Marketing a SaaS Product

Most people handed a SaaS product to market start by writing a launch plan. That’s the fifth step, not the first. The job runs in a sequence: understand the buyer, decide what the product means, shape how it’s sold and priced, put it in market, then make people use it. Skip a stage and the later ones absorb the damage, usually in the form of a messaging document nobody in sales can repeat back.

This page is written for the person who owns one product. Not a portfolio, not a category. One product, one roadmap, one set of reps to enable. If you own several, run this sequence per product and stagger the research months so you’re never doing two at once.

What does a product marketer own that nobody else does?

Three things: the buyer’s actual words, the narrative, and the bridge between what engineering shipped and what a rep says on a call. Everything else is shared. Demand gen owns the spend, product owns the roadmap, sales owns the number.

That ownership is why SaaS product marketing strategy sits upstream of campaign work rather than beside it. A campaign built on weak positioning still spends the budget, it just converts worse and nobody can tell you why for two quarters.

The research month: what to run and what it costs you

Research is the only stage you cannot delegate, and it’s the one that gets cut when the launch date moves up. Block four weeks and defend them.

Run three streams in parallel. Buyer interviews with people who bought in the last ninety days, ten to fifteen of them, thirty minutes each. Win loss calls, ten minimum, and insist on at least four losses because losses carry the information wins don’t. Review mining across G2, Capterra and Reddit for your product and the two competitors buyers name most, pulling the exact phrasing people use to describe the problem.

The interview questions that produce usable material are boring. What were you doing before you looked for this? Who else had to approve it? What almost stopped the purchase? That last one surfaces the objection your pricing page needs to answer. This is jobs to be done work in practice, not theory: you’re reconstructing the situation that created demand.

What it costs: about 35 hours of your time, plus $500 to $1,500 in incentives if you’re paying for non customer interviews. Cheap relative to a mispositioned launch.

What positioning and messaging should produce

Three artefacts, and if you produce more than three nobody will read any of them.

First, a one page narrative. The market shift, the problem it creates, why existing approaches fail, what your product does instead. Under 400 words. This is the thing a CEO repeats on a podcast and a rep opens a call with.

Second, a message map by persona. For each buyer role, the problem in their words, the capability that solves it, and the proof point. At $20K ACV you typically have three roles: the practitioner who feels the pain, the manager who owns the budget line, and a security or IT reviewer who can veto.

Third, a proof inventory. Every customer number, logo, benchmark and screenshot you’re allowed to use, with an expiry date on each. Proof rots. A 2024 case study quoted in 2026 does damage.

The packaging and pricing input product marketing owes

You don’t set the price. You do decide whether the buyer can understand it in eleven seconds, which is roughly how long a pricing page gets before someone hits back.

Bring three inputs to the packaging conversation. The competitor prices buyers quoted at you unprompted during interviews, which are more accurate than published list prices. The feature that appeared in every single upgrade conversation, which is your natural tier boundary. And the language test: read your proposed tier names to five customers and ask which one they’re on. If more than one hesitates, the names are wrong.

Tier boundaries drawn on engineering convenience rather than buyer logic are the most expensive quiet mistake in SaaS. The deeper mechanics live in SaaS packaging and tiering, but the product marketing job is narrow: make sure the packaging matches how buyers describe their own maturity.

Building the launch calendar backwards from enablement

Launches fail on enablement timing far more often than on asset volume. Work backwards: external launch date, then internal launch two weeks earlier, then enablement assets ready two weeks before that.

The full mechanics of the launch week itself sit in the SaaS product launch strategy playbook, and the distribution mechanics in the launch week social distribution checklist. Don’t rebuild either from scratch.

Enablement delivery that reps actually use

Four assets, delivered in the tools reps already live in, with usage tracked. That’s the whole standard.

The slide goes in the master deck, not in a new deck. The demo script lives next to the demo environment. Objection handling goes wherever your reps search during a call, which is usually the CRM sidebar or an enablement tool, never a shared drive folder.

Then check usage after thirty days. If fewer than 40% of reps have opened the kit, the problem is the asset or the placement, and asking sales leadership to mandate it will not fix either. Rewrite for length first. Most rep facing docs are three times longer than they need to be.

When you’re working alongside a sales team with its own motions, coordinating this properly needs a combined sales and marketing strategy rather than two calendars that collide in the same week.

Adoption campaigns for features already shipped

This is the stage almost everyone skips, and it’s where the value is. A shipped feature nobody uses is a support cost, a roadmap distraction and a renewal risk in one.

Pick the segment with the clearest job to be done for that feature. Not everyone. At $20K ACV that’s often a few hundred accounts, which is small enough to run a personalised in app message plus a two email sequence plus twenty outbound notes from customer success.

Measure feature adoption as percentage of eligible accounts using the feature at least twice in thirty days. Once is curiosity. Twice is a habit forming. Amplitude, Mixpanel or PostHog will give you this in an afternoon if your events are instrumented.

Promotion tactics that pull adoption and acquisition together are covered in how to promote a SaaS product. If your product is AI native, the adoption problem shifts because buyers need proof of reliability before habit, and marketing an AI native SaaS product handles that difference.

The 90 day plan, week by week

Here’s the plan for a mid market product at roughly $20K ACV, a sales assisted motion, one product marketer with a shared designer.

At $20K ACV with a 60 day sales cycle, don’t expect closed won attribution from this launch until roughly week 22. Report influenced pipeline at day 30 and say plainly that revenue reads later. Teams that promise closed revenue at day 30 end up explaining a shortfall that was never real.

The monthly dashboard worth defending

Three numbers on one page, reviewed monthly, with a trailing twelve month line on each.

Influenced pipeline: opportunities that touched a product marketing asset or campaign, tracked as a share of total new pipeline. Imperfect, and say so out loud in the meeting. Feature adoption: eligible accounts using the launched feature twice in thirty days. Rep asset usage: percentage of reps who opened the current kit in the last thirty days.

Add a fourth if you can get it clean: message consistency, sampled by pulling ten call recordings a month and counting how many reps used the narrative’s core line. That one is manual and worth the hour.

If you’re also setting spend targets alongside these, the demand generation plan template gives you the budget side of the same picture.

What to do next

Pick your product. Book eight buyer interviews this week before you write a single line of copy, because everything downstream is built from those transcripts. Then work through the 90 day table above in order and resist the pull to jump to launch assets, which is where the pressure always comes from.

If the launch date is already fixed and you have six weeks instead of thirteen, cut asset production and the adoption push, not research. Fewer, better assets built on real buyer language will outperform a full kit written from guesses every time.