Why phone calls fail with international prospects
A call to an international applicant fails for three reasons that stack on top of each other. You are rarely both awake and free at the same time, the attempt costs money whether or not anyone answers, and the person on the other end has spent years learning not to pick up a number they don't recognise. None of the three is a training problem for your admissions team β they are structural, and they apply to every school competing for candidates outside its home market.
UK institutions feel this acutely because their biggest international pipelines sit furthest away on the clock. India and China together account for 47% of all non-EU international students in UK higher education, according to HESA's 2024/25 student statistics. India runs 4.5 or 5.5 hours ahead of the UK depending on the season, China 7.5 or 8 hours ahead. A call slot that suits an admissions officer in Manchester at 10am lands at teatime or midnight for the candidate they are trying to reach.
This article sets out the three reasons calls under-perform with international prospects, and what changes when the first outbound touch moves to WhatsApp. It complements our pillar guide on recruiting more students in higher education.
Three reasons a call does not work across borders
Time zones turn one call into a scheduling puzzle
A single call to an international prospect is rarely a single call β it is a scheduling negotiation across two calendars that barely overlap. An admissions officer working UK office hours (roughly 9am-5pm GMT/BST) shares a workable window of two or three hours with a candidate in Delhi or Lagos, and close to zero overlap with one in Shanghai, Seoul or Los Angeles. Missed windows turn into a second attempt, then a third, each one competing with the candidate's own school day, work shift or sleep.
Email and web forms dodge the scheduling problem but pay for it in latency: the median reply to an online enquiry lands in about 47 hours, largely because responses wait for office hours on both ends before anyone can act on them (Source: Skolbot mystery-shopping audit, 2025, 80 institutions β see our UK response-time benchmark for the full breakdown by channel). A channel that does not require both parties to be present at the same moment removes the scheduling problem rather than working around it.
Every attempt costs money, whether or not anyone answers
Outbound calling has a cost floor that does not depend on whether the call connects. Twilio's 2026 pricing analysis puts outbound call-centre agent time at $10-50 an hour β a cost the school absorbs on every dial, answered or not. Layer on the connect rate. A 2026 analysis of more than 200,000 B2B cold calls found a connection rate of just 16.6% (Cognism State of Cold Calling 2026, via Scrap.io) β roughly five attempts in six end in no answer, voicemail or an immediate hang-up. Do the arithmetic and the effective cost of one actual conversation runs to several times the headline hourly rate, before any international call surcharge is added.
WhatsApp Business messages carry a per-conversation cost, not a per-attempt one β a school pays to open a conversation, not to redial a number that never picks up. For a recruitment team working a shortlist of several hundred international candidates a term, that difference compounds fast.
Prospects aged 16-25 do not answer numbers they do not recognise
Younger applicants have been trained by a decade of spam and scam calls to treat an unrecognised number as something to ignore, not answer. Industry telecom data backs this up at scale: Hiya's State of the Call report found that only around half of all calls received on a mobile phone get answered, and the figure drops sharply further for numbers the recipient has never saved β a pattern consistent across the markets Hiya tracks. A landline-style outbound call from an admissions office, arriving from a number the applicant has never seen, competes for attention with exactly the kind of call most 16-25-year-olds have learned to screen out by reflex.
Phone calls to prospects are answered only 34% of the time, and an answered call runs to about 3 minutes 20 seconds on average β against an AI chatbot response of about 3 seconds, 24/7 (Source: Skolbot mystery-shopping audit, 2025, 80 institutions). A channel where the recipient already recognises and trusts the app β because friends, family and, increasingly, other institutions already message them there β starts the conversation without that trust deficit.
What changes when the first message is WhatsApp, not a call
The direct answer: a WhatsApp message does not need a shared time slot, does not cost anything extra when it goes unread, and arrives inside an app the prospect already checks daily rather than a call log they've learned to ignore. ICEF Monitor's analysis of WhatsApp in the international student journey describes it as "the most immediate and practical starting point" for institutions recruiting internationally, citing open rates as high as 98% β applicants who would never place an international call are far more willing to reply to a free, familiar message.
The table below sets the three channels side by side for the specific job of first contact with an international prospect.
| Channel | Typical response window | Answer / open rate | Cost pattern | Time-zone dependency |
|---|---|---|---|---|
| Outbound phone call | 3min 20s when answered | 34% (Skolbot); ~50% of all mobile calls generally (Hiya) | Per-attempt, answered or not | High β needs a shared live window |
| Email / web form | ~47h median | High eventual open, slow reply | Near-zero marginal cost | Low, but reply still waits on office hours |
| WhatsApp message | Minutes, asynchronous | Up to 98% open (ICEF Monitor) | Per-conversation, not per-attempt | Low β read and replied to on the prospect's own time |
| AI chatbot on WhatsApp | ~3s automated first reply, 24/7 | Available at any hour, any time zone | Fixed platform cost, not per-message | None β always on |
The pattern across the table is the same one that shows up in every mystery-shopping run: a channel that requires synchronous presence loses to one that doesn't, and a channel with a trust deficit loses to one without it. Automating the first reply on WhatsApp β rather than staffing it manually around the clock β is what makes 24/7 coverage realistic for a team that isn't rostering night shifts. Our WhatsApp chatbot article on the Gen Z channel covers the deployment side of that in more detail.
Who you are actually trying to reach on WhatsApp
An international prospect pool is not one audience with one preferred language, which is exactly why an unstaffed WhatsApp inbox or a single-language script fails almost as often as a phone call does. Across Skolbot's partner-school base, 58% of international prospects are non-French-speaking β the panel breaks down to French 42%, English 28%, Spanish 11%, Arabic 7%, Portuguese 4%, Mandarin 3%, German 2% and 3% other languages (Source: automatic language detection across 8,500 Skolbot conversations, 2025-2026). Campus France's 2024 figures independently found close to 45% of candidates were not francophone, a separate dataset landing on the same broad conclusion.
This panel skews toward French and other European institutions rather than UK ones specifically, so treat the exact split as directional β but the underlying shape (a majority-non-native-language pool, split across half a dozen languages) matches what UK recruiters see too, given that India and China alone already account for 47% of non-EU international students (HESA, cited above). A WhatsApp channel only closes the phone-call gap if it can hold a conversation in whatever language the prospect opens with β see our guides on multilingual AI chatbots for international students and on building a multilingual chatbot for international recruitment.
How to introduce WhatsApp without dropping the phone entirely
WhatsApp becomes the default first contact for international outreach, and the phone stays as the channel for conversations that genuinely need a live human voice β a scholarship negotiation, a visa concern, a candidate who has explicitly asked to be called. Replacing every call with a message is not the goal; removing the phone as the default first attempt for a cohort that mostly won't pick up is.
A workable rollout has three steps. First, move initial outreach onto WhatsApp, opened through an explicit opt-in on the enquiry form rather than an unsolicited first message. Second, let an AI chatbot handle tier-one questions β fees, entry requirements, deadlines, visa document checklists β in whichever language the prospect writes in, so a reply lands in seconds rather than waiting for the next time zone to align. Third, route anything needing judgement or a scholarship discussion to a named human admissions officer, with the full conversation history attached so the candidate never repeats themselves. The chatbot complements the admissions team rather than replacing it: it absorbs repetitive, factual volume so human time goes to conversations where a human genuinely adds value.
Calls do not disappear β they move from default first attempt to a deliberate, scheduled step once a candidate has already engaged and wants to talk. That single change addresses all three structural problems at once: no shared time slot is required to start the conversation, no cost is wasted on unanswered dials, and the first message lands on a channel a 16-25-year-old candidate already trusts enough to open.
FAQ
Does WhatsApp actually work better than phone calls for international student recruitment?
For first contact, yes, because it removes the two biggest structural failures of a call: the need for both parties to be free at the same moment, and the trust deficit around numbers a prospect does not recognise. Phone calls remain useful once a candidate has already engaged and asks to speak to someone directly.
What does WhatsApp cost compared to international phone calls?
WhatsApp Business messaging is priced per conversation, so a school pays when a conversation actually happens, not for every dial that goes unanswered. Outbound call-centre agent time runs $10-50 an hour according to Twilio's 2026 pricing data, and with only around one call in six connecting, the effective cost of a single completed conversation by phone is several times the headline hourly rate before any international call surcharge is added.
Why don't Gen Z applicants answer the phone?
Years of spam and scam calls have taught most 16-25-year-olds to treat an unrecognised number as something to screen out, not answer β a pattern industry telecom data confirms at scale, with roughly half of all mobile calls going unanswered generally and the figure dropping further for unsaved numbers (Hiya State of the Call report). WhatsApp sidesteps the problem because the app itself, not the specific number, is already trusted.
Should a school abandon phone calls entirely for international recruitment?
No. The goal is to stop using the phone as the default first attempt, not to remove it from the process. Calls still matter for scholarship discussions, visa concerns and any candidate who explicitly asks to talk β they just work better as a scheduled follow-up once WhatsApp has already opened the conversation.
Does a WhatsApp chatbot need to support every language a prospect might use?
It needs to cover whichever languages make up the bulk of a school's international pipeline, which is rarely just one. Across Skolbot's partner-school base, 58% of international prospects write in a language other than the school's primary teaching language, split across roughly half a dozen languages rather than concentrated in one β a single-language script misses most of that pool.
See how schools are improving recruitment


