Twenty Years of Blue Tutors: What's Changed and What Hasn't
I started Blue Tutors in 2004, in a spare room, with a phone, multiple spreadsheets and a conviction that good tutoring made a real difference. Twenty two years later I still believe that. Almost everything else about the job has changed.
Back then, tutoring was a word-of-mouth trade. A parent would ask another parent at the school gate, or a teacher would quietly recommend someone on the side. It was definitely a quieter market, compared to now. National figures from the mid-2000s put the proportion of secondary-age pupils receiving private tuition at around 18%. Today that figure sits closer to 29%, and in London it is closer to 45%. Tutoring has gone from a slightly niche, slightly awkward thing you didn't mention at dinner parties to something most families have either used or seriously considered.
I had no idea how private tuition would develop. I hoped we could carve out a niche for Oxbridge tutors, but I had no idea I was getting in near the start of a trend that would eventually see tutoring discussed in government spending reviews.
Before I started Blue Tutors I tutored independently in London. One of my earliest clients was a Year 13 girl struggling with maths. I remember meeting the family and realising that simply saying I graduated from Cambridge was enough for them, half expecting them to ask for references I didn't yet have. There was nothing to check me against except a conversation at the door. Families now expect, reasonably, a level of professionalism and vetting that simply didn't exist as an industry standard back then. That change alone, the shift from an informal arrangement to something closer to a proper service, has probably done as much for tutoring's reputation as any technology has.
The exams themselves have changed too, more than people sometimes remember. GCSEs moved from the old A* to G grading to the 9 to 1 scale in 2017, alongside a switch from modular exams, sat in bite-sized chunks across two years, to linear ones, sat almost entirely at the end. That was a genuinely significant shift for tutoring. Modular courses meant a student could recover from a bad January module with a strong result in June. Linear courses put far more weight on a single set of exams in May and June, which meant the cost of a wobble in Year 11 or 13 became much higher, and the value of steady, well-timed support in the run-up to exams rose with it.
The first big shift was Skype, and then Zoom. For years, tutoring meant a tutor travelling to a family's kitchen table, book bag in hand, or a student being driven across town after school. That was the whole model, and it worked, but it was also limited by geography. If the best physics tutor for your child's specific exam board lived an hour away, you either accepted the journey or found someone less well suited who happened to live closer.
Online lessons broke that constraint apart. Suddenly a student in Cornwall could work with a tutor who had spent a decade teaching the exact syllabus they were sitting, without either of them leaving the house. I was sceptical at first, if I'm honest. I worried something got lost without a shared table and a shared piece of paper. Some of that scepticism was fair; a screen is not quite the same as sitting beside someone while they work through a problem. But I underestimated how quickly both tutors and students would adapt, and how much flexibility mattered to busy families and busy tutors. Online tutoring is not a compromise any more. For a lot of our families, it is simply the preference.
Then came the pandemic, which did more to reshape tutoring in eighteen months than the previous fifteen years combined. Schools closed, learning fragmented, and parents who had never considered a tutor before suddenly needed one, urgently, for a child who had missed months of proper teaching. The government's own National Tutoring Programme delivered nearly 4.7 million tutoring courses across the 2020/21 to 2022/23 academic years. Before Covid, tutoring was the spending priority for about 10% of secondary school leaders managing Pupil Premium funding. By 2022, that had risen to 34%. Private tutoring uptake, which had been broadly stable for years, jumped to its highest level since 2005.
What struck me most wasn't the volume, though the volume was real. It was the change in attitude. Tutoring stopped being seen as something only struggling students needed, or something only wealthy families could justify. It became, for a huge number of ordinary families, a sensible response to a genuinely difficult situation. That shift in perception has outlasted the pandemic itself. Parents who booked a tutor in 2021 out of necessity are often still booking one now, because they saw what a good tutor actually does: not just plugging a gap, but building a student's confidence in a subject they'd started to dread.
The most recent change, and the one I find genuinely harder to have a settled view on, is artificial intelligence. Some families we speak to now ask about it in some form. Should my daughter be using ChatGPT for her homework? Is an AI tutor a sensible alternative to a human one? Will this technology make tutors like ours redundant within a few years?
I don't think the answer is straightforward, and I'm wary of anyone who tells you it is. The government clearly thinks AI has a role to play; its AI Tutoring Pioneer programme, launched this year, is trialling AI tools designed to support several hundred thousand disadvantaged pupils with personalised, one-to-one style learning. Teacher use of generative AI has nearly doubled in the space of a year, mostly for planning and admin rather than direct teaching. That tells you something about how fast this is moving inside schools, whatever families decide to do at home.
But I also take seriously the concerns being raised by the people who actually stand in front of classrooms. A National Education Union survey this year found that two thirds of secondary teachers believed their pupils' critical thinking had declined because of AI use. That doesn't surprise me. There is a real difference between a tool that helps a student check their own reasoning and a tool that does the reasoning for them. And this isn't always obvious to a fifteen year old under deadline pressure at 11pm the night before coursework is due.
My own view, for what it's worth, is that AI will end up doing to tutoring roughly what online lessons did: it will remove some of the friction, and it will not remove the need for a person who genuinely understands both the subject and the student in front of them. A good tutor notices when a student's confusion is really about confidence and understanding, not knowledge. They adjust their questions based on a raised eyebrow or a long pause, not just a wrong answer. I haven't seen anything yet that replicates that, and I suspect the families who use us feel the same way, or they wouldn't keep coming back.
What hasn't changed, in twenty-two years, is the thing that made me start Blue Tutors in the first place. A student who is stuck on a subject, quietly convinced they're just not a maths person or a languages person, and a tutor who is patient and skilled enough to show them otherwise. That relationship is still, in my view, the whole point. Everything else, Skype, lockdown, AI tutoring pilots, is just the changing shape of the room it happens in.
I don't know exactly what tutoring will look like in another twenty years. I didn't predict the last twenty with any accuracy, so I won't pretend to now. But I'll take a reasonably confident guess that whatever tools arrive, the families we work with will still want the same thing they wanted in 2004: someone who genuinely cares whether their child understands the material, and who is good enough to make sure they do.