Digital marketing isn't learned by watching slides

By Miguel Maio · Published 2 October 2026

Watching vs doing: what sticks

1. 1Lecture: 1.5x more failures (Freeman)
2. 2Lab: +0.47 SD on exams
3. 3The feeling of learning misleads (Harvard)
4. 4Deliberate practice: do, with feedback

You learn to do by doing.

I've sat in plenty of rooms where the "hands-on course" was an instructor clicking through slides for three hours, with screenshots of a Google Ads account nobody in the group ever touched. Everyone left with the feeling they'd learned something. Almost nobody left able to build a campaign on their own come Monday.

And it's not rhetorical exaggeration. In my own teaching I've had reports of students subjected to nearly 170 slides in little over two hours. More than one a minute, with no pause to breathe. Their feedback was awful, and rightly so: at that pace you don't teach, you dump. You've barely read one slide and the next is up. It's one of the worst pedagogical experiences I know, and it happens precisely in courses that advertise themselves as hands-on.

That gap, between feeling you've learned and being able to do, is the subject of this article. None of this is opinion. In education, few things are as well evidenced, and this is one of them.

Watching isn't knowing how to do

The landmark meta-analysis on this is Freeman and colleagues, published in PNAS in 2014. They pooled 225 studies comparing lectures with active learning: students doing, not watching. Two findings, both solid: exam performance rose 0.47 standard deviations with active learning (about half a grade), and in pure lectures students were 1.5 times more likely to fail (33.8% failure against 21.8%). The authors were blunt: if this were a clinical trial, the lecture arm would have been halted for being inferior.

This was measured in science, engineering and maths, not marketing, and it's fair to flag that. But the mechanism isn't specific to physics: you learn to do something by doing it, with correction along the way. It's the same reason nobody learns to drive by watching driving videos. And in marketing's own literature, the work on work-integrated learning (Lu, Scholz and Nguyen, 2018) points the same way: students gain competence and employability when the course forces them to work on real problems, not abstract simulations.

Before anyone throws the "learning pyramid" at me (the one claiming we retain 10% of what we read and 90% of what we do), those numbers are fake. They have no empirical basis, were mistakenly attributed to the NTL and to Edgar Dale, and whoever went looking for the source never found it. I use real science, not the motivational poster. The case for practice doesn't need invented numbers; the real ones are more than enough.

Why the feeling of having learned misleads

There's a Harvard study, Deslauriers and colleagues (PNAS, 2019), that should be taped to the wall of anyone who buys training. They exposed the same students to both formats (lecture and active) with the same content. Those who had the active class demonstrably learned more on the tests. But they felt they'd learned less. The authors' line is sharp: actual learning and the feeling of learning were strongly anticorrelated.

The explanation is uncomfortable and it matters to whoever picks a course. A good speaker makes you feel you've understood everything. It's all tidy, fluent, effortless. That comfort is the feeling of learning, and it's deceptive. Real learning costs: it means stumbling, getting stuck, correcting. That friction gets read as "this course is hard" or "I'm not good at this", when it's precisely the sign that learning is happening.

The practical consequence is awkward: course reviews taken right at the end mostly measure the instructor's warmth and eloquence, not what stuck. The course that leaves you smiling may be the one that taught you least. That's why I distrust training sold on the "experience" rather than on what the student walks out able to do.

What a real lab looks like

"Hands-on" has become a worn-out word. Almost every course calls itself hands-on. When I ask what that means, the answer is usually "we show real examples". Showing isn't practicing. In a real lab, the one with hands on the keyboard is the student, and the mistake happens right there, in front of someone who can correct it. A few concrete examples:

1. Building a Google Ads account from scratch (campaign structure, ad groups, match types, extensions) and defending each decision. Not watching the instructor's account; building your own and hearing where the structure will burn budget.
2. Auditing a real site with Screaming Frog: running the crawl, reading the response codes, finding the wrong canonical, the hreflang serving the wrong country, the redirect chain. And then writing the audit, because spotting the problem and communicating it are different skills.
3. Opening a real business's GA4 and answering a business question with the data - not "what is a session", but "where did the conversion that matters come from and what held it back". Building the Looker Studio report someone will actually use.
4. Building a small tool - a rank tracker, an anti-waste Google Ads script, a sheet that joins data from two sources. Whoever has built the machinery understands it differently.

In all of these the pattern is the same: a real task, the student doing it, individual feedback while they do. This is what Ericsson (1993) called deliberate practice: not repetition, but practicing at the edge of your competence, with immediate correction. Without the feedback, it's just busywork.

Theory doesn't die - it moves

This isn't an anti-theory manifesto, that would be stupid. Anyone building campaigns without understanding the auction, Quality Score or attribution is pressing a button blind. The question isn't whether to teach theory, but when. In the slide model, theory comes all up front, just-in-case, and the student has to store it for a practice that may never arrive in class.

In the lab model, theory comes in just-in-time: the student hits a concrete wall and the explanation arrives the moment they need it and can grasp it. I explain Google's auction when the student sees their CPC spike and asks why, not on slide 14 of a morning. Theory sticks better glued to a problem the student has just felt first-hand. It's this articulation that separates training from informational entertainment: practice that raises the question, theory that answers it, practice again to lock it in.

Why few do it

If it's so much better, why do most courses stay projection? Because the lab is far more demanding for whoever teaches. Projecting slides scales effortlessly: the same deck serves 10 or 100 people, and the instructor controls the pace from start to finish. A lab doesn't scale that way. It means designing exercises that fail in useful ways, following each student, giving individual feedback, and withstanding the discomfort of a room that's stuck and frustrated before it gets it.

And there's the reputational cost the Harvard study exposes: the instructor who makes students sweat takes worse reviews, in the short term, than the one who entertains them. You have to know the learning is happening even when the room doesn't feel it yet. Most take the comfortable path for both sides. I don't.

How to tell if a course is really hands-on

Before you sign up, or buy training for your team, ask these questions and demand concrete answers. It's the same logic I use when I explain how to choose a course, tightened to the point that misleads most:

1. How much time will I have my hands on the keyboard, doing, rather than watching someone do? If the answer is vague, the practice is decorative.
2. What will I walk out able to do on my own, concretely? A serious course describes the outcome in action verbs, not topics covered.
3. Do I work with real data and accounts, or brochure examples? The real world is messy; that's where you need to know your way around.
4. Is there individual feedback on what I produce, or only generic correction for the room? Without feedback on your work there's no deliberate practice, just busywork.
5. Has whoever teaches actually done this, outside the classroom? You teach best what you've practiced with consequences.

What you lose by ignoring this

Digital marketing is a craft, and crafts are learned by practicing them, with someone beside you to correct. Whoever leaves a course with only the feeling of knowing will find out the truth on day one at work, facing an account to build and nobody to tell them where to start. The bill for the course that was "a nice presentation" comes due right there, with interest. If you want to see how I apply this, the courses I run are designed this way: fewer slides, more of your own hands on the keyboard.

Sources

• Freeman et al. (2014), PNAS - Active learning increases student performance in science, engineering, and mathematics
• Deslauriers et al. (2019), PNAS - Measuring actual learning versus feeling of learning
• Ericsson, Krampe & Tesch-Römer (1993), Psychological Review - The role of deliberate practice in the acquisition of expert performance
• Lu, Scholz & Nguyen (2018), Australasian Marketing Journal - Work-Integrated Learning in International Marketing

Frequently asked questions