What to ask a lead generation agency before you sign
Most questions buyers ask an agency are easy to answer well and predict nothing. These are the ones that actually separate them.
By The Coaching World
Published October 4, 2026 · 6 min read
Photo by S O C I A L . C U T on Unsplash
The questions that predict nothing
Most of what buyers ask an agency is easy to answer impressively and tells you nothing. How long have you been doing this. Who have you worked with. What results do you get. Every agency has polished answers, and the polish is unrelated to whether they will do good work for you.
The questions that separate agencies are the ones with no flattering answer available.
This article is for anyone about to sign with an outreach agency. It covers the questions worth asking, what a good and bad answer sounds like for each, the contract terms that matter more than the pitch, and the thing to check after you sign.
Ask who does the work
Not who is in the meeting. Who writes the copy, builds the list and reads the replies.
Agencies frequently sell with senior people and deliver with junior ones. That is not automatically a problem, since a well-run junior team with good process beats an overstretched senior one. It is a problem when it is concealed, because you priced the engagement on the wrong assumption.
A good answer names roles and is specific about what each person does. A weak answer talks about the team and moves on.
Follow up with how many accounts that person handles. The number tells you how much attention yours will get, and it is rarely volunteered.
Ask what they would need from you
An agency that needs nothing from you is planning to work from generic assumptions.
Good outreach requires things only you have. Which clients were most profitable and why. What objections come up repeatedly. Which prospects turned out to be a poor fit after signing. Language your buyers actually use.
An agency that asks for none of this will build a list from job titles and write copy from your website. That produces a campaign indistinguishable from the one they ran for someone else.
A strong answer includes a request for your time in the first weeks, and is specific about what for. If they say they can run it entirely without you, that is a warning rather than a convenience.
Ask what usually goes wrong
This is the single most revealing question available.
Every experienced agency has seen campaigns fail and knows the patterns. An answer that names specific failure modes, explains how they are detected, and describes what happens next is evidence of real experience.
An answer that deflects, blames clients generally, or claims nothing much goes wrong means either inexperience or a sales reflex. Both are reasons to keep looking.
Pay attention to whether the failures they name include ones that are their fault. Agencies that only describe client-side failures are telling you where the blame will land when the campaign struggles.
Ask what they will change and when
Outreach is iteration. The plan for iteration is the actual service.
Ask what gets tested first, and why that before anything else. A credible answer has an order and a reason. Targeting before copy is a common and defensible one. Having no order means variables will get changed together, which makes every result uninterpretable.
Ask how long a test runs before they act on it. Short answers suggest they will chase noise.
And ask what they do if month two is poor. The answer reveals whether they have a process or a hope.
The contract terms that matter
Three things matter more than the headline price.
What happens to the assets. Lists, copy, sequences, accounts. If you leave, do you take anything with you. Many agreements leave the client with nothing, which means switching costs restart the whole engagement.
What a guarantee actually commits to. Meeting guarantees are common and vary wildly. Find out what counts as a meeting, what happens to no-shows, and whether the remedy is a refund or more work. A guarantee of continued effort is not a guarantee.
Notice period against the sales cycle. A thirty day notice period on a service whose results take ninety days to appear puts you in a bad position in either direction.
After you sign
Ask for the list before the first send.
Not a summary, the actual list. Read fifty entries and check whether you would sell to those people. This is the only moment where the single most important input is easy to inspect and cheap to fix.
Most clients never ask, and most campaigns that fail on targeting could have been caught right there.
Ask what they will not do
Good agencies decline things, and knowing what they decline tells you what they stand for.
Some will not run accounts where the offer has never been sold to anyone. Some will not take clients below a certain price point, because the cold cycle makes the economics impossible. Some will not work in categories where the targeting data is unreliable.
An agency that says yes to everything has either no point of view or no capacity constraint. Neither is encouraging.
Ask directly: who is this not for. A specific answer, with a reason attached, is one of the strongest signals available in a sales conversation. A vague answer about being able to help most businesses usually means the process is generic enough to apply to anyone, which is another way of saying it is not built for you.
The reference call to ask for
Ask to speak to a client who is six months in, rather than one who just started.
Early clients are enthusiastic because nothing has gone wrong yet. Six months is long enough for the first difficult month to have happened, and the useful question is what the agency did about it.
If they will only offer a recent client, that is informative too.
What a trial month really tests
Short trials are common and they test less than both sides pretend.
A month is enough to see whether messages are being delivered, whether the list looks sensible, and whether the agency communicates well. It is not enough to see whether the campaign produces clients, because the cold cycle has barely started.
So treat a trial as an assessment of process rather than outcome. Did they ask good questions. Did the list match what you agreed. Did they change anything after the first week of data, and could they explain why.
Judging a trial on signed revenue will make you reject agencies doing good work and keep ones who got lucky with an early meeting.
Whatever you agree, write down what you expect to see at thirty, sixty and ninety days before the first message goes out. Agreed expectations are much easier to discuss than remembered ones.
Our articles are researched and reviewed by our editors. Read our editorial standards.