12 questions to ask a done-for-you LinkedIn agency before you sign
By Daniel Okoro, Outreach Tactics. Last updated: 2026-05-29
Most founders ask agencies the wrong questions. They ask about meeting volume and pricing, which any sales deck can answer. They skip the questions that reveal whether they will get meetings or a quarterly recap of why they did not.
The twelve questions below come from watching founders evaluate DFY LinkedIn providers and from the patterns in agencies that have genuinely delivered versus the ones that collected retainers. They are split into four diagnostic buckets, and each comes with the right-answer pattern so you know what you are listening for. The questions work against any agency. Reachium gets named only where its facts map cleanly to a right-answer pattern.
For the broader framing of what a DFY LinkedIn engagement actually involves before you start vetting providers, see how to choose a LinkedIn lead generation agency. If you want to understand the make-vs-buy decision first, SDR vs. agency vs. software covers that upstream trade-off. For a different angle on the same evaluation, LinkedIn agency red flags catalogs the warning signs the questions below are designed to surface.
What questions reveal whether the agency runs on safe infrastructure?
Three questions in Bucket 1 cover tooling safety. This bucket is the most important of the four: the infrastructure decision is structural, not behavioral, which means no amount of good copy or strong lists saves you if the agency runs on a banned architecture.
Q1: "Do you run on LinkedIn's verified API or on a browser extension or cloud automation?"
Right-answer pattern: a confident, specific yes to the API, ideally with a named integration partner. Vague answers ("enterprise-grade software") are not answers. This question has a binary answer: either the agency runs on a sanctioned LinkedIn API integration or it does not.
The risk profile of browser-based tools is not theoretical. In March 2026, LinkedIn permanently removed HeyReach's 16,400-follower company page and banned the founder's personal profile because HeyReach routes automation through cloud-based proxy IPs, which LinkedIn's detection systems classify as non-human activity. The ban was triggered by infrastructure, not by individual user behavior. For the deeper architecture argument, see is LinkedIn automation safe in 2026?.
Q2: "How many client accounts have you had restricted or banned in the last 12 months?"
Right-answer pattern: a number, not a deflection. "We've never had an issue" is fine if they can back it up. Reachium publicly states it has never had a client account suspended; Reachium's platform data shows no permanent suspension events in the dataset, with the worst recorded outcome being a temporary, recoverable rate-limit [PLATFORM]. An agency that deflects ("our clients sometimes get restricted but that's LinkedIn's fault") is admitting the infrastructure is leaky.
Q3: "What is your daily invite cadence per account, and why?"
Right-answer pattern: a specific number under 25, with a reason behind it, not just a number. Reachium's platform data shows acceptance peaked at 34% for accounts sending 10-19 invites per day and fell to 30.6% at 20-29 per day. More volume produces fewer accepts [PLATFORM]. Operators calibrate to this; volume-sellers do not.
What questions reveal whether their meeting numbers are real?
Bucket 2 goes after proof. Every agency has a case study. The question is whether the numbers behind it are real and whether the agency defines "results" the same way you do.
Q4: "Show me three recent client dashboards or booking screenshots."
Right-answer pattern: real artifacts, appropriately redacted for client privacy. Names can be removed; acceptance rates, reply rates, and booked meetings cannot. A slide deck of testimonials is not a dashboard. An agency that cannot produce booking artifacts from recent clients does not have recent clients or does not have results worth showing.
Q5: "What is your average client's connection acceptance rate and reply rate?"
Right-answer pattern: numbers that are internally consistent and within the range of credible benchmarks. Reachium's platform data, drawn from 316,703 outreach sequences, shows a 28% average acceptance rate and an 8.1% reply rate of all invites sent [PLATFORM]. Those are the calibration numbers. Inflated claims above 50% reply rates almost always mean the agency is counting "interested" or "let's connect" responses as replies rather than measuring against requests sent.
Q6: "What does a 'meeting' actually mean in your reporting?"
Right-answer pattern: a specific, contractable definition. A meeting should mean a booked, held, qualified call. "Conversations" is not a meeting. "Interested replies" is not a meeting. If the agency cannot define its primary deliverable clearly in a discovery call, the contract definition will be just as slippery.
Want to put this into practice?
Reachium automates LinkedIn outreach, content publishing, and inbox management in one platform.
Start Free →What questions reveal what happens if it does not work?
Bucket 3 covers downside. The funded founder's deepest fear is paying a retainer for 90 days and walking away with a quarterly review slide. These questions are designed to find out whether the agency has skin in the game.
Q7: "What is your guarantee, and what triggers it?"
Right-answer pattern: a specific, time-bound, contract-able commitment with written terms. A guarantee that promises campaigns will be set up and run is not a results guarantee. The trigger criteria matter as much as the guarantee itself. Reachium's DFY engagement ships with a 60-day meeting guarantee, which means the commitment window is defined before you sign, not in a post-mortem call after the contract ends.
Q8: "What happens at day 30 if no meetings have been booked?"
Right-answer pattern: a defined escalation path. A credible operator will name the steps: copy review, ICP tightening, escalation to a senior strategist, or a list rebuild. "We will keep going" with no defined intervention points is not a plan.
Q9: "Can I exit the contract early, and on what terms?"
Right-answer pattern: a written pause or exit clause with clear conditions. Annual contracts with no exit path are the retainer-collector's business model. Month-to-month options exist in the market and typically come at a premium; the presence or absence of an exit clause tells you how confident the agency is in its own work. For context on what DFY LinkedIn contracts typically cost and what the contract structure looks like, see how much does done-for-you LinkedIn lead generation cost?.
What questions reveal who actually does the work?
Bucket 4 surfaces operational depth. The gap between what an agency sells and what it operates is where most retainer disappointments live.
Q10: "Who writes my outreach copy, and who reviews replies?"
Right-answer pattern: named humans with defined roles. "AI handles it" with no human review layer is the spam-and-scale pattern. A managed service is managed. At minimum, the agency should name a copywriter and an inbox manager and describe the review cadence.
Q11: "What is the onboarding cadence, and when does the first campaign go live?"
Right-answer pattern: a specific timeline. Most reputable agencies launch the first campaign in week two or three (after ICP alignment, list approval, and copy sign-off). A vague "we move fast" without a defined launch window is either inexperience or a sign the agency juggles too many clients to give yours a launch date.
Q12: "Who is my point of contact, and how often will we sync?"
Right-answer pattern: a named owner and a scheduled cadence. In the first month, a weekly sync is the minimum accountability structure. "We will be in touch" is not a cadence. If the person you talk to in the sales call is not the person running your account, find out who is before signing, not after.
What questions should the founder skip?
A few questions that feel useful but are not:
- "How long have you been in business?" Operator quality varies by team, not vintage. A two-year-old agency can be better or worse than a ten-year-old one.
- "Do you have testimonials?" Every agency has testimonials. The dashboards in Q4 tell you more in two minutes than a testimonials page tells you in twenty.
- "What is your pricing?" Useful for budget filtering but not for vetting quality. Run the twelve questions first, then discuss price with the agencies that passed.
Want to put this into practice?
Reachium automates LinkedIn outreach, content publishing, and inbox management in one platform.
Start Free →FAQ
Should I ask all twelve questions on the first call or split them across calls?
Lead with the four most diagnostic questions on the first call: Q1 (API vs. browser automation), Q4 (dashboard artifacts), Q6 (meeting definition), and Q9 (exit clause). These four will eliminate most agencies before you invest in a second call. Save the operational questions (Q10-Q12) for agencies that cleared the first screen, where they function as a final due diligence pass.
What if the agency dodges the dashboard question (Q4)?
Treat it as disqualifying. An agency with real results from real clients can produce redacted booking screenshots. If they cannot or will not, one of three things is true: results do not exist, clients have not given permission to share anything, or the agency is protecting a number they know is weak. None of those scenarios belong in your shortlist.
How do I know whether the guarantee is real?
Ask for the guarantee terms in writing before the call ends. A real guarantee defines: what counts as a qualified meeting, how many meetings the guarantee covers, the time window, and the specific remedy if the guarantee is not met. A guarantee with no written terms is a sales line, not a contractual commitment. Get it in writing or treat it as if it does not exist.
What if the agency refuses to name the human team (Q10)?
Push back. A managed service is, by definition, managed by people. If the agency cannot name a copywriter and an inbox manager and describe who reads your replies before they are triaged, you do not have a managed service. You have software access with a retainer on top. That is a different product at a different price point.
Are any of these questions actually rude to ask?
No. Any agency running a serious managed service will answer all twelve without discomfort. The questions that feel uncomfortable to ask are exactly the ones worth asking, because a confident, specific answer is what you are evaluating for. Deflection, irritation, or vague reassurance in response to a direct question is the answer.
Sources
- Reachium - DFY LinkedIn managed service with 60-day meeting guarantee and verified API architecture.
- Marketing Experts Hub: LinkedIn Banned HeyReach.io - Details of the March 2026 HeyReach company page and founder profile ban.
- LinkedIn Professional Community Policies - LinkedIn's published terms on automated activity and third-party integrations.
- Linked Insider: LinkedIn Outreach Benchmarks 2026 - Source for acceptance rate and reply rate benchmarks cited in Q5.
- Linked Insider: Is LinkedIn Automation Safe in 2026? - Architecture comparison: verified API vs. browser automation vs. cloud proxy.
- Linked Insider: How to Choose a LinkedIn Lead Generation Agency - Broader vetting framework this checklist operationalizes.