The short version
Ask how the fee moves when your ad spend moves. Ask whose name is on the ad account. Ask who else in your category they work with. Ask who is doing the work once the pitch team leaves the room. Four questions, roughly ten minutes, and they tell you more than a case study deck will.
What makes them useful is that each one is awkward to answer straight. The published lists of questions are written by the side being questioned, which is why they tend to cover process, reporting and turnaround times. Those are worth knowing. They are also questions any competent firm has a polished answer ready for.
Why fifteen questions is the wrong number
The standard guide runs to fifteen or twenty questions across four neat categories. It is a genuinely useful document and almost nobody uses it as written, because a first meeting is a conversation and not an audit. You get through the first three, the discussion goes somewhere interesting, and the list stays in the notes app.
So the question is not which fifteen to ask. It is which four survive contact with a real meeting. The four below were chosen on one test: does an honest answer cost the person answering something. If it does not, the answer carries no information.
A question that is comfortable to answer has already been answered a hundred times.
How are you paid when my ad spend goes up?
Performance marketing in India is commonly priced one of three ways in 2026: a flat monthly retainer, typically in the range of ₹1.5 lakh to ₹6 lakh; a percentage of ad spend, usually ten to twenty per cent; or a hybrid that takes whichever is higher. On a ₹5 lakh monthly budget the percentage model puts roughly ₹50,000 to ₹1 lakh a month on the table.
The percentage model has a structural property worth naming plainly: the fee rises with the budget, and it does not rise with efficiency. A firm that halves your cost per acquisition and therefore halves the budget you need has just cut its own revenue. This is not an accusation about anyone in particular. It is arithmetic, and it is the reason the market has been drifting toward flat and hybrid structures.
What a good answer sounds like
A good answer names the model without being pushed, explains the incentive it creates, and says what stops that incentive running away. Something like: we charge a flat fee, so recommending a bigger budget earns us nothing. Or: we take a percentage, here is the cap, and here is the efficiency target we are held to alongside it.
The answer to be careful with is the one that treats the question as rude. Pricing structure is not a trade secret, and a firm that has thought about its own incentives will be pleased you asked.
Whose name is on the ad account?
This is the question that costs the most to get wrong, and almost nobody asks it in month one. When a firm sets up your campaigns inside its own manager account, that account belongs to the firm. Your campaign history, conversion data, audience lists, quality scores and negative keyword lists live inside an asset you do not own.
The consequence only shows up on the way out. One documented case had a company spend fourteen months trying to separate from its previous firm, with no admin access and no way to retrieve two and a half years of negative keyword lists it had paid to build. Nothing improper had happened. The account had simply been created in the wrong place on day one.
The correct structure is unglamorous and takes about fifteen minutes to set up. You own the ad account, the analytics property and the tag manager container. You grant manager-level access. If the relationship ends, access is revoked and every piece of accumulated learning stays with you.
- Ask to be made owner of the Google Ads account, not a user on it.
- Ask the same about Analytics, Tag Manager, Meta Business Manager and the domain registrar.
- Ask what an offboarding looks like in practice, and how long it takes.
- Get the answer in the contract rather than in an email thread.
Who else in my category do you work with?
Category experience is sold as a strength, and it genuinely is one. The same experience is also a constraint, and the two are rarely discussed together. A firm that has run campaigns for four brands in your category knows the auction, the creative that works and the seasonality. It is also bidding against you in some of those auctions, on behalf of someone else.
There is no correct answer here, which is exactly why the question is useful. What you are listening for is whether they have thought about it at all. A firm with a real position will tell you where its line sits: no direct competitors in the same city, or full disclosure before onboarding, or separate teams with no shared data. A firm that has never considered the question will tell you it is not an issue.
Who is doing the work once the pitch is over?
The people in a pitch are usually the most senior people available. The people on your account from week three are usually not, and there is nothing wrong with that as long as it is stated rather than discovered. What matters is how many other brands those people are carrying, because attention divides in a way that skill does not.
Ask for names and current load. Ask how many accounts your day-to-day contact handles today, not how many they are allowed to handle. Ask what happens when someone leaves, since in this industry someone always does. The specificity of the answer is the answer.
How we would answer these
It would be poor form to publish four uncomfortable questions and not sit on the other side of them, so here is Slate on all four.
Fees are flat and agreed for the term, so a larger budget earns us nothing extra and a more efficient one costs us nothing. Every account, property and container is created in the client's name on day one, and we hold manager access that can be revoked in a minute. On category overlap we put the current client list in front of you before anything is signed, and it is short enough to read in under a minute. The people you meet are the people who do the work, because there are two of us and a short list, which is the whole design.
That last part is also the honest limitation. A model built on a short list cannot absorb a brand that needs twenty deliverables a week across six channels. If that is the requirement, the right answer is a firm with the bench for it, and the four questions above are how you find a good one.
What a good conversation feels like
The pattern across all four questions is the same. You are not testing for the right answer. You are testing whether the person opposite has already thought about the awkward parts of their own model, because someone who has will handle the awkward parts of your business the same way.
A meeting that goes well tends to leave you with a shorter list of certainties and a longer list of good questions about your own business. A meeting that goes badly leaves you impressed and unable to explain, an hour later, what was actually agreed.