot-roi-metrics). Those figures are a broader benchmark, not an Australia-specific cost-per-enrolled-student study — treat them as a directional case, and model your own cost per qualified prospect before quoting a return to a finance committee.

FAQ

Can two people really fill a cohort in 90 days without hiring?

Yes, if the plan cuts scope rather than adding effort — triage the CRM hard in the first 30 days, automate first response so the humans only handle high-intent conversations, and run exactly one outreach push instead of several. The constraint isn't available hours; it's how ruthlessly the list gets cut before any outbound work starts.

What should the two-person team stop doing immediately?

Stop replying to every enquiry personally, and stop chasing cold, low-signal contacts at the same priority as recently engaged ones. Both consume hours the sprint can't spare, and both are jobs an automated first-response system and a properly triaged list already solve.

How much of the funnel drop-off is actually recoverable in 90 days?

The largest single leak — a 91% drop-off between a website visit and first contact — is also the fastest to close, since chatbot-enabled institutions cut it to 76% purely through faster response. That's the highest-leverage fix available in a short sprint, well before yield-management tactics on the back end.

What's the biggest mistake a lean team makes under a 90-day deadline?

Working the full CRM list instead of the triaged one. Treating every dormant contact as equally worth a phone call spreads two people's hours across prospects who were never going to enrol, at the expense of the smaller, higher-intent group the audit should have surfaced.

A 90-day cohort-fill sprint with two people only works if automation is already doing the volume work before day one. See how the full plan fits a year-round admissions calendar in the 12-month admission campaign timeline, and start from recruiting more students without adding headcount if this is the team's first lean sprint.