12 Red Flags When Hiring a B2B SaaS Marketing Agency

Most agency horror stories were predictable in the first sales call. Here is what to look for, and exactly what to ask.

The 12 red flags when hiring a B2B SaaS marketing agency fall into three groups: how the agency sells to you (tactics pitched before strategy, no questions about your sales cycle or ICP, guaranteed lead volumes, pressure to sign fast), how it prices and proves itself (no named case studies, percentage-of-ad-spend fees, opaque budgets, lock-ins with no out), and how it runs the work (one-size-fits-all playbooks, account-team turnover, vague communication cadence, vanity-metric reporting). Every one of them is checkable before you sign, in the sales call, while the agency is on its best behavior.

Getting this wrong costs more than the retainer. A typical engagement takes about a month to onboard and a quarter to ramp, and unwinding a bad one takes another quarter plus whatever the contract says. Pick wrong and you lose two or three quarters of pipeline generation at exactly the stage, Series A to C, when the board watches pipeline hardest.

This guide gives each flag the same treatment: what it looks like, why it predicts failure, what good looks like instead, and one concrete question to ask in the sales call. The questions are the point. Run them live and watch how the room reacts. One disclosure up front: this guide is written by an agency. The advice stands on its own, and the closing section tells you how to use the checklist against its author too.

Key takeaways

• The twelve flags cluster into three failure modes: selling before understanding (flags 1-4), misaligned money and unverifiable proof (flags 5-8), and templated, unaccountable delivery (flags 9-12).
• The single most predictive test is whether the agency asks about your sales cycle, ICP, and deal data before it proposes tactics or numbers.
• Guaranteed MQL volumes contradict how B2B buying works: up to 95% of business clients are not in the market for many goods and services at any one time, per Ehrenberg-Bass Institute research (Professor John Dawes, May 2021), so guarantees get met by degrading lead quality.
• Percentage-of-ad-spend pricing pays the agency more when you spend more, whether or not pipeline follows. A flat retainer removes the conflict.
• A minimum commitment (commonly six months) is not a lock-in. The red flag is a long term with no checkpoint, no out, and auto-renewal. Good agencies earn tenure instead: average client-agency relationship tenure reached seven years in 2025, per the 4As and ANA.
• Every flag comes with one question to ask in the sales call. The at-a-glance table lists all twelve.

The 12 red flags at a glance

Red flags you can catch in the first sales call

The first four flags show up before any contract exists, in how the agency sells. They are the cheapest to check and the most predictive, because an agency is never more careful with you than when it wants your signature.

1. They pitch tactics before strategy

An agency that opens with a channel list before it understands your business is selling deliverables, not pipeline. The tell is a first proposal that names tactics (cold email, LinkedIn ads, SEO, ABM) with volumes and budgets attached before anyone has asked what you sell, who buys it, at what price, or how deals actually close. Tactics only work downstream of those answers. When a tactic-first plan underperforms, the agency has no diagnostic frame to fall back on, so the fix is always more volume, and the burn continues.

What good looks like: discovery before prescription. A serious agency spends the first conversation on your ICP, your funnel data, and your sales motion, and its proposal explains why each channel earns a place for your specific economics, including which channel it would cut first if the data said so.

Ask them this: "Why this channel mix for us specifically, and what evidence would make you kill one of these channels in the first 90 days?" A good answer references your ACV, sales cycle, and ICP, and names a kill criterion. A bad answer defends the package.

2. They ask nothing about your sales cycle, ICP, or deal data

An agency that runs the sales call as a monologue about itself will run the engagement the same way. In B2B SaaS, pipeline math is deal math: without your average contract value, sales cycle length, win rate, and a picture of who actually sits on the buying committee, nobody can set an honest target or an honest timeline. B2B buying cycles are long by nature; corporations change service providers such as their principal bank or law firm around once every five years on average, per Ehrenberg-Bass Institute research by Professor John Dawes (May 2021). An agency that never asks how long your deals take will time everything wrong: budgets, sequences, expectations, and its own reporting.

What good looks like: the agency asks for funnel data, CRM access, or at minimum a structured intake before it commits to any forecast, and its questions get more specific as the call goes on.

Ask them this: "What do you need to know about our sales cycle and ICP before you would put a number in a forecast?" The good answer is a list. The bad answer is some version of "our playbook works across B2B."

3. They guarantee a specific number of leads or MQLs

A guaranteed lead volume quoted before anyone has seen your data is a sales tactic, not a forecast. Up to 95% of business clients are not in the market for many goods and services at any one time, per Ehrenberg-Bass Institute research by Professor John Dawes (May 2021), so the pool of genuinely in-market buyers in any given month is small and outside the agency's control. A guarantee like "30 MQLs a month" therefore gets met the only way it can be: by degrading quality. Targeting loosens, definitions stretch, and the CRM fills with contacts your sales team learns to ignore.

What good looks like: honest channel-level ramp expectations and leading indicators instead of volume promises. Credible agencies quote ramp timelines, with cold email taking roughly three to four weeks to launch cleanly, LinkedIn outreach two to three, and paid media faster, and they commit to what they control: the system, the iteration speed, and transparent kill criteria.

Ask them this: "Tell me about a client who missed the guaranteed number. What changed, the plan or the definition of a lead?" If they claim no client has ever missed it, that is the answer.

4. They pressure you to sign fast

Manufactured urgency in an agency sales process predicts how you will be treated after the signature. Expiring discounts, proposals valid for 48 hours, and "one slot left for your category" are pressure mechanics borrowed from transactional selling, and they signal a volume business model: sign many, churn many, replace the churn with the next founder in the pipeline. An agency with real capacity planning and a real onboarding process (about four weeks is normal for a full go-to-market engagement) has no reason to rush you past diligence, because the engagement only works if the fit is real.

What good looks like: comfort with a two-week evaluation, references offered before you ask, and a willingness to scope the first 90 days in writing before any signature.

Ask them this: "Can we talk to two current clients and one former client before we sign?" The answer matters less than the reaction. Hesitation on the former client is the tell.

Red flags in proof, pricing, and the contract

The next four flags live in the commercial layer: what the agency can prove, how it charges, and what the paper commits you to. Each is a question of incentive alignment, and misaligned incentives beat good intentions over a six-month engagement every time.

5. No relevant case studies or named proof

A logo wall is not proof, and proof from a different motion is not proof for yours. B2B SaaS demand generation is a specialized discipline: long cycles, buying committees, high contract values, and outcomes that have to show up in a CRM. Results an agency earned in ecommerce, local services, or consumer apps do not transfer, and results it will not attach a name to cannot be checked at all. "We can't share details due to confidentiality" for every single client is a pattern, not a policy.

What good looks like: named clients, on-the-record testimonials attributed to a person, a role, and a company, and at least one case from a company at your stage with a comparable contract value and sales motion. Video testimonials outrank logo walls, and a reference call outranks both.

Ask them this: "Show me a named client with a similar ACV and sales cycle to ours. What happened in their first 90 days?" Vague averages across an unnamed portfolio are a no.

6. Pricing is a percentage of your ad spend

Percentage-of-spend pricing pays the agency more when you spend more, whether or not pipeline follows, and that conflict shapes every recommendation you will receive. When the fee is a cut of media, "increase the budget" is never a neutral suggestion, and efficiency actively costs the agency money: cut wasted spend in half and you have cut the agency's revenue in half. The model is a legacy of media buying at enterprise scale, where some large shops manage the conflict with guardrails. At Series A to C budgets, there is no reason to accept it.

What good looks like: a flat retainer scoped to the service level and workload, so the fee is identical whether your monthly media budget rises or falls, and the only way the agency grows the account is by results earning an expansion.

Ask them this: "If you cut our ad spend 30% next quarter while holding pipeline flat, what happens to your fee?" A flat-retainer agency answers "nothing" without pausing.

7. They cannot tell you where your budget goes

An agency that will not break down your budget is hiding either margin or effort, and usually both. The tells: one bundled invoice line, media markup folded invisibly into "management," tool costs nobody can itemize, and, most seriously, ad accounts, sending domains, and dashboards that live under the agency's login instead of yours. Opacity is not a billing annoyance; it is an evaluation blocker. If you cannot see what portion of spend actually reached the market, you cannot compute CAC or payback, which means you cannot evaluate the agency at all. For some agencies, that is the point.

What good looks like: you own every account (ad platforms, CRM, sending domains, analytics), the invoice separates fees, media, and tools line by line, and reporting reads from systems you can audit yourself.

Ask them this: "Do we own every account and domain from day one, and can you split the retainer into fees, media, and tooling on the invoice?" Two clean yeses, or walk.

8. A long lock-in with no checkpoint and no out

A minimum commitment is not a lock-in; a lock-in is a term the agency can only justify with a signature. Minimum terms exist for a legitimate reason in B2B SaaS: outbound infrastructure takes weeks to warm up, sales cycles run longer than a quarter, and neither side learns anything real in 90 days. That is why six-month minimums are common and defensible. The red flag is the asymmetric version: a twelve-month term, renewal by silence, no defined performance checkpoint, and a termination penalty that makes leaving more expensive than staying.

Agencies that trust their work do not need legal retention, because earned retention is what the industry actually runs on when the work is good: average client-agency relationship tenure reached seven years in 2025, more than double the 2016 figure, per the 4As and ANA Client-Agency Relationship Tenure report (April 2025).

Ask them this: "After the minimum term, what is the cheapest and fastest way out if this is not working, and which checkpoint do we review against?" Good answers name a specific review point (day 90 is common) and a 30-day out.

Red flags in how the work actually runs

The last four flags fully reveal themselves only after the start, but each casts a shadow you can catch in the sales process if you know where to look.

9. A one-size-fits-all playbook

If the proposal would read the same with another company's logo on it, you are buying a template. The same channel mix, the same sequences, the same "proven system" pitched identically to a product-led dev tool and an enterprise fintech is a process optimized for the agency's margin, not your market. What works in B2B SaaS is a function of ACV, sales cycle, category maturity, and how your buyers actually research, and a play that fills pipeline for a $60K ACV sales-led motion will burn cash for a $6K ACV product-led one. Templates also leak: your prospects have already received the same cold email skeleton from the agency's other clients, which costs you replies and domain reputation at the same time.

What good looks like: a plan whose assumptions are stated and falsifiable, built from your data, with a first 90 days that could not belong to another client.

Ask them this: "What in this proposal would change if our ACV were half what it is?" If the answer is nothing, the proposal was finished before you ever spoke.

10. Seniors sell it, juniors run it, and the team keeps changing

The people in the sales call and the people who will run your account are often different people, and that gap is a margin structure, not an accident. The related flag is turnover after kickoff: every account-team handoff resets the context that makes an engagement compound, the ICP nuance, the message learnings, the record of what was tested and failed. Each reset is invisible on the invoice while you pay full rate for re-learning. If your day-to-day contact changes twice in six months, results follow the context out the door.

What good looks like: you meet the actual operating team before signing, the pod is named in the agreement, and the agency documents its systems well enough that knowledge survives any individual departure.

Ask them this: "Who exactly touches our account week to week, can I meet them before signing, and how long has each of them been with you?" Watch for a pitch team that cannot answer the third part.

11. A vague communication cadence

How an agency communicates during the sales process is the best it will ever communicate with you, so vagueness now is a promise of silence later. If "what does a normal week look like" gets answered with "we'll set up a monthly call," you have just learned the operating rhythm: a monthly deck, assembled the day before, describing a month you can no longer influence. B2B campaign feedback loops are weekly at most. A failing sequence or a broken tracking setup discovered five weeks late is a month of budget gone.

What good looks like: the cadence written into the agreement. A weekly or biweekly working call, a shared Slack channel with a stated response expectation, a live dashboard you can open at 7 a.m. without asking anyone, and a named owner for when something breaks.

Ask them this: "Walk me through week three of a normal month: which meetings happen, what do I see without asking, and who do I message when something breaks?" Specific answers only exist where a real operating system exists.

12. Reporting on vanity metrics instead of pipeline

A report that leads with impressions, clicks, and open rates is a report designed to survive a bad quarter. Vanity metrics can all rise while revenue falls, and every experienced B2B SaaS operator has watched it happen. The deeper tell is where the data lives: an agency reporting from its own tools, on metrics it defines itself, has structured the engagement so it can never lose an argument. MQL counts with no downstream tracking invite gaming, since definitions stretch and targeting loosens until the number is met.

What good looks like: reporting that reads from your CRM, not the agency's slides. Opportunities created, pipeline dollars by source, influence on win rate, CAC and payback where the data supports it, with every touch landing on contact and account records you own, surfaced in a live dashboard rather than a monthly PDF.

Ask them this: "Show me a real monthly report, anonymized is fine. Where does pipeline appear, and does that number come from the client's CRM or from your tools?" If pipeline is on slide nine and the source is the agency's own tool, you have your answer.

How do you actually pick a B2B demand gen agency?

To pick a B2B demand gen agency, invert the twelve flags into a filter and run it in order: shortlist only agencies with named proof in your motion, put every finalist through the sales-call questions in this guide, and weight how they answer over what they answer.

1. Shortlist on proof, not polish. Two or three finalists with named B2B SaaS results at a stage and ACV comparable to yours. Comparison research helps here; start from category roundups like the best GTM engineering agencies and the best B2B SaaS paid media agencies, then verify the proof yourself.
2. Run the questions live. The twelve questions take about twenty minutes of a discovery call. Specific, falsifiable answers signal a real operating system; polished evasions signal a sales machine.
3. Talk to a former client, not just current ones. Current clients tell you the ceiling. Former clients tell you the floor, and how the exit was handled.
4. Read the first 90 days before the contract. A serious agency will write down what happens in the first quarter, onboarding, ramp milestones, and the checkpoint you will both review against, before you sign anything.
5. Decide on the operating system, not the deck. The best predictor across every flag in this guide is the same: agencies that understand your deal math, align their fees with your outcomes, and report in systems you own tend to be good at everything else too.

What separates the good ones is not a secret channel or a proprietary AI. It is discovery before prescription, incentives that do not fight yours, a team that persists, and reporting that lands in your CRM where it can be checked. Everything else is presentation.

The bottom line: make the agency pass its own checklist

Every agency that publishes an article like this one has an interest in the outcome, so here is ours, stated plainly. Understory Agency is an allbound GTM agency for funded Series A to C B2B SaaS, and this list doubles as the standard we ask prospects to hold us to: custom flat retainers for each service, never a percentage of spend; named, on-the-record client testimonials and video case studies; a defined pod (a GTM engineer, an ops manager, and a paid media strategist at minimum); and reporting that lands on CRM contact records with live dashboards, not a monthly PDF.

Bring all twelve questions to your first call, with us or with anyone else you are evaluating. The agency that welcomes the checklist is the one you want. The agency that bristles at it just showed you flag thirteen for free.

Related reading

• A step-by-step evaluation framework
• How agency pricing models actually work

FAQ

What are the red flags when hiring a B2B SaaS marketing agency?

The twelve red flags: tactics pitched before strategy, no questions about your sales cycle or ICP, guaranteed lead volumes, pressure to sign fast, no named or relevant proof, percentage-of-ad-spend pricing, opaque budgets and account ownership, long lock-ins with no out, one-size-fits-all playbooks, account-team turnover, vague communication cadence, and vanity-metric reporting. Each one is detectable in the sales call with a direct question, and the strongest single test is whether the agency asks about your sales cycle, ICP, and deal data before proposing tactics or numbers.

What questions should you ask a B2B marketing agency before signing?

Five questions expose the most: Why this channel mix for us specifically, and what would make you kill one? What do you need to know about our sales cycle before you would forecast? Tell me about a client who missed a target, and what changed? If we cut ad spend 30% while holding pipeline, what happens to your fee? And: show me a real monthly report, and where pipeline appears in it. Weight how they answer over what they answer; specific, falsifiable answers signal a real operating system, and polished evasions signal a sales machine.

Is percentage-of-ad-spend pricing a red flag?

For Series A to C B2B SaaS budgets, yes. Percentage-of-spend pricing pays the agency more when you spend more, whether or not pipeline follows, so budget recommendations are never neutral and efficiency costs the agency revenue. The model is a legacy of enterprise media buying, where some large shops manage the conflict with guardrails, but the cleaner structure is a flat retainer scoped to service level and workload, where the fee does not move with the media budget and account growth has to be earned with results.

How long should a B2B SaaS marketing agency contract be?

A six-month minimum commitment is common and defensible in B2B SaaS. The minimum exists because outbound infrastructure takes weeks to warm up and B2B sales cycles run longer than a quarter, so 90 days proves little either way. The red flag is not the minimum; it is a twelve-month lock-in with auto-renewal, no defined performance checkpoint, and termination penalties. Look for a named review point (day 90 is typical) and a 30-day out once the minimum term has run.

How quickly should a B2B marketing agency show results?

Channel by channel, on honest timelines: cold email typically takes three to four weeks to launch cleanly, LinkedIn outreach two to three, and paid media faster, with first replies and booked meetings landing as early as the end of month one and the compounding effect of an integrated program visible around month three. Onboarding itself usually takes about four weeks before campaigns launch at full speed. An agency promising meaningful pipeline in week one is overpromising against how B2B buying actually works, and one refusing to name any timeline at all is hiding the ramp. What you want is a stated ramp, leading indicators per channel, and a defined checkpoint to review them.