Growth Agency vs Demand Generation Agency: A Decision Framework for B2B Founders

Last quarter a founder told me their agency was “great but not working.” It was a growth agency, and the founder needed a demand generation agency. Fair enough. The agency had delivered a clean brand refresh, a content calendar and a landing page test plan. The founder needed 40 qualified demos a month. Nobody had lied to anybody. The shape was simply wrong.

This is the most common expensive mistake in the growth agency vs demand generation agency decision. The two labels get used as synonyms on agency websites, but they describe different delivery shapes with different teams, different reporting and different failure modes. Pick the wrong one and you lose a quarter, which at the $50K to $150K MRR stage is a lot of runway.

Here is the framework we use to decide which shape fits, and the questions that expose which shape an agency really is, whatever its homepage says.

Growth Agency vs Demand Generation Agency: The Core Difference

Strip away the marketing language and the split is simple.

A demand generation agency is an execution engine. You already know who buys and where to reach them. The agency builds campaigns, content, paid programs and outbound sequences that turn that knowledge into qualified pipeline, then optimizes conversion rate and cost per opportunity.

A growth agency is a discovery engine. You have revenue but no repeatable way to add more. The agency runs structured experiments across the funnel (positioning, pricing page, onboarding, channels, retention) to find which levers move revenue, then hands you the ones that work.

Both want more revenue. The difference is what they assume you already have.

Dimension	Demand generation agency	Growth agency
Core assumption	ICP, offer and channel are validated	Something in the funnel is unvalidated
Primary job	Create qualified pipeline	Find and fix the biggest revenue constraint
Typical scope	Paid, content, email, outbound, events	Full funnel, including product and retention
Main metric	SQLs, pipeline created, cost per opportunity	Experiment win rate, conversion deltas, revenue per cohort
Time to first signal	30 to 60 days	60 to 90 days
Output you keep	Running programs and assets	Playbooks, tested channels, a prioritized backlog
Biggest risk	Scaling a broken offer faster	Experiments that never become volume

If you want depth on the demand generation side, we wrote about the delivery model behind a demand generation agency in detail, including retainer, hybrid and performance structures. For the growth side, how to evaluate a growth agency covers the scoring dimensions. This post sits above both and answers the earlier question: which shape do you need at all?

Why Founders Buy the Wrong One

Three patterns cause most mismatches.

Pattern 1: The label is the pitch. Agencies pick whichever label ranks. A shop that runs paid social calls itself a growth agency because the term has more search volume. A shop that runs experiments calls itself demand gen because that is what the budget holder searches for. You cannot trust the name.

Pattern 2: The founder describes the symptom, not the constraint. “We need more leads” is a symptom. The constraint might be a weak offer (conversion problem), a thin channel (volume problem) or a leaky onboarding flow (retention problem). Only the second one is a demand gen job.

Pattern 3: The agency answers the question it is good at. Ask a demand gen shop why pipeline is flat and you will hear about campaign volume. Ask a growth shop and you will hear about conversion experiments. Each is right inside its own frame and blind outside it.

The benchmark data makes the stakes concrete. Vendor surveys from 2026 put the median B2B MQL to SQL rate near 13 percent, with SaaS running higher at 18 to 22 percent and top quartile teams reaching 39 to 40 percent (the SaaS Hero benchmark roundups are the source for those figures, and they come from a vendor, so treat them as directional). Sales cycles run about 75 days under $20K ACV and 90 to 120 days between $20K and $100K. If your funnel converts at the median and your cycle is 100 days, a demand gen agency launched today will not show closed revenue until January. That is normal. It is also why a mismatch costs two quarters instead of two weeks.

The Four Questions That Decide It

We use a short diagnostic before recommending either shape. It is not clever. It forces the founder to say what they actually know.

Question 1: Can you name the channel behind your last five closed deals?

If yes, and four of the five came through one channel (say outbound or organic search), you have a validated channel. Demand gen can scale it.

If the answer is “referrals, a conference, a Product Hunt launch and one inbound,” you have no repeatable channel. A demand gen agency will spend your retainer guessing on your behalf, which is exactly what a growth agency is structured to do properly.

Question 2: Does your demo to close rate hold when volume goes up?

Take your last 90 days. If demos convert to closed won at a stable rate across weeks, your offer and sales process can absorb more pipeline. If conversion swings wildly or collapses whenever you run a campaign, more leads will not help. The constraint is downstream, and that is growth agency territory (or a sales systems fix, not a marketing agency at all).

Question 3: Is your ICP written down with numbers?

A usable ICP names company size, role, trigger event and the pain in the buyer’s words. Demand gen agencies need this on day one, because every targeting decision flows from it. If you cannot hand over a one page ICP, the first 30 days of any engagement is ICP work, and you should price it as a growth engagement.

Question 4: Who owns the output when the agency leaves?

A demand gen agency leaves behind running programs that need maintenance. A growth agency leaves behind learning and a backlog. Decide which you can staff. If you have no marketer in house, a growth playbook you cannot execute is shelfware. If you have a strong marketing lead, a demand gen team plugs into their roadmap with little friction.

The Fit Scoring Framework

Turn the four questions into a score. We call this the Constraint Fit Score. Score each line from 0 to 2 and add them up.

Signal	0 points	1 point	2 points
Validated channel	None	One channel, mixed results	One channel behind most closed deals
Close rate stability	Swings widely	Stable on small volume	Stable across volume changes
ICP documentation	Verbal only	Rough doc, no data	One page doc with firmographics and triggers
In house execution capacity	No marketer	Part time owner	Dedicated marketing lead
Tracking quality	CRM has gaps	Source tracked, stages messy	Pipeline by source reported weekly
Sales cycle known	No	Rough guess	Measured by segment

Read the total like this:

• 0 to 4 points: Buy a growth agency, or fix the basics first. You are missing the inputs demand generation depends on.
• 5 to 8 points: Hybrid. Start with a 60 to 90 day growth engagement scoped to the weakest two lines, then move to demand gen.
• 9 to 12 points: Buy a demand generation agency. Anything else is paying for discovery you have already done.

Notice that the score is about your readiness, not the agency’s quality. A great demand gen agency fails on a 3. A great growth agency wastes money on an 11.

What Each Model Costs and What the Money Buys

Pricing in this market is messy, and most published numbers come from agencies selling the service, so read them with that in mind. Vendor surveys put B2B SaaS demand generation retainers between roughly $3,000 and $25,000 per month, with the low end buying basic campaign management and the high end buying fractional CMO level strategy. Growth focused agencies serving Series A to C SaaS are often quoted near $8,000 to $15,000 per month. Ad spend is almost always on top.

Budget band	What a demand gen shop does with it	What a growth shop does with it
$3,000 to $5,000 per month	One channel, light reporting	Rarely offered; too thin for experiments
$5,000 to $10,000 per month	Two channels, content plus paid or outbound	Focused experiment program on one funnel stage
$10,000 to $15,000 per month	Multi channel program, pipeline reporting	Cross funnel experiments with analytics support
$15,000 to $25,000 per month	Strategic lead plus full execution team	Embedded growth lead and engineering support

Two things matter more than the sticker price.

First, what the retainer is measured against. A demand gen retainer should tie to pipeline created and cost per opportunity. A growth retainer should tie to experiment velocity and measured lift. If a growth agency reports MQL volume as its headline number, it is behaving like a demand gen shop without the accountability.

Second, who carries the learning cost. In a growth engagement, many experiments lose. That is the point. If you cannot stomach paying for a losing test, you are not buying growth, you are buying demand gen with extra steps.

Reporting Tells You Which Shape You Bought

Ask both types of agency to show a sample monthly report. The structure gives it away.

A demand generation report leads with pipeline created, SQLs by source, cost per SQL and spend efficiency. It should connect to your CRM, not to the ad platform. If the headline is impressions or clicks, you have an activity shop.

A growth report leads with hypotheses tested, results, decisions made and the next backlog. The best ones show the expected value of each experiment before it ran and compare it to what happened.

A mismatched report is the earliest warning sign. If you bought pipeline and the report is a list of tests, or you bought experiments and the report is a lead count, the shape is wrong, and no amount of optimization fixes it. Our guide on questions to ask a SaaS marketing agency before signing includes the contract level checks that catch this before the first invoice.

The Hybrid Path: Sequence, Don’t Blend

Founders in the middle of the scoring range often ask for an agency that does both. Sometimes that works, but in our experience it works best as a sequence, not a blend.

Phase 1 (days 1 to 60): Constraint hunt. Run a growth style engagement against your two weakest signals. Typical outputs are a written ICP, a fixed tracking setup, one channel test with a clear pass or fail threshold and a recommendation.

Phase 2 (days 60 to 90): Decision gate. Either a channel passed and you scale it, or none did and you go back to Phase 1 with new hypotheses. Do not let momentum carry you into scaling something that failed its own threshold.

Phase 3 (day 90 onward): Scale. Switch the scope to demand gen. Same agency or a different one, whichever executes best. Now the retainer buys volume on a validated motion.

The mistake is skipping the gate. We have seen teams sign a six month “growth and demand” retainer and spend month five still debating what the channel is.

Where Other Agency Types Fit

Neither model is the right answer for every pipeline problem. Three adjacent options are worth knowing about.

• ABM agency. Better when your market is a few hundred named accounts and deal sizes are large. We compared the tradeoffs in our ABM vs demand gen decision guide.
• Outbound or cold email specialist. Better when the validated channel is outbound and the constraint is list quality and deliverability, not strategy.
• AI native marketing agency. Better when you want the same execution at a different cost structure. Ask any AI native agency how its pricing changed when it adopted AI; that separates real AI delivery from a traditional agency using AI tools.

If you are scanning the wider market first, the founder’s guide to B2B marketing firms maps every type side by side.

Red Flags by Agency Type

Red flag	Demand gen version	Growth version
Vague scope	”Full funnel pipeline” with no channels named	”Growth experiments” with no hypothesis format
Wrong headline metric	Leads and impressions	Traffic or follower growth
No stage fit	Pitches the same plan to a $20K and $200K MRR company	Promises a channel before auditing yours
Opaque ownership	Ad accounts held in agency name	Experiment data held outside your analytics
No exit plan	Programs cannot run without the agency	Learnings never documented

Two rules cover most of this. Insist that every ad account, analytics property and CRM integration stays in your name. And insist on a decision gate at day 90, either to scale, change course or stop.

A 30 Day Process for Choosing

If you are about to start an agency search, this is the sequence I would follow.

Week 1: Run the Constraint Fit Score. Be honest on every line. Write down your number and the two weakest signals.

Week 2: Write the brief around the constraint. Not “we need more leads.” Something like “channel one drives most closed deals, we want to double qualified demos from it without lowering close rate.” Or “we have no repeatable channel and need three tested hypotheses by day 90.”

Week 3: Interview three agencies of the matching shape. Ask each for a sample report, a named team and a reference from a company at your stage. Ask what they would refuse to do for you. Good agencies have an answer.

Week 4: Decide on scope and gate. Agree what success looks like at day 90 in writing, who owns each account and what happens if the gate is missed.

Common Mistakes

1. Buying on the label. Read the report sample, not the homepage.
2. Scaling an unvalidated offer. Demand gen amplifies whatever is there, including a bad offer.
3. Expecting growth shops to run volume. They are built to learn, not to produce 500 touches a week.
4. Expecting demand gen shops to fix the funnel. They will report on it, rarely redesign it.
5. Ignoring cycle length. With a 100 day sales cycle, judge the first quarter on pipeline created, not closed revenue.
6. No day 90 gate. Without one, every engagement drifts.

What to Do This Week

Run the Constraint Fit Score before you take another agency call. If you land at 4 or below, spend your first 30 days on the inputs: ICP, tracking and one channel test. If you land at 9 or above, write the pipeline brief and shop for a demand generation agency with a CRM connected report.

If you want a second opinion on your score, we run this diagnostic as part of a free growth audit, and we will tell you plainly which shape fits, even when the answer is not us. You can also try the free AI growth tools at app.momentumnexus.com to pressure test your funnel numbers first.