The day I put every exam on a sheet is the day I stopped studying at random. Until then I did what everyone does: the next exam, then the next. At 41, with a company to run, that does not work, because you do not decide your time and the bad weeks arrive without notice.
So I did what I do with a business plan: I started from the end. I sat every exam of the three years in two, with no credits recognised for experience and no shortened track, and the part that made the difference was not studying more. It was deciding beforehand what to study and when.
How many hours does an exam really take?
The starting point is a legal convention: one credit corresponds to 25 hours of total student workload, lectures included (Italian Ministerial Decree 270/2004, art. 5). A 180-credit bachelor’s degree is therefore 4,500 theoretical hours. But hours are not spent by the month: at university time is counted in exam sessions. In my degree there were two a year, winter in December-January and summer in June-July, plus the partial tests of some exams in late October and late March: four windows a year, and I used every one of them. The standard pace is therefore 180 credits over six sessions, 30 per session. To finish in two years you need 45, that is 50% more. It is not about running faster: it is about never leaving a window empty, partials included.
The 25 hours, though, are a convention, not your hours. For every exam I started from there and corrected with two coefficients: how hard the subject was and how much of it I already knew. On corporate finance and management control I had twenty years of business behind me, and I declare it because without that head start my timings cannot be replicated. On mathematics, after twenty years without touching it, the coefficient went the other way.
The most useful rule is the recalibration after the first two exams: compare the estimated hours with the real ones and redo the plan. No first plan is right, and the difference is having the data to correct it.
How do you plan backwards?
Seven steps, in this order.
- Set the target graduation date. Not “as soon as possible”: a date.
- List every exam with its credits, perceived difficulty and your familiarity with the subject.
- Estimate the hours per exam starting from 25 per credit and correcting with the two coefficients.
- Map the available exam sessions, which are the truly scarce resource: time can be found, exam dates cannot.
- Assign every exam to a session starting from the last one and working backwards, balancing heavy and light exams in the same session.
- After the first two exams recalibrate your personal coefficient and redo the plan.
- Track your pace in credits per session: it is the only number that tells you whether you are on course before it is too late.
Then there are the practical rules. A buffer exam in every session, easy or close to what you know, because a session closed at zero is what kills the plan. Never two calculation-heavy exams in the same week: you only have one head. Short, frequent study blocks, because for someone who works the whole Sunday does not exist, and an hour a day yields more anyway: spaced repetition is among the few high-utility study techniques in the 2013 review by Dunlosky and colleagues. And above all: check the prerequisites and the rules of your own programme before planning, not after. Finding them out once the plan is made costs a session.
Why didn’t I attend lectures?
Because time was the scarce resource, and in a lecture hall time is not yours. I tried one lecture for every subject: a lecture is designed for hundreds of people, it moves at the pace of the class, and for the way I study one hour in the hall was worth ten minutes at home with the right material, the right procedures and an AI that quizzes me. It is not a criticism of the professors, it is the arithmetic of my calendar.
With one exception. There are professors who have the gift of explaining a subject and making you like it, and in one hour they give you a context that no handout and no AI will give you. Those lectures I attended on purpose, few and chosen, and it was worth it every time.
What happens when the plan breaks?
It breaks, that is certain. A week of work that explodes, an exam gone wrong, a subject that turns out twice as long as expected. The backward plan is not there to avoid surprises: it is there to tell you at once what they cost. If an exam slips, the sheet tells you which session to move it to and which other exam shifts to make room. Without the sheet you find out in June, looking at the credits you are missing.
The extreme case I did in June 2025, ten exams in ten days, and I tell it separately: it is not an example to follow, it is proof that the plan holds even when you load it beyond reason.
My first mistake, though, was not about time. It was about presumption. I underestimated the marketing exam because in twenty years of companies I had done a great deal of marketing and paid for even more, and I already felt like a king. I went without studying and passed with 18 out of 30, which for someone who lives on marketing is a mark pride cannot tolerate. I retook it, and the second time I got 30. The real lesson came when I actually studied it: inside a theory I thought far from practice there were foundations that would have spared me mistakes, even trivial ones, paid dearly for in the field. Since that day, when I estimate how well I know a subject, I take one step down: the familiarity coefficient is the easiest one to get wrong, exactly where you feel strongest.
Does it work for everyone?
No, and I would rather say it myself before the reader finds out. My pace is one case, not an average, and it depended on three things: twenty years of experience in a good part of the subjects, a job that lets me decide my hours, and a university that allowed me to sit exams early. Without one of these three conditions: a higher coefficient and a later date, and no drama.
What can be copied is the method: start from the end, estimate the hours, recalibrate with real data, protect every session with a buffer exam. It works for a degree and it works for any project with a deadline and scarce resources, which is the definition of almost everything I have done for twenty years.
I built companies without ever having studied planning, and I planned them badly more than once. At 41, with a spreadsheet and a graduation date, I understood that the method was the same one I should have used at 25 with business plans. Better late. And if you are reading this with an exam two sessions overdue, the right moment to put everything on a sheet is this afternoon. The study loop and the prompts I then use to prepare each exam are written up in twelve pages, the Quaderno 3in2, which I send for free to anyone who asks (in Italian): the plan decides when, those pages say how.
Sources: Italian Ministerial Decree of 22 October 2004, no. 270, art. 5 “University credits”, Official Gazette no. 266 of 12 November 2004 (in Italian); Dunlosky, Rawson, Marsh, Nathan, Willingham, “Improving Students’ Learning With Effective Learning Techniques”, Psychological Science in the Public Interest, 2013. The rules of your own degree programme count more than any plan: check them in the regulations, not in this article.
