Skip to the good stuff
Sign up for Alley-Oop

The product launch playbook for 2027: launch day is the easy part

Season 1 of Product Marketing for You was 19 conversations with people who've shipped more launches than they can count. I asked them how they actually do it. This is the whole playbook, in the order you'll need it, with every episode one click away. Steal all of it.

Start here

What is a product launch plan?

The short versionA product launch plan is your go-to-market schedule. It runs next to the dev roadmap, starts with a goal and a tier, and keeps going well after launch day with more launches and a 30, 60 and 90-day review.

Here's what surprised me after 19 interviews: almost nobody wanted to talk about launch day. Daria Love wanted to talk about the months before it. April Dunford wanted to talk about the year after it. Haidee Griswold won't grade a launch until day 30. Launch day matters. It just matters a lot less than we stress about.

So the playbook runs in five phases. Here's who covers what.

The five phases of a product launch and the guests who cover each Nineteen guests across five phases. Plan: Tamara Grominsky on goals, Daria Love on getting involved early and tiering, and Adam O'Brien on AI-speed releases. Position: Drew Giovannoli on buyer research, April Dunford on positioning, Talya Heller on templates, Emily Pick on messaging, Shannon Kearns on message testing. Build: Dmitry Shamis on the creative brief, Natalie Marcotullio on interactive demos, Allison Rona and Michele Nieberding on enablement. Launch: Devin Reed on exec posts, April Dunford on rolling thunder. Learn: Haidee Griswold on the 30-60-90 day review. // THE LAUNCH SEASON FIVE PHASES. NINETEEN GUESTS. ONE HOST. 1 · PLAN2 · POSITION3 · BUILD4 · LAUNCH5 · LEARN Tamara · goalsDaria · get in earlyDaria · tiersAdam · AI releases Drew · buyer researchApril · positioningTalya · templatesEmily · message mapAshley · repeatableShannon · testing Sushmita · commsDmitry · creativeNatalie · demosAllison · beliefMichele · AI enablementSharon · AI trust Suz · launch dayDevin · exec postsApril · rolling thunder Haidee · 30-60-90Zach · flopsCollin · relaunch
Figure 1. The playbook at a glance. Scale each phase down for smaller launches. Keep the order.

Phase onePlan it

Plan · Goals · Tamara Grominsky

How do you set goals for a product launch?

The short versionTie the launch to one company goal, find out why it's being built, put numbers on your assumptions, then have people who know more check your math.

Most PMMs set launch goals by picking a number out of a hat and hoping nobody holds them to it. Tamara Grominsky, founder of PMM Camp and a former product marketing leader at Unbounce and Kajabi, says it's way more logical than it feels. Her four steps:

“Part of it is art and science, but it's much more logical than you might think.”

Tamara GrominskyTamara GrominskyFounder, PMM Camp
Connect on LinkedIn
  1. Start with the company goal. New market? Churn? Expansion? If you chase a goal the company doesn't care about, hitting your number won't get you any credit.
  2. Ask why it's being built. If the PM says "it helps everyone across the whole lifecycle," keep digging. Something specific put it on the roadmap. Come in curious, not accusatory, so you and the PM end up giving the same answer.
  3. Do the math. At Unbounce, Tamara sized a launch like a mini market: how many customers in each segment, how many already A/B test, how many might upgrade. If all that math ends in four upgrades, you know what kind of launch you're running.
  4. Get your math checked. Sales knows close rates. Finance knows how similar launches went. That's how a SWAG (a "sophisticated wild-ass guess," per her old FreshBooks manager) becomes a goal you can defend.

Her homework: build a KPI tree

Map how your business makes money. Start with acquisition, conversion, engagement, retention and expansion, then branch each one until you hit something a PMM can move. Nobody else has to see it. Next launch, you'll have a list of options to pick from instead of a blank page.

How a KPI tree works A company goal at the top branches into five level-one metrics: acquisition, conversion, engagement, retention and expansion. Each keeps branching down into smaller drivers until you reach levers a product marketer can move. A launch goal should sit on one of those branches. COMPANY GOAL AcquisitionConversionEngagementRetentionExpansion Keep branching until you hit a lever a PMMcan move. Your launch goal lives there.
Figure 2. Tamara's KPI tree, simplified. Map yours one level deeper than feels necessary.

// Run it back

How to Build Your Launch Goals From Scratch in 4 Steps

with Tamara Grominsky · 18 min

Plan · Getting in early · Daria Love

When should product marketing get involved in a launch?

The short versionWhen it shows up on the roadmap. If you're not in roadmap planning yet, ask to be. Then write down how PM and PMM work together and track launches on a go-to-market roadmap.

A goal doesn't help much if you hear about the feature two weeks before it ships. You know the move: product finishes, walks over and says "do your marketing magic." Daria Love has built launch processes from scratch at multiple companies, and she's done with that move.

“Start simple. First you need to know what you actually want to automate.”

Daria LoveDaria LoveDirector of Brand and Product Marketing
Connect on LinkedIn
  1. Get in the roadmap meeting. Ask to sit in. You'll hear what's coming and earn trust with product.
  2. Write a partnership agreement. Who owns what, the timelines, and what PMM needs from product before launch work starts. Startups will tell you writing it down isn't their style. Write it down anyway. Nobody remembers what got agreed in a Slack huddle.
  3. Run a second roadmap. The dev roadmap says when it's production-ready. Your go-to-market roadmap says when it's market-ready: enablement done, customers told, campaign live.
  4. Start in a spreadsheet. When every team lives in a different tool, a shared sheet is the fastest way to see how messy things are. Move to launch templates in a project tool later. Daria's team uses Asana synced with engineering's Jira.

Plan · Tiers · Daria Love

How do you tier a product launch?

The short versionTier 1 is new to the market and can change your positioning. Tier 2 is something the market expects or competitors already have. Tier 3 is a nice update. Agree on the definitions with product.

Once you're in the room, the next fight is "this is huge" versus "this is a bug fix with a press release." Daria's first tiering meeting involved weighted scores and revenue math, and she thought it was overcomplicated. Now she mostly goes on gut after reading the product brief. Until your gut gets there, use this:

Launch tiers, Daria Love's version
TierWhat qualifies
Tier 1Innovative, not in the market yet, can change how customers see you
Tier 2Something the market wants but already expects, or competitors already offer
Tier 3A nice-to-have feature, or an update to one you already have

Two rules keep tiering from turning into a turf war. First, write the tiers into a launch SOP with product, so the call comes from a shared rule and not PMM's opinion. Second, treat each tier's template as a starting point. If a tier 2 release lands the week of a big conference, announce it there.

Plan · AI-speed releases · Adam O'Brien

How do PMMs keep up when AI speeds up product releases?

The short versionDon't launch everything. See releases before they ship, score each one on customer impact and pipeline impact, bundle the small ones into a story, and remember a release isn't a launch.

Season 1 didn't have a full episode on this one, so it's on me, with backup from a few guests. Daria called it: AI is shrinking the notice PMMs get. Product used to ship a few things a quarter. Now something ships most weeks. Treat every release like a launch and you'll be cooked by February, and your customers will stop reading your emails.

“Every launch includes releases, but every release doesn't earn a launch.”

Sushmita BandaSushmita BandaSr. PMM, KEV Group
Connect on LinkedIn
  1. See it before it ships. Daria's roadmap seat matters more now. Ask PMs to flag work the day it's prototyped, not the day it merges, and give it a "coming soon" row on your go-to-market roadmap.
  2. Score it on two things. Customer impact: does it change how people work, and do they have to do anything? Pipeline impact: does it open deals, beat a competitor or unlock a segment? Where it lands decides the play.
  3. Bundle the small stuff. Group releases by the problem they solve and launch the theme. "Six updates that cut onboarding in half" beats six release notes nobody opens. This is also where packaging gets decided: which plan it belongs in, and whether it's an add-on.
  4. Don't talk about it yet. April's rule, sped up. Ship it quietly, let customers and reps use it, and save the story for when there's a story. Collin Mayjack's team at Sybill shipped every week and only made noise about the releases that pushed their main story.
  5. Keep a release log. One row per release with its score. It feeds your rolling thunder calendar and your 30-60-90 reviews.
The release impact matrix: customer impact versus pipeline impact A two-by-two grid. High customer impact and high pipeline impact: launch it, as a tier 1 story with rolling thunder. High customer impact, low pipeline impact: tell customers in-app, through CS and email, and track adoption. Low customer impact, high pipeline impact: arm sales with a talk track and competitive card, and bundle it into the pitch. Low on both: release notes, then bundle it into a theme later. // THE RELEASE IMPACT MATRIX TELL CUSTOMERSIn-app, CS, email.Track adoption, not pipeline. LAUNCH ITTier 1 story, full playbook.Plan the rolling thunder. RELEASE NOTESChangelog now.Bundle into a theme later. ARM SALESTalk track, competitive card.Fold it into the pitch. CUSTOMER IMPACT PIPELINE IMPACT
Figure 3. Score every release, then run the play for its box. Most weekly releases land bottom-left, and that's fine.

Cracked this already? I want to hear how. Shoot me a DM on LinkedIn.

Phase twoPosition it

Position · Buyer research · Drew Giovannoli

How do you do buyer research before a launch?

The short versionGive it 20 days. Interview recent closed-won customers first, recruit like an SDR, pay people for their time, and share what you learn as short video clips.

You've got a goal and a tier. Before you write a word of positioning, go talk to buyers. Drew Giovannoli, founder of Buried Wins, has run more than 1,000 buyer interviews, and he thinks research is where a PMM earns a seat at the table. Here's his 20-day plan:

“Your credibility comes from knowing the market, the customer, and the competitors better than anyone else.”

Drew GiovannoliDrew GiovannoliFounder, Buried Wins
Connect on LinkedIn
  1. Start with closed-won deals. Everybody wants to know why you lost. But you'd rather find 100 more buyers like the ones who said yes, and they're far more likely to take the call.
  2. Pay people properly. A $10 Starbucks card is why "research didn't work." Drew suggests $85–$150 for 30 minutes, a few follow-up emails, a LinkedIn message, and a note from an exec when it counts.
  3. Automate closed-lost outreach. Set up a CRM workflow that emails lost deals about 10 days after close with your calendar link. Two or three calls a week adds up fast.
  4. Ask your partners first. Before any interviews, ask the heads of sales, marketing and product what they want to know. Research that answers their questions gets used.
  5. Share clips, not a 30-page doc. A 30-second clip of a lost buyer explaining why the competitor looked better moves an exec team faster than any slide. Find the theme first, then pick the clip that shows it.

// The slightly diabolical move

Find people who sold for or marketed your competitor in the past year on LinkedIn. Message them directly, say why you're reaching out, and offer to pay for their time. Hearing how they pitched against you is the interview that makes execs' jaws drop.

Position · Positioning · April Dunford

How do you position a new product before launch?

The short versionFor a brand-new product, your positioning is a thesis: an educated guess. Let real customers buy it first, check where you were wrong, then do the big launch.

Research gives you a starting point. April Dunford, author of Obviously Awesome, explained why it's still only a starting point. Your positioning says who you compete with, what makes you different and who'll love you. Then the product ships and buyers compare you to a competitor you never thought about. Six months later you're rewriting the website and everyone's sad.

Her fix is to separate the software release from the marketing launch:

“Our thesis is generally incorrect because it was based on the best information we had at the time.”

April DunfordApril DunfordAuthor, Obviously Awesome
Connect on LinkedIn
  1. Release to a limited first wave. Sales can sell it and customers can use it. Marketing stays quiet.
  2. Check the thesis. Did the customers you expected buy it? Did they compare you to the competitors you expected? Was the value what you thought?
  3. Then do the big launch with positioning that's been tested. Call the first wave a soft launch if that keeps your CEO calm. Just put it on the calendar from day one so nobody thinks you're stalling.
  4. Check again every six months. Competitors, differentiators and customers all shift. Building your positioning with a repeatable process makes it easy to spot which one moved.

We'll come back to April in phase four. She also has the best answer I've heard for what happens after launch day.

Position · Templates · Talya Heller

Should you use templates for launch positioning and messaging?

The short versionYes, as a starting point. Then pressure-test it against your own business before anyone else sees it.

Before you open a positioning template, hear from Talya Heller, B2B sales deck strategist and founder of Down to a T. We've all found the perfect template, followed it to the letter, and gotten back "...okay, but what does this get us?" Talya stopped calling them frameworks and started calling them "thought processes": something you shape to fit, not a recipe.

“The odds are if your BS meter goes off, then someone else's will too.”

Talya HellerTalya HellerFounder, Down to a T
Connect on LinkedIn

Her test before you share anything: what does this miss for our business? What will the sales lead ask? What will engineering push back on? Trust that gut, and apply it to AI critiques too. Claude saying it's good doesn't make it good.

Position · Messaging · Emily Pick

How do you keep launch messaging from drifting?

The short versionKeep one master message map, from company message down to products, and make every tier 1 launch fit inside it. Write it so AI tools can use it, because that's how your team will read it.

Positioning gives you the story. Messaging is how you keep 40 people telling it the same way. Emily Pick took Docebo through a rebrand, an acquisition and a full repositioning in about six months, using a message map she first built at Clari. She calls the problem messaging drift: a new product line, a new persona, a stakeholder rewriting copy on their own, and suddenly everyone's telling a slightly different story.

“Your customers, your internal teams, and now we're starting to write for machines.”

Emily PickEmily PickPMM Leader, Docebo
Connect on LinkedIn
Emily Pick's message map, top to bottom Five stacked layers, widest at the top. Message house: core message, three pillars, proof points, elevator pitch. Platform differentiators. Personas: pains and lead value proposition. Use cases: jobs to be done. Products: playbooks and key capabilities. Everything rolls up to the message house. 1 · MESSAGE HOUSE2 · PLATFORM DIFFERENTIATORS3 · PERSONAS4 · USE CASES5 · PRODUCTS core message · 3 pillars · proof · pitchvalue-led, not product-specificpains · lead value prop · pillarsvalue prop · jobs to be doneplaybooks · key capabilities Everythingrolls up.
Figure 4. The why comes before the how. Link out to full persona docs and product playbooks instead of cramming them in.

Her twist for 2027: she assumes nobody will read the doc line by line. They'll upload it to Claude, ChatGPT or Glean and ask it questions. So she writes it with the context a model needs, and she built a Claude skill that flags drift in new assets. Sometimes the drift check shows the map itself needs updating. (Her PSA: get IT's OK before you put company info into any AI tool.)

// Run it back

Messaging Drift Is Killing Your Launch: Here's How to Fix It

with Emily Pick · template in the show notes

Position · Messaging that travels · Ashley Guerra

How do you make launch messaging stick when you're not in the room?

The short versionWrite it to be repeated, not admired. Keep it simple enough to explain to a friend over a beer, build it with the people who'll retell it, and test it by having your team record themselves pitching it.

Emily's map keeps the story consistent on paper. Ashley Guerra, a PMM at LinkedIn, worries about what happens after it leaves your hands. Sales simplifies it. Marketing stretches it into campaigns. Customers reinterpret it for their own teams. None of that is wrong. It just breaks a message that only works when you tell it.

“Messaging is really the story that gets told when you're not in the room.”

Ashley GuerraAshley GuerraSr. PMM, LinkedIn
Connect on LinkedIn
  1. Pass the beer test. Could you explain what this does to a friend at a bar who knows nothing about your world? If not, it won't survive a sales call either.
  2. Get alignment, not just clarity. Clarity is how well it's said. Alignment is whether everyone agrees to say the same thing. The second one is harder, and it's the job.
  3. Co-build early. Bring in field leaders, individual reps, customers and analysts while it's still a draft. It's slower up front, and much faster than finding the problems after launch.
  4. Record the pitch. Ask everyone on the go-to-market team to record a sub-one-minute pitch of the launch. You'll hear how far one changed word drifts, and sometimes you'll steal a better line.

// Run it back

Your Sales Team Is Breaking Your Message: Here's Why

with Ashley Guerra · 22 min

Position · Message testing · Shannon Kearns

How do you test launch messaging?

The short versionTurn themes from your research into hypotheses, run each as a concept with three or four ad variants in a short paid sprint, and read the results as direction, not proof.

Even a message your team can repeat is still a guess until buyers react to it. Shannon Kearns has run message testing across more than 15 launches. She started when she had to brief a launch with almost no customer insight behind it. It works best for PLG and direct-to-consumer products. B2B teams can adapt it by tracking clicks instead of acquisition cost.

“Leaning really hard into differentiated messages and quickly finding out they don't resonate. To me, that's a really big win.”

Shannon KearnsShannon KearnsPMM Consultant, More Than Said
Connect on LinkedIn
  1. Mine your inputs for repeating themes: reviews, Reddit, G2, sales calls, support tickets, competitor ads.
  2. Write hypotheses about what motivates buyers. Do they react more to losing something or gaining something? What role does trust play? Who do they want to be seen as?
  3. Build concepts. Each gets a lever (emotional, functional or social), the insight behind it and a one-line hypothesis.
  4. Run a sprint with your acquisition team. Three or four ads per concept. One setup Shannon has used: about $500 per concept over five days.
  5. Bring what you learn to everyone. PMM owns the test so web, sales and product all learn from it.
Customer acquisition cost ranged from about 80 dollars to over 200 dollars across message concepts From one of Shannon Kearns's message tests for a luxury direct-to-consumer product: across four concepts, customer acquisition cost ranged from about 80 dollars for the best concept to more than 200 dollars for the luxury-language concept, against a typical average of about 120 dollars. // ONE TEST, FOUR CONCEPTS SAME PRODUCT. CAC FROM ~$80 TO $200+. $50$100$150$200$250 Best concept ≈ $80Typical avg ≈ $120Luxury framing $200+ CAC by concept · luxury DTC client · 4 concepts, 2 ads each · source: Shannon Kearns
Figure 5. The worst performer taught the most. Luxury language flopped, which raised a real question: does this audience even want luxury positioning?

// Run it back

Stop Guessing on Launch Day: Meet Your New Message Testing Framework

with Shannon Kearns · 24 min · template in the show notes

Phase threeBuild it

Build · Internal comms · Sushmita Banda

How do you run internal communications for a product launch?

The short versionTreat internal teams like customers with different needs, bring each one in at the right time, and work backward from the release date so nobody gets surprised or buried.

Positioning's set. Before anything gets built, line up the people who'll carry it. Sushmita Banda, a Sr. PMM at KEV Group running launches across a multi-product portfolio, says internal comms is usually the least structured part of a launch. Bring a team in too late and they're surprised. Too early and they tune out. Her fix is to treat each team like a customer: product wants it represented well, sales wants to know how to sell it, CS wants to know what customers will ask, and leadership wants outcomes.

“A product launch is not a moment. It's a timeline.”

Sushmita BandaSushmita BandaSr. PMM, KEV Group
Connect on LinkedIn
Sushmita Banda's internal comms timeline, worked backward from launch About three months out, release readiness calls start with every team except sales. Six to eight weeks out, align with product and development. Three to four weeks out, finish messaging and the launch brief and brief marketing. Two weeks out, train sales and customer success. On launch day, give the whole organization visibility and credit. After launch, gather feedback. // WORK BACKWARD FROM THE RELEASE DATE ~3 MONTHS6–8 WEEKS3–4 WEEKS2 WEEKSLAUNCHAFTER Release readinesscalls, all teamsbut sales Product + dev:what changes forcustomers Messaging, launchbrief, briefmarketing Train salesand CS Org-wide visibility,credit the team Feedback: is thestory working?
Figure 6. Sushmita's best-case timeline. Map the people, map the dates, then copy it for every launch.

Two habits keep it repeatable. Give launches a home: a recurring release readiness call, plus one channel (Slack or Teams) where sales always finds what's coming. And scale it by tier: tier 1 gets the full timeline, tier 3 bug fixes may only need release notes. In her words, there's no perfection, just repeatability.

Build · Creative · Dmitry Shamis

What goes in a product launch creative brief?

The short versionAudience, goals, key messages, budget, timeline and what's driving it, inspiration, and the assets you can't skip. Bring creative in early and tie every note to the strategy.

Internal teams are lined up. Now the message needs to look like something. Dmitry Shamis (ex-HubSpot, co-owner of creative agency OhSNAP!) put it bluntly: brand can't make a bad product good. What creative can do depends on how much of the why you give them. Here's his brief:

“Treat them like partners, not service providers, because they want the launch to be successful too.”

Dmitry ShamisDmitry ShamisCo-owner, OhSNAP!
Connect on LinkedIn
The launch creative brief, per Dmitry Shamis
FieldWhat goes in it
AudienceWho exactly this is for
GoalsWhat the launch has to do, and what you want people to do next
Key messagesStraight from your message map
BudgetWhat you can spend
Timeline + whyThe date and what's driving it. What breaks if it's April 15 instead of April 1?
InspirationReal references. "Make it pop" doesn't count.
Must-havesProduct page, ads, templates, email headers

The space between "must-haves" and the deadline is where the good ideas show up. When the work comes back:

  • Decide who gets a vote. Use a RACI. People put on the spot for feedback tend to invent some.
  • Skip "I don't like it." Say "This doesn't connect to the strategy," or ask "Walk me through this choice." Either it makes sense, or the designer admits they were flexing.
  • Warn them early. If the CEO is talking about changes, tell creative now, not three days later.
  • No creative team? Set up a Canva brand kit properly (colors, type, photo style) before you make a single asset.

Build · Interactive demo · Natalie Marcotullio

How do you build an interactive demo for a launch?

The short versionPick where it'll live, pick the two or three moments that get buyers excited, capture first, script second, and keep it to 5–13 steps.

The fastest way to show a new feature is to let people click through it. Natalie Marcotullio of Navattic personally clicks through the top 1% of demos on the platform for their annual report. Her first piece of advice is what not to do: open the product, click around and record a tutorial.

“The biggest mistake we see with demos is ones that are way too long.”

Natalie MarcotullioNatalie MarcotullioHead of Growth and Product Marketing
Connect on LinkedIn
  1. Decide where it'll live. Homepage, launch page, customer email or internal training changes the length, the depth and how you capture it. 86% of the top 1% use full HTML captures, partly because you can edit the data so the dashboards look real.
  2. Pick two or three moments that get buyers excited, from sales calls or feature requests. Confirm them with sales before you build.
  3. Capture first, write the script second.
  4. Keep it to 5–13 steps. That number hasn't changed in four years of reports.
  5. Put a CTA halfway through, not only at the end. Lots of people are sold by step four.
Interactive demo length: the 5 to 13 step sweet spot Fewer than 5 steps is too short. 5 to 13 steps, about one to two minutes, is the sweet spot. Past 13 to 15 steps, drop-off rises. Split longer stories into 3 to 6 chapters, or a demo center beyond 6. // HOW MANY STEPS? 5–13 STEPS PER FLOW. THEN CHAPTER IT. TOO SHORTSWEET SPOT · 1–2 MINRISKYDROP-OFF 15131520+ Bigger launch? Split it into 3–6 chapters. More than 6? Build a demo center.
Figure 7. Demo length, per Natalie and Navattic's State of the Interactive Product Demo 2026 report.

Gate it or not? Usually not. 65–70% of the top demos are ungated. If you have to gate, ask for an email and nothing else. You're not a bad person for gating. Just make it worth the email.

Build · Enablement · Allison Rona & Michele Nieberding

How do you enable sales for a launch so reps actually use it?

The short versionReps sell what they believe. Show them the product winning in front of real buyers, then give them an AI assistant trained on your best material so they stop Slacking you 10 minutes before a call.

Great assets don't matter if reps don't use them. Two guests looked at that from opposite sides.

Allison Rona: belief beats slides

Allison, a former seller turned PMM, had an AE who was scared to demo one complicated product. At a conference, he watched her demo it live and a prospect said it might solve all her problems. Allison asked to record her saying it again and walked away with a 10-second clip. That AE built every conversation around the product and doubled his quota. Spread across the team, it led to some of the best months in company history.

Her advice for tomorrow: get close to your sellers, SEs and sales leaders. They trust PMM when PMM is in it with them.

“When they believe it, they go out and they sell it, and they sell the crap out of it.”

Allison RonaAllison RonaDirector of Product Marketing, PetDesk
Connect on LinkedIn

Michele Nieberding: get out of the enablement black hole

Michele's fix for the rep asking for the deck you've linked five times is to let reps help themselves. She builds a NotebookLM notebook for each key product and each key competitor. The trick is what goes in. Generic inputs get generic answers, and with no context an AI can end up quoting your competitor's comparison page back at you. Feed it what public AI doesn't have:

  • Customer call transcripts, ideally connected straight from Gong or Chorus.
  • A "dirty laundry list" of where competitors break.
  • An if-then objection table: if they say X, pivot to this value and this pain.
  • Specific case studies: industry, problem, metric, quote. Not "improved efficiency."
  • Who not to sell to, plus a glossary of your team's internal nicknames for features.

// Start Monday: Michele's 3×3

Ask your sales leader for the 3 questions reps freeze on. Write 3-sentence answers. Load them into a custom GPT or notebook, tell it to stay inside your material, and give it to your 3 best reps for a week. Then bring a real customer onto your next enablement call. Michele says nothing beats it.

Ready to go past the pilot? The AI sales enablement playbook walks through the full build.

“As cool as AI is, nothing beats that human-to-human conversation.”

Michele NieberdingMichele NieberdingHead of Product Marketing
Connect on LinkedIn

Build · AI launches in regulated industries · Sharon Liu

How do you launch AI products to buyers whose job is to say no?

The short versionTreat it as education, not just a launch. Win your own sales team first, put your experts in the room with theirs, and earn trust with proof, controls and honest talk about what it takes to run.

Enablement gets harder when the product is AI and the buyer is a bank, a hospital or an insurer. Sharon Liu markets agentic AI into regulated industries, where buying committees now include AI governance boards. She says the launch checklist still matters, but most of the work is education: first for yourself, then your company, then your customers.

“In a sales-led motion for upmarket enterprise, sales is your first customer.”

Sharon LiuSharon LiuPositioning & GTM Strategy Lead
Connect on LinkedIn
  1. Form a launch steering group for big bets: every go-to-market leader plus the person who actually executes for them. Open with the founders on why this, why now. Then have each team say what success looks like and what they need, like a working demo environment for SEs.
  2. Don't wait for 100% agreement. Surface the concerns internally first. If something will slip, decide together whether to push the launch or stagger it.
  3. Match experts to experts. If their CISO joins the call, bring yours. If their architect wants to see how data flows, bring an engineer. Your talking points won't hold up with specialists.
  4. Answer the trust questions up front. Are you training on their data? How does the AI decide? What controls limit what agents can do? Translate product's technical answers for sales.
  5. Read the signal, not the hype. Pilot customer KPIs beat big ROI claims. For the market, Sharon reads implementation war stories on Reddit, watches funding round sizes and listens for buyers reorganizing their procurement.

Phase fourLaunch it

Launch · Launch day · Suz Koch

How do you run launch day without burning out your team?

The short versionKnow your product cold, run the day from a checklist with an owner and a backup for every task, and protect the team's energy. A great launch that burns everyone out isn't a great launch.

Everything's built. Now it has to go out the door. Suz Koch, a solo PMM at Goody who worked on the very first Amazon Prime Day, thinks launches don't have to suck. The fun part depends on the boring part.

“If I have a killer launch, but the team is just burnt out, that's not a good launch.”

Suz KochSuz KochSolo PMM, Goody
Connect on LinkedIn
  1. Know your product cold. Get in the beta and use the features before they ship. When leadership asks a detailed question, or you have to give product hard feedback, that knowledge is your credibility. Suz once shipped a pricing and packaging change across four SKUs plus ten features at once. You can't keep that straight without it.
  2. Run it from a checklist. Pilots run the same checklist whether they have 30 hours or 30,000, because brains skip steps. Give every task an owner and a backup, and have people check items off live. When something breaks, you can fix it knowing everything else is still on track, and leadership can check status without pinging you.
  3. Be yourself. "Business Suz" never landed. People make better things with people they enjoy working with. Read the room, and dial it up or down.
  4. Steal her creative warm-up. Write the five most obvious ways to say it, then five you could never publish. The good stuff lives in the middle. Share the wild ones with your marketing partners so they feel free to go there too.

Launch · Exec posts · Devin Reed

How do you get executives to post about a launch on LinkedIn?

The short versionMake the launch their first post. Bring a tight brief, write it about 85% done, expect them to tear it up, and judge results over two weeks, not two hours.

The checklist's running. Meanwhile, your CEO has posted on LinkedIn zero times. Devin Reed, founder of The Reeder and former head of content at Gong and Clari, says it's not too late, and a launch makes the easiest first post because it's about the product, not about them. One technical co-founder he worked with got around 30,000 views on his first post. (Don't promise that.)

“There's no such thing as too late. A product launch is probably the perfect jump-off point.”

Devin ReedDevin ReedFounder, The Reeder
Connect on LinkedIn

Bring three things to the content team

  1. Who it's for. "Everybody" is not an audience.
  2. The before and after. Before: forecasting deals in spreadsheets. After: one click. Then show the click. Say "AI workflows" here and Devin may throw his laptop out the window.
  3. What success looks like, and by when.

Stop refreshing the dashboard at 8:05 a.m.

Buyers see the launch, talk about it internally and do some research before they book anything. Devin checks demand at one day, two days, one week and two weeks, compared with a normal week. If you usually get 20 demo requests a week and this week you got 35, figure out what worked and do it again.

And when your exec rips up your draft? Normal. Write it about 85% done, ship it when it's 60–70% of what you pictured, and use their own past posts to match their voice. Do get sign-off on numbers. Devin once published a CEO post about crossing $100M ARR without approval. It was taking off right before it got deleted.

Launch · Rolling thunder · April Dunford

What is a rolling thunder product launch?

The short versionInstead of one big launch, plan a series of them over a year: a pre-launch, the launch, a momentum launch about three months later, a launch for the next big feature, and more after that.

Told you April would be back. At IBM, her launch plans had multiple launches inside them, and once she'd done it that way she never went back. You'll only reach so many people on any one day, so you plan to show up again.

“Once you've done it once you're like, okay, launch day is bullshit.”

April DunfordApril DunfordAuthor, Obviously Awesome
Connect on LinkedIn
Rolling thunder: five launches across twelve months Five waves across a year: pre-launch ("it's coming"), launch at month zero, a momentum launch around month three with customers and metrics, a point-release launch around month six, and a follow-on at months nine to twelve. A dashed line shows a one-day big-bang launch that spikes and drops to nothing. PRE-LAUNCHLAUNCHMOMENTUMPOINT RELEASEFOLLOW-ON "It's coming"Month 0~Month 3~Month 6Month 9–12 analysts, light PRevent, press, customerscustomer list, metrics"finally, it's here"progress, roadmap tease One-day big bang Rolling thunder
Figure 8. The shape is illustrative. The idea: every wave reaches people who missed the last one.

Two things make this a solo PMM's best friend. You don't need every case study ready on day one, so save some customers for the momentum launch. And your execs stop judging a launch on day-one numbers when your sales cycle is three months long. April ran it for her own book: she held the audiobook for a later launch and got angry emails in the meantime. She called the whole thing ridiculous, and it worked.

Rolling thunder covers the public side. Next comes the internal side: figuring out what worked.

// Run it back (watch this one)

Your Next Product Launch Could Use a Little Rolling Thunder

with April Dunford · 30 min · If you know April, you know this one needs to be watched, not just heard.

Phase fiveLearn from it

Learn · Post-launch review · Haidee Griswold

How do you run a post-launch review?

The short versionHold checkpoints at 30, 60 and 90 days. Ask the same three questions every time, and grade how you executed separately from whether the market wanted it.

Haidee Griswold has run launches everywhere from startups to Amazon, and she runs the same review no matter the size. Small teams need it most: with fewer resources, repeating a mistake costs more.

“This is not an evaluation. It's a checkpoint.”

Haidee GriswoldHaidee GriswoldDirector, Technical Product Marketing
Connect on LinkedIn
The 30-60-90 day post-launch review Checkpoints at 30, 60 and 90 days, each asking what went well, what didn't, and what to change now and next launch. A two-column scorecard separates execution, whether the launch went out well, from market fit, whether it was the right product for this market right now. 306090 Product has settled.First real feedback.Same questions,more data.Becomes an inputto the next launch. Every time: What went well? · What didn't? · What do we change now, and next launch? COLUMN A · EXECUTIONCOLUMN B · MARKET FIT Did we ship it well?Assets, enablement, timing.Credit the team's effort here.Right product, market, moment?Reactions, adoption, purchases.Did our hypothesis hold?
Figure 9. A bad launch can hide great product-market fit, and a great launch can hide a fit problem. Keep the columns apart.
  1. Hand out the three questions on day one so people write things down as they happen.
  2. Write your hypothesis before launch. You expected A and got B. Working out why is how you build PMM judgment.
  3. Send the data ahead of time. Nobody should see a bad number for the first time in front of their peers.
  4. Open with "what surprised you?" It's about the process, not the people.
  5. When it flops, narrow down. Her CEO's pet Wi-Fi product missed its sales targets. The 30-day review found a handful of obsessed customers. The team focused on them, figured out what they had in common and expanded from there.

And when the review says it flopped? That's Zach's story, next.

Learn · When it flops · Zach Roberts

What do you do when a product launch flops?

The short versionLook for the signal, not someone to blame. Find the objection you kept hearing, stop treating the launch like your baby, and pay the people you meant to serve to tell you what's actually hard for them.

Haidee's review tells you what happened. Zach Roberts, founder of Scope & Sequence (and my go-to PMM sounding board), lived the version nobody posts about. He put $12K into launching a use-case page offer, ran the campaign, did the research. Thirty days later: no pipeline. The objection he kept hearing was "AI can build that for us."

“If someone tells you something is hard, that's a really positive signal to explore what it is.”

Zach RobertsZach RobertsFounder, Scope & Sequence
Connect on LinkedIn
  1. Find the objection on repeat. "AI can build that" was fair. But most teams saying it had never had a process for the work to begin with, so AI had nothing to follow.
  2. Stop calling your launch your baby. You'd never abandon a baby, and sometimes you need to walk away from a launch. Ask "did it work, yes or no?" and take the emotion out.
  3. Work in reverse. Zach admits he thought like a product manager: heard a signal, built the thing. The second time, he talked first, offering a $50 gift card for four questions about how people do the work today and how they measure it.
  4. Lean into "hard." People kept saying building AI workflows was hard, which is odd for a tool meant to make work easier. That tension pointed him to his next offer.

Learn · Relaunch · Collin Mayjack

How do you relaunch a feature nobody's using?

The short versionTreat it like a launch even though it isn't new. Show people what it can do, put it in front of them every week, cut everything else out of the story, and use your competitor as the contrast.

Sometimes the launch was fine and the problem is that nobody found the feature. Collin Mayjack, then a PMM at Sybill (now Head of Product Marketing at Storylane), joined to find most customers using the product as a call recorder. The feature that let you ask questions across every call, deal and email, Ask Sybill, sat at roughly 10% adoption. Some customers were exporting transcripts into ChatGPT to do what the feature already did. There was no big launch video and no big budget. Here's what he did:

“I'm less of the opinion that you need to get people to go to things that aren't working, and more that you should just pour gas on what's working.”

Collin MayjackCollin MayjackHead of Product Marketing, Storylane
Connect on LinkedIn
  1. Build a use-case library. Ready-to-run prompt templates, filtered by role and use case, that load into the product in one click. It retrains customers to see the product differently, shows off real output, and removes the blank-text-box problem.
  2. Bring email back, weekly. He went from about one email a month to one a week, each about 40 minutes of work. The best performers pair a point of view with a use case they can steal right now. He picks topics from the templates people already use most.
  3. Put blinders on the homepage. The hero talks only about the relaunched feature. Other features are still there, further down.
  4. Only launch what supports the story. The team ships features every week. They made a big deal only of the ones that pushed the main feature further, like Slack access and scheduled prompts.
  5. Use your competitor as the background. The market understood Sybill best in contrast to Gong, so Collin ran his most aggressive competitive campaign yet.

The result: about 70% of customers now use the feature week over week, and Sybill's ARR grew 70% in Q4.

Steal this

The 2027 product launch checklist

Paste this into your launch template. Cut items for tier 2 and 3 launches.

21 itemsDownload PDF ↓

Plan

Position

Build

Launch & learn

Last call

Product launch FAQ

What's the difference between a release and a launch?

A release is when the product is production-ready. A launch is when it's market-ready: positioning validated, sales enabled, customers told. Daria Love tracks the two on separate roadmaps, and April Dunford often schedules the marketing launch months after the release.

Who owns a product launch?

Product owns the release. Product marketing usually owns the launch: the goal, positioning, messaging and getting sales, marketing and CS on the same page. Drew Giovannoli's point is that PMM earns that seat by knowing buyers and competitors better than anyone else in the room.

Can a solo PMM run a full launch?

Yes, if you scale it by tier. Start with a spreadsheet (Daria), do your own interviews over LinkedIn (Drew), use a Canva brand kit if there's no designer (Dmitry), and start enablement with a 3×3 (Michele). Save the full playbook for tier 1.

How are PMMs using AI for product launches?

Emily Pick uses Claude skills to fill in message maps and check for drift. Michele Nieberding builds a NotebookLM notebook per product so reps can help themselves. Shannon Kearns has a Claude skill that turns your research into a first draft of a message test. Talya Heller's warning applies to all of it: check the AI's work.

How long should you keep promoting a launch?

Longer than you think. April Dunford's rolling thunder model runs about a year of follow-up launches, and Haidee Griswold reviews results through day 90.

Keep listening

The whole season, one place

Every guest above has a full episode with show notes, templates and links at Product Marketing for You. Listen on Spotify. Season 1 was sponsored by PMM Camp. Season 2, on PMM relationships, starts January 2027.

Next up after launch day: the AI sales enablement playbook, on building a Marketing Brain your GTM team actually uses.

Technically for solo PMMs

Built for solo PMMs. But nobody's checking ID.

One big problem off your plate, the first Monday of every month. Our ICP is the only PMM in the building. Our TAM is anyone who's ever been asked for "a quick one-pager."

One email a month. Unsubscribe anytime. Free, like good advice from a friend.

// Who wrote this

Adam O'Brien

I've built product marketing from scratch five times across 15+ years, at companies from seed stage to $250M+ ARR. I host Product Marketing for You and write Alley-Oop. Everything above comes from the guests. I asked the questions and took notes.

Say hi on LinkedIn