Most articles on this are a list of questions with model answers. The problem is that the interviewer has seen the same list, and a memorised answer is audible from across the table.
What helps more is understanding that the questions fall into a few types, and each type measures something different. Get that, and you can answer a question you have never seen before, which is precisely what is being tested.
Type one: did you actually do the work
These are the "tell me about a campaign you worked on" and "what is the worst result you have had" questions. The story is not the point. The interviewer is working out whether you did the work or sat next to someone who did.
What gives people away is talking only in outcomes with no numbers and no decisions. Someone who actually ran it remembers boring specifics: what the budget was, what they changed, why, and what happened next.
If you have no work experience, the answer is not to speak in generalities. It is to have made something yourself that you can talk about, however small, which is the subject of building a marketing portfolio with no experience yet.
Type two: do you understand the mechanism
These are the technical questions, and they trip people up because it is easy to memorise a definition without understanding why the platform behaves that way.
A recurring example: why does a platform need a certain number of events before you can judge a campaign? Meta explains in its own note on the learning phase that an ad set stabilises after roughly 50 optimisation events within a seven day period at most, because during that time the delivery system is "exploring the best way to deliver your ad set".
The differentiator is not reciting the number fifty. It is being able to add that repeated edits restart learning, so someone who adjusts the campaign every day is not optimising it, they are preventing it from ever stabilising. That shows you understand the system rather than having memorised a figure.
Type three: how do you think when the data is incomplete
The "performance dropped this month, what do you do" question. There is no correct answer, and that is deliberate. The interviewer is watching the order in which you think.
A weak answer jumps straight to a fix: "I would change the creative." A strong answer asks first. How far did it drop, since when, across every campaign or one, and did anything change at the same time?
Asking before proposing is not dodging the question. It is exactly what separates someone who will work correctly from someone who will work quickly.
The certification question, answered honestly
Certifications come up often, sometimes as a test in themselves. It helps to know what one actually claims.
Google states on its page about Google Ads certifications that a certification "will remain valid until the product area certification expires (one year)", and that you have to retake and pass the assessment to renew it.
That is useful in an interview because it shows the limit of the credential. It is evidence that you studied the platform at a point in time, not evidence that you have run accounts. Saying that yourself, before it is said to you, reads as maturity rather than as undercutting your own CV. I went into it further in whether marketing certificates are worth it.
The questions you should be asking
An interview runs in both directions, and what you ask says as much as what you answer.
Who sets the budget, and how does a decision to increase it get made?
What number will my work be judged on after three months?
Who reviews my work, and what is their background in marketing?
What has been tried before and did not work, and why?
That last one is unusually revealing. A company that cannot answer it is usually not measuring anything, and the work inside it will be driven by opinions.
One last thing
An interview is not a knowledge test. It is an attempt to picture what working with you would be like. That is why "I do not know, but here is how I would find out" is a far stronger answer than a memorised one delivered slightly wrong.
And if you are still deciding where you are heading in the field, the choice between specialising and staying broad changes the kind of questions you will face, which I covered in the digital marketing career path.