A diploma ends with a certificate, and the certificate does not tell you whether it worked. The useful test is not what was covered in class. It is what you can now do on your own, with nobody to ask, in front of a real account.
So this is a checklist you run on yourself, three or four weeks after the last session. Each item is phrased as a task rather than a topic, because "we covered analytics" and "I can set up a key event and explain why the number differs from Google Ads" are not the same claim. Where the answer is no, that is not a verdict on the course. It is the list of what to practise before an interview.
How to use this honestly
Do it without opening your notes. The point of the exercise is to find the gap between recognising something and being able to produce it, and notes hide exactly that gap. Recognition is what a multiple choice exam measures and it is a weak proxy for competence.
Score each item as one of three: I can do this unaided, I could do it with documentation open, or I would not know where to start. The middle category is fine for most things and not fine for a few, and the list below says which.
Measurement, which is where most graduates are weakest
This section first because it is the one that decides whether you get past probation. Everything else can be learned on the job more easily than this can.
- Mark an event as a key event and say what that does. Google's documentation defines a key event as an event that measures an action that is particularly important to the success of your business, and the mechanism is that you create or identify the event first and then mark it. If you cannot describe that two step sequence, you do not yet have it.
- Explain why Google Ads and Analytics disagree. This is the single most common question a junior gets in their first month and the most common one they fumble. The documentation on key events is explicit that you can create a Google Ads conversion from a key event, and that Google Ads conversions do not appear in standard Analytics reports. You should be able to say that out loud before anyone asks.
- Read a report and name what it cannot see. Specifically: when a table exceeds its row limit, Analytics surfaces only the most common dimension values and condenses the rest under an (other) row, and the data quality icon can carry a warning even when no (other) row is visible in the report. A graduate who does not know this will eventually present a conclusion drawn from a truncated table.
- Set up a conversion from scratch on a site you did not build. Unaided. This is the one item on the list where "with documentation open" is not good enough, because it is the task most likely to be handed to you in week one.
If this section is thin, the fix is not another course. It is one real property, your own or a friend's business, and a fortnight of configuring things and checking whether the numbers arrive.
Paid media
- Build a campaign end to end on one platform, including audience, placements, budget and a conversion event, and explain each choice.
- Read a performance table and say which number is the symptom and which is the cause. Cost per result going up is a symptom. Frequency, creative fatigue and audience size are candidate causes.
- Say what you would turn off first when a budget is cut by half, and why that and not something else.
- State plainly what the platform cannot attribute. If you have never said the words "we cannot tell from this data", you have not worked with real data yet.
One platform properly beats four shallowly, and that is not a consolation prize. It is what a hiring manager is looking for, and it is the premise of the module order in the diploma's own breakdown of what gets taught.
Content and writing
- Write a single clear sentence describing an offer, with no adjectives doing the work.
- Produce a landing page outline for a product you have not sold, and justify the order of the sections.
- Take a brief that is too vague and turn it into three specific questions for the client.
- Edit your own work down by a third without losing the argument.
The judgement items, which no course can certify
These are the ones that separate two graduates of the same programme a year later, and they are harder to self-assess, so phrase them as things that have actually happened rather than capacities you believe you have.
- Have you said "I do not know" about a number, in front of someone senior, and then gone and found out? The alternative habit, inventing a plausible explanation for a figure you do not understand, is the most expensive one in this field.
- Have you let a test run long enough to be boring? Turning a campaign off after two days feels like control and destroys the measurement.
- Can you explain what you did to someone outside marketing, without jargon, in a minute?
- Have you changed your mind because of data that disagreed with you? If not, you have not yet been in a position where the data could.
Diagnosis is the skill underneath several of these, and it is worth seeing how a practitioner sequences it. Mostafa Faried has written a working order for the specific case of a conversion rate that drops suddenly, a diagnosis order to run before touching the campaign. It is in Arabic, and it is a good example of what the finished version of this skill looks like: check the measurement before you believe the drop.
The portfolio test
One question settles whether the diploma converted into something employable: can you show three pieces of work with a result and an explanation of what you would do differently?
Not three certificates. Three artefacts. A campaign with its numbers, a page you wrote and the reason for its structure, an analysis that reached a conclusion. The final project from the course counts as one of the three, and the ideas in the final project list are chosen to be showable rather than merely gradable. The other two should come from outside the classroom, even unpaid, even small.
If you have none, that is the first thing to fix, ahead of any further study. It is also the honest answer to why two people from the same cohort get different outcomes, which is covered in more detail in what the roles after a diploma actually ask for.
What a low score does and does not mean
It does not mean the diploma failed. A taught course has a fixed number of hours and the items above include several that only accumulate with repetition, so nobody finishes any programme scoring full marks on all of them. Expecting otherwise is how people end up buying a second course to fix a problem that needed practice instead.
It does mean you now have a specific list rather than a vague feeling, and specific lists are tractable. Two items a week, on a real account, is a month of work and it closes most of this.
And if the measurement section came out empty while everything else was fine, that is a known shape and worth naming: it is the most technical part of the syllabus, it is the part most often taught last when attention is lowest, and it is the part employers test first. Start there.