# Growth Marketing for SaaS

> What growth marketing owns in a SaaS org, how it differs from demand gen and product marketing, and the first three hires that make the function produce.

Source: https://saas-marketing.net/guides/growth-marketing-for-saas/
Topic: SaaS Growth Marketing
Type: guide
Published: 2026-09-11
Last updated: 2026-09-11
Publisher: SaaS Marketing (saas-marketing.net)
License: CC BY 4.0. Quote or republish with attribution and a link to https://saas-marketing.net/guides/growth-marketing-for-saas/

## Short answer

Growth marketing in SaaS owns the conversion surfaces between a visitor and a paying, expanding account: signup flow, activation, monetisation screens, lifecycle triggers and the experiment programme that improves them. Demand generation owns getting people to the site. Product marketing owns what the product means. The function fails most often because nobody wrote that boundary down, and because the team has no dedicated engineering capacity, which turns it into a research group that produces decks instead of shipped changes.

## Key takeaways

- Growth owns signup, activation, monetisation surfaces, lifecycle triggers and experimentation. Demand gen owns traffic.
- A growth team without dedicated engineering capacity will produce recommendations, not shipped changes.
- The three org models each break differently: inside marketing, inside product, or as a standalone pod.
- First three hires in order: growth generalist, lifecycle or product marketer, then growth engineer or analyst.
- Judge the team on activation rate, free to paid conversion and experiment velocity, not on traffic.
- Write the charter before the first hire, including an explicit list of what growth does not own.

---

Most failed growth teams did not fail at experimentation. They failed because two months in, nobody could say whether growth or demand gen owned the pricing page, and six months in the team had shipped four experiments and written nineteen documents. The charter is the whole game. Write it before the first hire.

## What growth owns, precisely

Five surfaces. Everything else is somebody else's.

**The signup and trial flow.** From the moment someone clicks the button to the moment they are inside with an account. Form fields, verification, plan selection, the first screen.

**Activation.** Getting a new account to the moment of first real value, which you have defined as a specific event. If you have not defined that event, the growth function has nothing to aim at and should not be hired yet.

**Monetisation surfaces.** Pricing page, upgrade prompts, paywalls, seat expansion flows, usage limit messaging. These are product surfaces with commercial consequences, which is why they belong to a commercial team with engineering access.

**Lifecycle triggers.** Behaviour-driven messaging in product and over email. Not the newsletter. The sequence that fires when someone creates their second project and then stops.

**The experiment programme.** The instrumentation, the queue, the analysis, and the discipline of killing losers quickly.

Write the list of things growth does not own and get it signed by the demand gen and product marketing leads. Typically: paid media budget, the blog, category positioning, pricing levels (as opposed to pricing page layout), sales collateral. Most charters describe scope and omit the boundary, which is like drawing a country without borders.

## Growth versus demand generation versus product marketing

The three functions overlap on exactly one thing, which is the buyer, and they are otherwise doing different jobs with different tools.

The most common real-world collision is the pricing page. Demand gen wants it optimised for conversion from paid traffic, product marketing wants it to communicate positioning, growth wants to test layouts weekly. Resolve it by giving growth the layout and test rights, product marketing veto over claims and tier descriptions, and demand gen the right to ask for a variant for a campaign. Written down. In advance. The deeper comparison is in [product marketing vs growth marketing](/comparisons/product-marketing-vs-growth-marketing/), which is worth reading alongside this if you are redrawing both charters at once.

## Three org models and how each one breaks

**Growth inside marketing.** The CMO owns it, the team sits with demand gen, budget is easy and engineering is hard. Works well when your most influential surfaces are marketing-controlled pages. Breaks when the real constraint is in-product, because now a marketing team is negotiating with product for backlog space every sprint and losing.

**Growth inside product.** Reports to a product leader, has an embedded engineer, ships fast. This is how most product-led companies eventually arrange it. Breaks on the other side: the team drifts toward product work, loses its commercial framing, and stops caring about where the traffic came from or what it cost.

**Standalone pod.** Reports to a founder or COO, has its own engineer and designer, chooses its own targets. Genuinely effective under about 150 people, because the founder can settle boundary disputes in a hallway. Past that headcount it becomes a political problem, because a team with no functional home collects enemies and no defenders.

If the company has not defined an activation event, all three models fail identically. Growth optimises toward whatever it can measure, which will be signups, and signups without activation is a metric you can double while revenue stays flat. Define activation first. It is a product and data decision, not a marketing one.

Large product-led organisations tend to run several pods rather than one team. HubSpot, Miro and Atlassian all built PLG-era growth groups this way, with pods attached to specific surfaces (acquisition, activation, monetisation) rather than one team responsible for everything. That structure needs enough engineering capacity to staff each pod, which is why it does not work at 5 million ARR no matter how appealing the diagram looks.

## The first three hires, in sequence

Sequence matters more than titles. Hire in this order and each person makes the next one more effective.

**Hiring sequence**

On compensation, ranges in 2026 vary sharply by stage and location, and growth engineers cost what engineers cost, not what marketers cost. That is the budget surprise most first-time growth leaders hit. Write the roles precisely using the [job description templates](/templates/saas-marketing-job-descriptions/) rather than posting for "a growth person", which attracts either paid media specialists or product managers and rarely the person you meant.

## The weekly operating cadence

Growth is a rhythm function. When the rhythm goes, the output goes with it, usually within a month.

| When | What | Who | Output |
| --- | --- | --- | --- |
| Monday, 30 min | Experiment stand-up: what shipped, what is live, what is blocked | Full pod | Updated board, one escalation |
| Wednesday, 45 min | Results review on anything that reached its sample | Growth lead, analyst | Ship, kill or iterate decision |
| Friday, 60 min | Queue grooming, next two experiments briefed | Full pod | Two written briefs |
| Monthly, 90 min | Funnel review with demand gen and product | Leads | Reprioritised queue |
| Quarterly | Charter and metric review | Leads plus exec sponsor | Boundary adjustments in writing |

Two rules keep this honest. Every experiment gets a written brief before it is built, including the hypothesis, the required sample and the duration. And every experiment gets a decision at the end, recorded, including the ones that did nothing. Use the [experiment brief template](/templates/experiment-brief/) so the format is consistent and old experiments are searchable, because the most expensive thing a growth team does is rerun a test it already lost two years ago.

## The metrics the team is judged on

Four, and traffic is not among them.

- **Activation rate** against a defined event, measured by weekly cohort. The single most important number the function owns.
- **Free to paid or trial to paid conversion**, broken out by acquisition source so you can see whether a demand gen change is masquerading as a growth result.
- **Expansion revenue** where growth owns upgrade surfaces, otherwise excluded.
- **Experiment velocity**, meaning tests reaching a decision per month. A health metric, not a goal. Gaming it is easy and obvious.

Model what those rates need to be for the economics to work before you set targets. The [growth model spreadsheet](/templates/growth-model-spreadsheet/) will show you quickly whether a two point activation improvement matters at your ACV or whether the constraint is actually somewhere in demand gen, in which case the [budget allocator](/calculators/demand-gen-budget-allocator/) and the [demand generation plan template](/templates/demand-generation-plan-template/) are the more useful tools this quarter.

## Where the function gets stuck

Almost always engineering capacity. A growth team with a shared backlog and no committed engineer becomes a research team within two quarters. It writes good analyses, proposes sensible changes, and watches them sit behind roadmap items. Morale goes, the best person leaves, and the company concludes that growth marketing does not work here.

I will state it flatly: a growth team without dedicated engineering is a research team, and it will produce decks instead of shipped changes. If you cannot commit an engineer, narrow the charter to marketing-controlled surfaces, name it demand gen plus conversion optimisation, and do not call it growth. That is not a lesser job. It is just an honest one.

Two other stall points worth naming. Event data quality, which usually costs a full quarter to fix properly and which no exec wants to fund. And a company that has not reached product market fit, where growth work amplifies a leaky bucket; [scaling growth after product market fit](/playbooks/scaling-growth-after-pmf/) covers the sequencing, and the short version is that optimising activation before the product retains anybody is wasted money.

## What to do next

Write the one page charter this week: surfaces owned, surfaces not owned, metrics, engineering capacity, decision rights, cadence. Get the demand gen and product leads to sign it. Then define your activation event if you have not. Only after both of those should you open the first requisition. The broader operating picture sits in [SaaS growth marketing](/saas-growth/), but the charter is the part that determines whether any of the rest of it works.

## Frequently asked questions

### What is the difference between growth marketing and demand generation?

Demand generation creates and captures demand outside the product: paid media, SEO, events, outbound support. Growth marketing works on what happens after arrival: signup conversion, activation, upgrade paths and lifecycle messaging. The handoff point is usually the signup or demo form. Companies that leave that boundary undefined end up with two teams both optimising the same landing page.

### Should the growth team sit in marketing or product?

Inside product if your motion is product-led and the work needs engineering every week. Inside marketing if the most influential surfaces are pages and campaigns rather than in-product flows. A standalone pod reporting to a founder works at early stage and tends to create political friction once the company is past about 150 people.

### What does a SaaS growth marketer actually do day to day?

Runs experiments on conversion surfaces, writes and ships lifecycle messaging, builds the instrumentation to measure both, and reports on activation and conversion. In practice a lot of the week is spent getting engineering time and cleaning event data. The ratio of analysis to shipping is the clearest signal of whether the function is healthy.

### How big should a SaaS growth team be?

At 3 to 10 million ARR, two to four people plus dedicated engineering is typical. The headcount matters less than whether engineering capacity is committed. Three growth marketers sharing zero engineers will do less than one growth marketer with one engineer, because only the second arrangement can ship a change to the signup flow.

### What metrics should a growth team be judged on?

Activation rate against a defined activation event, free to paid or trial to paid conversion, expansion revenue where the team owns upgrade surfaces, and experiment velocity as a health measure. Traffic and MQLs belong to demand generation. Giving growth a traffic target is the fastest way to blur the two functions permanently.

### Do you need engineers on a growth team?

Yes. A growth team without committed engineering time can only recommend changes to surfaces other people control, which means every idea waits in someone else's backlog. One embedded engineer changes the output of the function more than two additional marketers. If you cannot fund that, narrow the charter to things marketing can ship alone.

### What does a growth marketing charter contain?

The surfaces the team owns, the surfaces it explicitly does not own, the metrics it is judged on, its engineering capacity, its decision rights on shipping without approval, and the weekly cadence. One page. The section that prevents the most conflict is the list of things growth does not own, which most charters omit.