How Singapore Parents Choose Tuition and Enrichment Centres: The Research Behind a $1.8 Billion Decision
A mother in one of our parent interviews scrolled back through her class chat for us, to the week after the PSLE preliminary exams. She wanted to show how she had found her son's maths tutor. It took a while (the chat had more than 400 unread messages that week). Somewhere between a lost water bottle and a reminder about a school form, another parent had written three words about a centre near their block. "Very strict, results good." Nine replies followed within the hour, asking for the number. She had booked a trial class that evening, before she had even looked up the centre's website. "Actually, I never check the website," she said, and laughed, as if that part were embarrassing.
That scroll seems to hold most of the story. Tuition in Singapore is a very large purchase that behaves like a small one. It gets decided in a group chat, confirmed in a single trial lesson, and reviewed with brutal honesty once a year when results come out. Operators I speak with invest in exactly the things parents skip, the polished website, the credentials page, the brochure. This post looks at how the decision really moves, who holds the veto at each stage, and why results season is when the switching happens.
What Singapore families spend on tuition
The scale is worth pausing on. The Straits Times reported that families in Singapore spent $1.8 billion on private tuition in 2023, up from $1.1 billion in 2013 and $1.4 billion in 2018, based on the Household Expenditure Survey. The average household spent $104.80 a month, and the top fifth of households by income spent $162.60 a month, more than four times the $36.30 spent by the bottom fifth. SingStat's own Household Expenditure Survey 2023 report attributes the rise in education spending (from $374 to $404 a month) to pre-primary education and private tuition. I read the averages twice, because they hide the real picture. Most families with a child in upper primary are spending well above $104.80, and plenty of households spend nothing at all.
The Ministry of Education has been fairly direct about its discomfort. In a February 2025 parliamentary reply on private tuition, MOE said the reasons for taking tuition vary, that some students take it even though they are coping well, and that it is "not in a position to intervene in such personal decisions". And in a November 2024 reply on PSLE pressure, MOE noted that when it removed mid-year examinations, some parents turned to tuition centres that offered mock mid-year exams. That detail tells you something important about the buyer. Parents are not only purchasing teaching. They are purchasing information about where their child stands.
Who actually makes the tuition decision
You could assume the parent decides and the child attends. In the early stages that is roughly right. The trigger is almost always parental (a disappointing test, a teacher's comment at a parent-teacher meeting, another child's results heard secondhand), and so is the shortlist. But my first reading of our transcripts, that parents hold the decision throughout, turned out to be wrong. Or rather, it was half right. Parents choose which centres get a trial. Children decide which one survives it.
The veto is rarely stated. A ten year old doesn't say "I reject this provider". She says the teacher talks too fast, or that the other kids already know each other, or nothing at all, while taking a very long time to put on her shoes on Saturday morning. Across our sessions, parents described reading these signals carefully for younger children and much less carefully for teenagers, who, somewhat against expectation, have more influence and are listened to less. By secondary school the student often picks the centre (usually where friends go) and the parent approves the bill.
Credentials matter at the shortlist, and probably less than operators hope (I say this as someone who has read a lot of tuition brochures). "Ex-MOE teacher" is the phrase parents quote most, because it answers an unspoken question about whether the tutor knows what the exam wants. Some centres push that logic further. Keynote Learning in Ang Mo Kio has tutors who sit the O-level exams every year, co-founded by a former MOE teacher, with fees starting at $240 for four weekly primary lessons. But parents rarely check the official record. MOE's register of private schools states that registration "does not in any way represent an endorsement or accreditation of the quality of the courses offered". In our interviews, parents rarely mention it. The group chat does the vetting.
The loop that runs from one results season to the next
A funnel, from awareness down to enrolment, was the obvious shape to draw. It didn't fit, because nobody leaves the funnel. A family that enrols is back at the start within a year (sometimes within a term), re-judging the choice against a new set of grades. The loop below goes by the Results-Season Loop in our project notes, and I suspect the stages compress or stretch depending on the child's level.
The Results-Season Loop
Trigger
A weak test, a teacher's remark, or another family's results heard in the class chat. Parent owns this stage.
Shortlist
Two or three names from parent chats and neighbours, filtered by distance and "ex-MOE". Websites are checked last.
Trial class
One lesson decides. The child's mood in the car home carries more weight than the lesson plan.
Verdict
Results season in November. Stay, add a second centre, or switch before the new school year.
Stage four is where operators win or lose most of their revenue (and most of their referrals). For the 2025 cohort, PSLE results were released on Nov 25, and the O-level and A-level releases follow in the new year. Parents describe late November and December as the time they "settle next year", which in practice means a narrow window in which a centre's reputation is rewritten by a few dozen result slips shared (or pointedly not shared) in the chats. The PSLE scoring change seems to have sharpened this. Under the Achievement Level system, a child's score is the sum of four subject bands from 4 to 32, and AL1 needs 90 per cent or more. One band in one subject can change the conversation about secondary schools, so parents look hard at the subject that dropped and at the tutor who taught it.
There is also a quieter anxiety driving stage one. SEAB has explained that challenging questions make up about 15 per cent of total marks in each PSLE subject, and that all of them are based on the primary school syllabus. Parents we speak with accept this in principle. They still believe that those marks are where tuition earns its fee (at least, that's how most describe it), and they judge centres by whether their child can handle the hard problems.
What parents say about choosing tuition, and what they do
| What parents say | What their choices show |
|---|---|
| "We only want tuition if she really needs it" | Tuition starts after one weak test, and rarely stops when grades recover |
| "The teacher's qualifications are most important" | The shortlist comes from a friend's recommendation, and credentials are checked afterwards, if at all |
| "I look at the centre's track record" | A handful of results heard in a group chat outweighs any published results page |
| "My child has no say, it's for his own good" | A child's reaction after the trial class decides which centre is chosen |
| "We're happy with the centre" | Satisfaction is renegotiated every November, and a single disappointing band can trigger a switch |
The gap between stated and actual behaviour here is less about honesty and more about self-image. Parents want to see themselves as measured, and they describe the decision the way a measured person would. The same pattern appears in our qualitative work on AI in Singapore education, where parents endorse "learning, not shortcuts" and then quietly reward whatever lifts the grade. It also shapes advertising. MOE said in a February 2026 reply on tuition advertising that ads "should not capitalise on anxieties to induce demand for tuition services", pointing providers to the Singapore Code of Advertising Practice. I think operators who rely on fear-based messaging are betting against both the regulator and their own buyers, who say they dislike it even when it works on them.
How we research tuition decisions without asking parents to perform
Ask parents in a room why they chose a centre and you get the measured answer. So we lean on in-depth interviews in the home, often with the child present for part of the conversation, and we ask to see the actual trail, the chat messages, the trial-class booking, the result slip that started it. Mobile ethnography helps across results season, when parents can record the week the verdict is made rather than recall it months later. For operators, it also helps to test the enquiry experience itself, since so much of it now runs over WhatsApp, and digital mystery shopping of WhatsApp sales and webchat shows what a parent actually receives when she messages at 10pm.
Two interviews, one decision: "Open your class chat and scroll back to the week you decided on your child's current centre. Show me the message that made you book the trial." Then, separately, we ask the child: "What was the first lesson like, and what did you tell your mum in the car?" The two stories rarely match, and the gap shows where the real decision was made.
We use focus groups more sparingly for this category (parents in a group tend to compete on how relaxed they are about grades), usually to test messaging once the decision journey is mapped. Deciding between the two methods is its own question, covered in our guide to choosing between in-depth interviews and focus groups. And because the parent buys on behalf of someone else, the dynamics resemble those in our caregiver research on hidden decision-makers, with the roles reversed. Here the buyer is visible and the decision-maker is ten years old.
Affordable options sit on the same map, which operators sometimes forget. Self-help groups run their own programmes. The CDAC Tuition Programme has offered affordable tuition since 1993 at 40 locations, with classes of 10 to 12 students. For lower-income families the choice set looks very different, and research that recruits only from mall-centre catchments (think of Oasis Terraces in Punggol, with at least 11 enrichment or student care centres on one floor) will miss them entirely. Our Singapore consumer research expertise starts from that sampling question.
Winning the trial class and surviving November
If you run a tuition or enrichment business, the practical reading is uncomfortable. Your marketing probably reaches parents at a stage where they have already shortlisted, and your credentials page is read after the decision rather than before it. What seems to matter is the sentence another parent types about you in a chat, the child's experience of one lesson, and what your students' result slips say in late November. HealthHub reminds parents that unrealistic expectations of academic performance can put children under unnecessary stress, and I suspect the centres that last are the ones parents feel calmer around, not only the ones with the best results. I'm still working through how those two things trade off. But nine replies in an hour, against a website nobody opened, says a lot about where the next marketing dollar should go.
What education brands ask about how parents choose tuition centres
How much do Singapore families spend on private tuition?
How do Singapore parents choose a tuition centre?
Do tuition centres in Singapore need to register with MOE?
When do Singapore parents switch tuition centres?
What research methods work best for education and tuition brands?
Understanding why parents pick your tuition centre, and why they leave after results day
The tuition decision moves through parent chats, a single trial class and an annual verdict in November, and most marketing reaches parents too late in that loop. We run in-depth interviews and focus groups with Singapore parents and children to show where your centre wins and loses families, and what would change the outcome.
Request a quote →
