Productside Webinar

How to Kill Zombie Products

Post Launch Success Series: Part 1

Date:

10/28/2026

Time EST:

1:00 pm
Watch Now

That product nobody wants to own? It died years ago, but it keeps shuffling along on a few loyal customers and engineering hours you’ll never get back. Nobody wants to pull the plug, so it haunts your roadmap quarter after quarter. 

Just in time for Halloween, Productside Principal Consultant Dean Peters with Kenny Kranseler shows you how to make the kill call with data, then bury the product without scaring your customers straight into a competitor’s arms. 

What You’ll Learn:  

  • How to spot a zombie product before it drains another quarter of roadmap capacity 
  • How to put a dollar figure on what it costs to keep the undead on life support 
  • How to build an end-of-life business case that gets leadership and sales on your side 
  • How to run a phased sunset that moves customers to your newer offering, not a competitor’s

Introduction and Housekeeping

Welcome and Speaker Introductions

Kenny Kranseler | 00:00:00 – 00:04:15

Welcome, everybody. Let’s introduce ourselves. All right, I’ll go first. I’m Kenny Kranseler. I’m a principal consultant and trainer here at Productside, and have been for about the last seven years or so. Prior to that, I spent lots of time in lots of product management careers, from junior assistant product manager to chief product officer, at places like Amazon, Microsoft, and a bunch of startups up in the Seattle area, where I usually reside. Dean, you want to introduce yourself?

Dean Peters | 00:04:15 – 00:04:50

Yeah, I’m Dean Peters. I am also a principal consultant and trainer at Productside, on the other side of the country, right outside of Raleigh, North Carolina, in a little town called Apex. I’ve been with Productside for about four and a half years, though it feels longer for Kenny. Before that, about 20 years of product management in a variety of areas, including early renditions of AI, connecting a lot of devices, and before that I was a software engineer, doing all that. So lots of fun doing all that. But enough about us. You’re here to hear about what Productside has to offer.

About Productside

Kenny Kranseler | 00:04:50 – 00:06:15

So, before we dive into our topic, I’m going to give you a little bit of a welcome from Productside. If you’ve been with us on webinars before, it’s great to see you again. If this is your first webinar with Productside, hopefully we’ll get to know you better through the course of this.

Here at Productside, we believe that bad thinking kills more products than bad execution. The teams we meet and the folks we work with are rarely short on effort or speed or agility. What they tend to be short on is the thinking that happens before the work starts. That’s where we should ask key questions:

  • Who is this product for?
  • What problems does it solve?
  • Does it deserve to exist at all? (which we will talk about a little bit in this webinar)

If people haven’t heard of AI, they’re living under a rock, but it has made every product team faster. It hasn’t necessarily made them smarter or more right. That belief is the reason we exist at Productside. We’ve spent more than 20 years or so helping product people of all sorts make decisions that hopefully survive contact with the room they are in, and we tend to measure the outcomes that result from the work they’re doing. Our training, assessments, and advisory services all serve that one idea: your outcomes matter most.

Dean Peters | 00:06:15

That’s so Kenny: ask questions.

Webinar Logistics & Engagement Guidelines

Kenny Kranseler | 00:06:19 – 00:07:20

The first question we get every webinar is, can I watch this video later? And the answer is an absolute yes. By showing up for this webinar, you will get a recording link in your inbox after we wrap up. So you can sit back, enjoy, listen to Dean and me ramble on, and know the recording is coming your way.

Second, while we’re live, please, please, please talk to us. We want this to be a conversation between friends. We want you to be able to interact with us and us to interact with you:

  • You can drop a question in the Q&A as they come to you. You don’t have to wait until we put up that magic Q&A slide at the end of the session
  • There’s a chat available in Zoom. Please chat with each other, compare notes, swap war stories, tell someone about your AI rollout and what sounds familiar. You’ll get as much from the room as you do from us

We are monitoring the Q&A and we’re monitoring chat, so if something is relevant at a certain point in time, I will not hesitate to interrupt. Dean,

Dean Peters | 00:07:20

Yes, we like a very chatty session here. We’d like you to connect to us.

Kenny Kranseler | 00:07:21 – 00:07:58

We would. We really appreciate you being here. We’d love to continue to stay in touch after the session. The best way to do that is to connect with us on LinkedIn. We really are everyone’s partner in product, and we want you to get the invitation to our party. It’s where we put a lot of stuff that is hopefully worth stealing: tips, tricks, ongoing arguments about what makes product management work, and a room full of people wrestling with some of the same problems you might be. Hopefully we deliver value as part of this. You can scan the QR code on the screen or hit the link that we have put into chat.

Real-World End-of-Life Stories

Agenda Overview & Citrix GoToMyPC

Kenny Kranseler | 00:07:58

And without any further ado or delay, I’m going to let Dean dive into today’s topic on…

Dean Peters | 00:08:05 – 00:09:39

Yeah, let’s go to our very spooky agenda here. So we’re going to have:

  • A tale from the crypt from Kenny
  • Five other zombies from the past
  • Your five rules for surviving that zombie product apocalypse
  • A demo with a skills repo we put together, so you can play the home edition

All righty, let’s just dive right in. And Kenny, before I hear your story, I think we’ve all got end-of-life product stories. I had one when I was at Citrix. It was a product called GoToMyPC, where you could remotely control a PC. It was part of their whole GoTo line. Now, by the time I got there, three CEOs in, they realized they were fragmented, and they got rid of the whole GoToMeeting, GoTo whatever, GoToMyPC product line. But for whatever reason, this product stuck around. After several years of it sticking around and just getting bundled with everything, I got a tap on my shoulder saying, “Dean, please bring this to end of life.” And that’s where I learned: how do I get rid of a legacy product without getting rid of the legacy customer?

Now, the short story is that the people we sold the GoTo line off to actually had the exact same codebase. We worked out an arrangement with them where they could get those customers, and then I could tell some people in our organization to quit bundling this to sweeten the pot. But I think you have an even more interesting end-of-life story.

The Tale of the KFC Spork

Kenny Kranseler | 00:09:39 – 00:11:40

Yeah, I had one that didn’t quite work out. I told you I was a product manager for a long time. My first role as a product manager was at a place called Kentucky Fried Chicken, which most of you have probably heard of. At the time when I joined Kentucky Fried Chicken, they were owned by Pepsi Cola. Pepsi Cola had recently purchased Kentucky Fried Chicken and had a concern that it was kind of getting dated, not up with trends. So they wanted to get rid of everything that made the brand feel old and dated. They didn’t quite understand the concept of retro at that point. So they did things like change the name from Kentucky Fried Chicken to KFC, officially. They changed all of our branding, our logos. Our restaurants got updated to meet new views, and they didn’t look like the Colonel’s image of Kentucky Fried Chicken.

One of the things that they also thought was part of that dated image was a thing called a spork. Most of you have probably used a spork. It’s this wonderful thing that’s a blend between a spoon and a fork, and it shows up in every utensil kit coming out from Kentucky Fried Chicken. The folks at PepsiCo said to me, “Please kill this thing. We think it’s part of their outdated image.” I went through the process. I talked to some operators. I talked to some customers. I talked to some franchisees. It turns out the franchisees and the operators loved it, because they had much cheaper utensil packets as a result: they didn’t have to put both a spoon and a fork into those packets. It made them a lot cheaper. And it turns out customers really liked it. They liked the concept that they could jump in, cut up their chicken, and then eat their potato salad or coleslaw and mashed potatoes, all with the same thing. Yeah, it was ugly. Yeah, it was kind of traditional, but it was kind of retro. So the good news is I pushed back on the team, and I was able to get us to not kill the spork, fortunately. Which I think is a wonderful thing, because if anybody has ever watched Toy Story 4, if I had been successful in killing the spork, Sporky never would have been alive.

Dean Peters | 00:11:40 – 00:11:55

Right? I mean, we have you to thank for that. Hey, listen. When I was raising my daughter, she loved a spork. When I was a camp counselor, sporks. So, yeah, I’m glad that you stood in the gap there and said, “No, the spork must not die.”

Kenny Kranseler | 00:11:55

So, not all old products are ready for death.

Famous Zombie Products & Audience Poll

Historical Zombie Examples: Skype, Revolv, IE & Kodak

Dean Peters | 00:11:56 – 00:14:05

And so, that gets us into our five zombies and five broken rules here. The first one is to recognize what you’re paying for. So, for example, is anyone counting the cost of your products? Is anyone counting whether the orange is still worth the squeeze? Is it still making revenue? Like with the spork, right? Are we asking the question: is it still bringing delight? Is it still bringing some differentiation? I like your story: they nearly killed the spork until they started looking at it and saw that what it was costing us is certainly superseded by the benefit and the delight it brings.

Here are the zombie examples:

  • Skype: It hung around forever. It was basically a vampire on the resources and a distraction from MS Teams until finally someone said, “No, no, kiss it goodbye.”
  • Revolv Hub: Oh, I like this one. Nest took this product over. They decided to brick the lifetime users, basically take the device and make it unusable, which then led to an FTC investigation, which means lawyers and all sorts of fun refunds that followed. Not really the economical move that they thought it was going to be.
  • Internet Explorer: We talk about zombies and killing them as a team sport. And of course, we all know the pain of the Internet Exploder, Internet Explorer browser that never died, especially for those of us in the emerging age of HTML5 and dynamic CSS. The end of sale was declared, but the sales kept bundling it. It just kept going on. It took forever to kill that zombie.
  • Kodak: We talk about “free the living.” What that basically means is that if you’ve got products in your product line that are cannibalizing each other, in either direction, you need to understand what they are. Kodak, I think, is the most egregious, where the old product just kept them from moving into digital photography.

Kenny Kranseler | 00:14:05

Interesting, because they invented the camera back in, was it ’75? Dean, I have a story about that, too.

Dean Peters | 00:14:10

Yeah, go for it. I want to hear it.

Kenny Kranseler | 00:14:15 – 00:14:55

When I left KFC, I was looking for a new role, and I wound up at Microsoft. But I was actually interviewing with Kodak at the time, and they wanted to hire me as their product manager for a new product they were introducing called APS. I don’t know if anybody’s ever heard of APS, but it was Kodak’s last vestige to hold on to their film revenue and their processing revenue. They created a digital film, but that still required processing at their little Kodak shops. It was sort of their new way of doing things, and they were doing everything they could to avoid going to digital.

Dean Peters | 00:14:55 – 00:15:09

Yeah. So you’ve got issues there, and we just want to give you five different zombie examples we’re going to be talking about here.

Poll 1: Which Zombie Nightmare Are You Facing?

Dean Peters & Kenny Kranseler | 00:15:09 – 00:16:14

I think what we’d like to do first of all is a poll. So if we can do a quick poll: which zombie nightmare are you facing the most?

  • You fund the product by default, sort of like Skype for all that time
  • The loudest voice wins: the person who wanted to kill the spork even though they had no data
  • We lose customers when we retire the product
  • Our exit falls apart in handoffs
  • It starves the next big bet

Give everybody a chance to answer their polls. We’ll see how many pop in. Looks like a couple of them are becoming pretty big. I think we can end the poll now and see what we got from this one. Let’s end the poll and share the results here.

It looks like “the loudest voice wins” and “we lose customers when we retire it.” I think those are the big ones, and “it starves the next big bet.”

No surprise there, and I think you and I have lived through that more than once.

Defining Zombie Products & The Product Life Cycle

What Is (and Is Not) a Zombie Product?

Dean Peters | 00:16:14 – 00:17:49

So with that in mind, let’s talk about the five rules of survival here. First, let’s define what a zombie product is and is not. Basically, it’s a product that’s stopped creating enough value. In other words, the orange is no longer worth the squeeze. So it’s not something that’s ugly like the spork, which was beloved by all. It could be a product that’s in decline but not necessarily dead yet. It’s what I call “anyone’s fault”: in other words, it’s just been left unattended or unmanaged. It just means somebody’s got to take a look at it. So those are not what I would call qualifications. What I would define is a product that continues to eat capacity, whether it’s engineering, support, compliance, tokens, whatever.

I have to wonder how many feral agentic cats some organizations have running around. They put the agents out last year, and now they’re wondering what’s all this spend happening here. It’s something that nobody’s willing to fund or sponsor anymore. It’s alive because nobody decides on it. Again, nobody’s taking responsibility. No one’s owning it. So again: it’s stopped creating value but keeps consuming resources, and nobody’s there to help make the right decision on it. I don’t know, do you have anything to add to that? I think that’s sort of my…

Kenny Kranseler | 00:17:49

I think you’ve got it.

Understanding the 6-Stage Life Cycle

Dean Peters | 00:17:50 – 00:18:55

So, you can’t talk about zombie products without talking about their relationship to the product life cycle. Now, most product life cycles only have four stages. I’ve included six, because we need to talk about design, and we need to talk about how it gets introduced and how it grows. You’ll notice, Kenny, that I talk about the revenue line and the profit line. The profit line doesn’t go positive right away, because you’re spending to get the product out the door, which is why I showed these earlier phases here. But then notice the zombie here: it has almost the same impact. It’s eating into the profits. So as you look at the product life cycle, and you’ve got a product that’s going through these phases, what you want to guard against is landing in this area here, where all of a sudden the zombie is eating the resources and puts you in a negative P&L situation. So let’s talk about how products get zombie-fied, as I like to call it.

How Products Get Zombie-fied

Dean Peters | 00:18:55 – 00:19:20

There are a few ways products get zombie-fied:

  • Inertia: Nobody… it’s just the product’s there. It’s just chugging along. It’s part of bundles. It’s part of kits. It’s part of packages. It’s just running along, and nobody’s really making any evaluations on it
  • The sunk cost fallacy: We’ve spent so much on this, we absolutely can’t pull the plug on it now
  • Using hope as a strategy

Kenny Kranseler | 00:19:20 – 00:19:30

I ran into a couple of those, where I tried to kill a product, but it turned out the CEO of the company was the product manager who introduced it.

Dean Peters | 00:19:30 – 00:21:10

Yeah, it’s their pet, and I think that goes from sunk cost maybe even into politics. It was their pet.

I can imagine there are some product managers right now, Kenny, who are dealing with what I call the seagull manager: the one that swoops and poops. They’re dealing with some products, maybe some agents, that their bosses vibed over the weekend. They need to pull the plug, but they can’t, either because they’ve sunk cost into it or because it’s a certain executive’s baby. We’ll talk about how to get past that. And of course, there are legacy commitments. This is the one where lawyers get involved and finance gets involved, so it can get a little messy and ugly there as well. Those are about the four different categories, and I would imagine there’s some nuance, or you’ve got some combinations, but let’s talk about what to look for here. The biggest thing is to understand that as your product goes through its life cycle, your strategies need to change, your goalposts, what you measure in terms of success or health metrics. As your product goes from design and introduction, your competition is going to change. How are you going to make those adjustments? Your cost and your margins are going to change. How are you going to accommodate those? Your investments change. So as you start getting to mature is when you should really start asking some questions, like, what’s decline going to look like? That’s what I would suggest looking for. I don’t know if you’ve got anything else you want to add.

Kenny Kranseler | 00:21:10 – 00:21:29

No, it’s a time to start thinking about it, and there are some companies who actually obsolete their products by design. Think about Hewlett-Packard in their printers. They obsolete their products by design, so they’re planning those all through the process to know when and where and how and why they…

Strategic Reasons & Options for Product Retirement

Four Reasons to Retire a Product

Dean Peters | 00:21:29 – 00:23:05

And so that gets us to four good reasons to retire a product:

  • The cost beats the revenue. It just costs more to put the product out. I had a person in a class who worked with agricultural drones. Their earliest drones were beloved by their customers, but they took twice as many chips and cost twice as much to make, so they had to bring them to end of life just because of the cost structure. It wasn’t that there was anything wrong with them. It’s just that they couldn’t afford to do it as a business. It just wasn’t viable or feasible anymore
  • Strategies get moved. With AI coming in, I would imagine a lot of strategies have changed. Like I said, when I was at Citrix, they brought in a new CEO, new strategies, new lines of thinking. The same thing when I was at companies like Aprimo, as we went through a series of mergers and acquisitions on both sides of that: some of the product line gets elevated and some of it is brought to end of life because it’s no longer relevant to where the organization’s moving
  • There’s a replacement. There’s something better. I think we’re seeing a lot of that happening right now. I feel sorry for people who launched products in late 2022 and early 2023. I think they’re probably feeling that pinch
  • The market changes. Sometimes the need is gone

I don’t know if you’ve got any stories along those lines, but…

Kenny Kranseler | 00:23:05 – 00:23:40

No, I mean, there’s one that I can relate to. I spend some time down in Palm Springs, California, and one of the things it’s famous for is a lot of wind down there. So they put in a bunch of wind turbines in the ’80s, ’90s, and 2000s. They put in these big, huge wind turbines, and there were thousands of them. And guess what? Wind turbines got a lot better. So if you look on the hills now, there are a lot fewer wind turbines up there, because they knocked down the big, huge, inefficient ones, now that the newer ones are smaller and a lot more efficient.

Dean Peters | 00:23:40 – 00:25:21

Well, yeah. The cost just to keep them up there. I knew a group that was working with cryogenic labels. These are the labels that go on, for example, when they take a blood test, and sometimes they have to go into a freezer and sometimes they have to go into a furnace, and they need to be able to survive both of those extremes. Every time there’s new technology, for example with the vials they’re going into, or with the polymers that are coming out, or with the autoclaves they’re going into, this company needs to take a step back and ask: is there an issue of compliance? Is there an issue of regulation? Is there an issue of technology that makes us more obsolete? Those are the types of things they need to think about. So there’s a whole variety of things that can help bring a product to end of life. As you’re thinking about it, think about the different factors, internal and external, that are pushing on your product, whether it deserves to be continued or needs to be brought to retirement. And basically, think about what it earns versus what it eats. At the end of the day, as you can see from this iceberg image, we tend not to look at what’s below the waterline as we get very excited. But below the waterline there’s:

  • Support
  • Maintenance
  • Compliance
  • Technical debt

I’ve heard way too many stories of, “We need to upgrade our product,” and engineering saying, “Every time we change something, it breaks. They have so much technical debt that they keep saying we need to start over again.” So there are those conversations as well.

Four Strategic Plays: Retain, Retire, Revive, Replace

Dean Peters | 00:25:21 – 00:25:40

Now, that leads us to four plays, or four moves, we can make:

  • Retain, if the product still fits, which is what you did with the spork, right?
  • Retire: just bring it to end of life. There are all sorts of products that are no longer valid
  • Revive: sometimes products are worth reviving, worth giving a new life, creating a version 2.0, creating the next version of that
  • Replace

Dean Peters & Kenny Kranseler | 00:25:40 – 00:26:31

HP printers, they just kill them.

Yeah, I can never forget the sound of a dot matrix printer, but I would like to. Nobody needs…

You don’t want to wait for that thing to print out.

No, you don’t want to wait for that thing to print out either, right? And so it got replaced. It got replaced by laser printers, remember? It got replaced by inkjet printers.

Think about Arm & Hammer baking soda.

Yeah, right. So I think that happens with physical products as well as software products. So these are four plays you can make.

Timing the Extension and Replacement Moves

Dean Peters | 00:26:31 – 00:29:10

With those four plays, let me get into the two that I think matter most. When you bring a product to end of life with no replacement, or if you’re going to keep the product, those are sort of obvious. But notice where I say we should start thinking about our extension play, on this timeline, on these time horizons. It could be for any variety of reasons:

  • Our organization’s expanding into new segments
  • We’re trying to meet the changing needs of our customers (none of our customers’ needs are the same as they were two or three years ago)
  • Differentiation from our competitors
  • Boosting sales
  • Strengthening brand loyalty

So you really want to start thinking about that as your product gets into a mature state. Yes, if you’re in decline, you need to think about it, too. The best time to plant a tree was 20 years ago; the next best time is now. So if you’re in a product that’s in decline, you need to start asking yourself which of these four plays we’re going to make. And if you’re going to make a total replacement play, you might want to avoid some conflicts with people in finance and even maybe in legal by thinking, while this thing goes from growth to mature, what does the next product replace? For example, I talked about the people making the drones and the people making the labels. They were constantly doing this, because what they didn’t want was that revenue gap. What they didn’t want to happen was for the product to mature and go end of life and then spend time spinning up the replacement. They wanted that replacement out the door already, introduced. So by the time the product was brought to end of life, they were pushing more toward a growth model than an introduction type of play.

Again, it could be for a variety of reasons: suppliers, channel partners, tariffs. Things just get obsolete, or unused. I don’t know about you, but at the companies I worked with, I insisted on analytics instrumentation, and one of the things it let me do, not just on products but on features, was figure out what’s not earning its keep, what’s not being used, and when and how I should bring that to end of life. So it’s not just entire products; it could be feature sets as well. I would imagine you ran into that more than once or twice.

Kenny Kranseler | 00:29:10

We got rid of some capabilities, or we merged them into other groups.

Dean Peters | 00:29:15 – 00:29:38

Oh yeah. I had one company that, for whatever reason, decided to get themselves in the analytics game, even though it wasn’t an analytics company. I think they had all of two users after so many months, and I was like, okay, we’re going to pull the plug on this. But then that gets you into the conversation of what other conversations are needed here.

End-of-Life Execution Strategy & Internal Stakeholders

Losing the Product Without Losing the Customer

Dean Peters | 00:29:38 – 00:30:35

Here’s the rule, and it’s rule number three, probably the most important rule: how do you lose the legacy product without losing the legacy customer? That’s the real key here, and that’s the crux of this conversation. You can:

  • Surprise them
  • Strand them
  • Brick them, like the people at Revolv
  • Or give them a little bit of a heads-up

I would imagine, Kenny, you’ve seen more than once of this. I’ve had products that were bricked. And we’re probably going to start seeing some surprises, plugs pulled, as various AI companies are spending more than they’re making and either get absorbed, merged into other companies, acquired, or just have to shutter their product. I think there are going to be more surprises and more people stranded than I think they would like. But this is not a new problem, folks.

Kenny Kranseler | 00:30:35 – 00:31:01

No, acquisition is a big piece of this. When I acquired a company years and years ago, they had a legacy product that we were not interested in. The brand name was still there. The customers were still there. We had to figure out how we took care of those customers and made sure that they had a path to success, even though we were essentially killing all the old capabilities of that product line.

End-of-Life Lifecycle Worksheet

Dean Peters | 00:31:01 – 00:33:27

And I think that gets us into this worksheet here, because I don’t think a lot of people realize you don’t just send out the email saying, “Hey, in a couple of months, end of life. Good luck.” For example, your product, if you’re going to bring it to end of life, is more than likely not always in a general availability state. It’s gone through the alphas, it’s gone through the betas, it’s gone through the pre-release, and the product has now gone to market, good, bad, or indifferent. Here’s the progression:

  • Announce it internally first. You probably want to announce that internally first and get your ducks in a row before it goes out
  • Send a notice of status change. This goes out to your customers. The first thing you have to do is figure out, if you’ve got a replacement play, what are the old versus new features? You don’t want to leave your channel partners or your distributors in the lurch. That’s a good way to lose them. You lose support from them
  • End of sales. You need to make sure sales sticks to this. When I got to GoToMyPC, they had announced end of sale in, I think, 2016 or 2017, and going into 2019 they were still bundling the product, still bundling it to sweeten deals, which adds some complexity
  • End of expansion. No new features. This is the last time
  • End of (service) life. Just because you bring that product to end of life, you have service contracts, which means finance, legal, and other people get involved as well

I would imagine there are other stages here, but this is pretty much the game plan I ran the few times I had to bring things to end of life. You’ll notice it’s progressive. I know lots of large companies that actually publish what these stages are, and it’s contract language as well, because otherwise you can leave the customer hanging.

Stakeholder Roadmap: Legal, Finance & Teams

Dean Peters | 00:33:27 – 00:33:50

Now, I’ve got the answers up here, but Kenny, would you be surprised? When I teach end of life in Unit 6 of Optimal, I like to ask the question: what’s your first stop? Who do you talk to first? What’s the normal answer you get?

Kenny Kranseler | 00:33:50

Engineering.

Dean Peters | 00:33:51 – 00:35:55

Engineering. I had someone try to get slick: DevOps first. I did have some people say sales first as well. All of those answers, although they’re well intended, will get you into trouble. So you’ll notice I’ve got legal and finance. I don’t care what order you meet those people in. I just want to make sure that those are your first and second stops along the way. When I was at Citrix, legal was just a few desks away. I stopped there first because I had to go to another floor to get to finance. That was the only reason that happened.

One other thing, and this is not just for zombie products, but it’ll help you with zombie products: when you get to a new organization, or even at your current organization, how good is your network outside of the product team? When I went to bring the product to end of life at Citrix, the lawyers knew me by name. They already knew who I was, because I had taken time to make sure the first time they saw me wasn’t me hitting them up for a favor. Having those types of personal networks and office networks is really valuable to the product person, especially if you’ve created strong networks with finance and legal, which I did, and it made the process much easier, because it was complicated. We had bundles, which meant they were part of contracts, which meant they were part of financial projections, which meant there had to be a bunch of readjustments. It took me a couple of weeks just working with them to get the ducks lined up in a row.

Here’s the order I followed:

  • Legal and finance first
  • Then, and only then, sales. In this case, for that product, I talked to the support group first. Don’t leave your people in customer support or customer success hanging. Not a good idea
  • Talk to sales, and talk to partners
  • I also talked to a few key customers somewhere in between
  • Then, and only then, engineering and our DevOps group

I would imagine you went through a similar set of stages.

Kenny Kranseler | 00:35:55 – 00:36:05

Talk to those legal beagles first. Make sure you know what your commitments are before you even think about it.

Dean Peters | 00:36:05 – 00:36:38

Yeah. The last thing you want to be is in a deposition saying, “Why did you pull the rug out from under this client when they had a two-year contract that included this, and you didn’t make it right or offer them a discount?” Those really thorny types of items need to be pulled out. They need to look at the contracts, and both groups, legal and finance, you’d be surprised just how much paperwork end of life can bring them. So you want to give them a heads-up way ahead of time, before you start talking to others.

Evaluating ‘Earns vs. Eats’ & Freeing Capacity

Dean Peters | 00:36:38 – 00:38:30

Now, this is a slightly different cadence here. This is “free the living.” I would suggest that every product manager, at least every quarter, depending on your product, if not every month, check to see if your product is still worth the squeeze. The orange is still worth the squeeze. Is the product still meeting its marks? Because think about it: is it costing you more in people, in budget, in time? Is it costing you opportunity cost? It’s like those people with the agricultural drones. They could have kept building the older drones, but the newer drones had new features on them. They could go longer distances on less fuel or less battery charge. They were a lot more sophisticated. They had a lot more built into them, but they couldn’t exploit that opportunity if they were still building the old units. I had the same thing with a group building IoT units for upstream oil and gas and wind turbines. They had to get rid of these old units because they were eating into their opportunity cost as well.

All righty. So what to do? If you want to start socializing bringing a product to end of life, you don’t want to get beat up by the seagull manager whose baby it is, or the HiPPO who has an opinion about the spork. If you need to say, “No, this product doesn’t deserve to exist anymore,” you need to:

  • Name the freed capacity
  • Give them an idea of the best bets
  • Show leadership the trade-offs as well

I don’t know if you’ve got anything to add to that brief list.

Kenny Kranseler | 00:38:30 – 00:38:40

No, I think you’ve covered it. Make sure that the freed-up work that you set free, you do something with it. It’s not just fill in the hole. Right.

Dean Peters | 00:38:40 – 00:39:06

Right. And I’m talking about not just what to do with it, but what to socialize as well. In other words, bringing your product to end of life isn’t just a technical battle. It is a diplomatic and political battle as well. So you need to be able to speak to the freed capacity, the next best bets, and the trade-offs as well. Just keep that in mind.

Productside Zombie Survival Kit Demo

Introducing the AI Skills Library

Dean Peters | 00:39:06 – 00:41:46

But enough talk about it. It’s like, “Yeah, Dean, okay, we’re sold here. I’ve got some zombie products. How do we do this?” And I’m like, okay, behold the Productside Zombie Survival Kit. Of course, I’m not going to leave you in the lurch. I’m going to give you a skills library. You can run it, and I’ll even show you how. I’m going to pull the slide deck off for a moment, and I’ll put it back on. I’m now in my browser. I am now in the Productside Zombie Survival Kit, which has been shared with you in the comments. You’ll notice it looks like a lot of agent skills libraries. For those of you using Claude Code, Claude Desktop, Codex, Antigravity, or Cursor, you’ll be able to use this skills library. You’ve got a set of skills:

  • One skill that runs the entire loop
  • One that helps you with your product life cycle mapper, to recognize whether you’ve got a zombie or not
  • What type of play you’re going to make
  • What type of communications you’re going to do
  • How you’re going to exploit the capacity transfer

So this is a nice little skills library. Like I said, it has these different skills, and it has the steps and the utilities. I’ll show you really quickly this one skill here. This is the communications plan. It’s pretty much what I showed you here. It’s going to help you figure out who I’m communicating to, and next. By the way, the nice thing about when we build skills is that you have three modes:

  • Expert mode: You just dump the context and say, “Do it.”
  • Learner mode: “Hey, it’s my first time doing end of life. Walk me through this.”
  • Spitball mode: “Hey, let’s walk through a simulation, just so I can see how this operates and get some inspiration.”

All of this is built into these skills. We’ve also got evaluations, so you have different criteria and cases you can run them against. For example, I’ve got some actual live case studies here. There’s my Citrix LogMeIn transfer. There’s the Revolv bricking. I don’t think I have one for the spork, so Kenny, I think I need to create one for that. And I even have a demo you can run. I’m going to go ahead and run that demo and show it to you, because yes, I learned how to take a browser tab and turn it into a console for Claude.

Live Demo: Classic Reports Case Study

Dean Peters | 00:41:46 – 00:46:29

I’m going to go EOL step one, zombie ID, and I’m just going to paste in this case here. This is a fictional case of Classic Reports. Kenny, look: an entire fintech suite written in COBOL in 1994, patched for Y2K, and based off this FFIEC standard that came out back then, under the Clinton administration, which has since been overridden. How many times have you taught a class or dealt with somebody who says, “Hey, I got a mandate from my boss, go ahead and convert all the reports,” and instead of saying, “Hey, can I check to see if these reports are still valid?” what do we do? We go ahead and mindlessly port the reports, even though somebody doesn’t want them, or nobody’s using them. So, for example, what this skill just did is it looked at the case in a few different ways, asking: Is this thing still financially viable? Is it still strategically aligned? Is there a replacement? Is it irrelevant? It asks those questions, and what it does in this mode is give me recommendations on what to do, based on how much data I’ve given it and how much I’ve asked it to search. It’s undetermined for this one, but I already know that the legal standard went away a couple of decades ago. So I’m going to go ahead and figure out which motion we want, so I’m going to go to step two. I’m going to give it this information. I’ll say, “Hey, listen, FFIEC was discontinued in 2006. It was about insider lending restrictions and so forth. It’s no longer there.” So it’s going to take that information, and now it’s going to give me some recommendations that are stack-ranked. Basically it’s saying, “Look, you’ve got a strong reason to retire this. You could do a replacement.” Here are some of your other choices, and it tells me what I want to do. But notice, Kenny, it recommends, but I have to decide.

So again, it gives me all the options, and then if I decide to retire it, say I go to step three. This is just a way, when you learn things, if you do skills here. We are going to retire it, and then I give it the context. I’ll say, “Actually, we’re going to retire it.” It says replace here, but I want to retire it. That’s okay. Let’s see what it does now that I give it conflicting instructions. Maybe it’ll go all HAL 9000 on me. It’ll give me some communications on who to talk to: legal, finance, compliance. So anyway, while that cooks, I just want to let you know that you can use that skills repo to walk your own products through this process. It is not going to tell you what to do. It’s going to give you recommendations along each of these steps, and it’s going to give you information, and you can even say, “Go out and do some more searching as well.” And by the way, who should hear what? Oh, look at that. There we go. Who I need to talk to, and when:

  • Legal and compliance
  • Then finance
  • Then sales
  • Then customer success
  • Then enablement
  • Then licensing
  • DevOps and engineering last

It doesn’t give me any time horizons, so I might ask it for time horizons as well, but again, it’s just giving me recommendations on helping me do end of life. It’s also giving me some ideas of how to know:

  • Whether I’m dealing with the walking dead
  • What call we need to make: replace it, revitalize it, retain it as is, or retire it
  • How to make sure we don’t strand our customers, so that we lose the legacy product without losing the legacy customer
  • That we never go it alone: it’s a team sport
  • That we either bury it or we plant something in its place

Poll 2: What Will You Do With Your Zombies?

Dean Peters & Kenny Kranseler | 00:46:29 – 00:47:20

So I think we need to take a poll, and this is a multiple choice: what are you going to do with the zombies you’re faced with, or the suspected zombies?

You can pick more than one.

You can pick more than one, because, you know, zombies sort of propagate anyway. So if you want to do all of these, you’d select them all. But please do choose. We’ve got some people selecting more than one, and that’s great. We’ll wait another minute or so, a couple of… 30 seconds, to get those votes in. All right, let me close the poll and share the results. The top responses:

  • Naming their zombies
  • Warning the customers early
  • Freeing the capacity

Yes, you can’t go wrong with that. Nicely done, folks.

Upcoming Events, Courses & Q&A

Productside Resources & Training Schedule

Dean Peters & Kenny Kranseler | 00:47:20 – 00:50:15

Kenny, I think we’ve got some goodies, and then we’ll get into some Q&A, won’t we?

We will. We do have questions, a bunch of questions beforehand and afterward, but let’s go ahead and give them some goodies, and then we’re going to answer some really cool questions we got. Here’s what we have for you:

  • Product management skills report: We did a report on product management skills a couple of months ago. If you want that report for all 15 skills that we picked, please grab it, or you can take the skills assessment yourself, which means you can add to next year’s report if you so desire. Click the link in the chat or scan the QR code to get a copy of our wonderful report on what happens to PMs and where they are at this point in time
  • Austin class: We’re going to be live in Austin in early December. One of our co-trainers, Tom Evans, is going to be teaching a class there. We still have a couple of spaces left. If you’re in Texas, or want to spend December in Austin, which is a wonderful place to be (if you’re in Minnesota or Wisconsin, it might be a good option), head on down for three days and get an intense introduction to Optimal Product Management
  • Post Launch Success Series: This is actually the first in a series. We’re talking a lot about what happens once I launch my product, and we have a whole set of webinars over the next month or so on what to do after launch. This is the first one, about how to kill it. On November 11th, we’ll cover pricing and monetization for products once they’re actually out in the market: how do you make sure you’re getting the most value from them? And on November 20th, we’ll talk about postmortems
  • Courses: If you’re not going to be in Austin but still want to learn Optimal Product Management or AI Product Management, we have a whole bunch of courses for you in the coming months, through February, with Dean, myself, and some of our other trainers. We have a few classes on Optimal Product Management and a few on using AI as part of your product management practice, with a little more detail on becoming an expert in AI. Dean’s teaching starting next week, I think, right?

Yeah, next week. Still time to get in, folks. A lot of fun happening there. And if you’re really interested in courses, go to our website.

So now I’m going to stop sharing my screen as soon as I can find the button. We’ve got some questions we need to answer, don’t we?

I think so. Do we have them somewhere?

I think… nope, still sharing. I’m going to stop sharing. There it is. Stop share. Zoom has made this really hard here.

Q&A: Multi-Year Contracts & On-Prem to SaaS Transition

Dean Peters & Kenny Kranseler | 00:50:15 – 00:55:08

Dean Peters:
We got a couple of questions, I think, early on and maybe during the course of this.

Kenny Kranseler:
I’ve got one that just came in while we’re looking for some of the others.

Question:
I did what you asked. Dean and I actually went and talked to the legal guys, and they said, “I’ve got one customer that has a commitment that’s like four years out, and they’re the only last customer that’s going to be there.” What should I do?

Dean Peters:
Just to read that back to you: I’ve got a product we’ve brought to end of life, and I’ve got one customer left.

Kenny Kranseler:
No, there’s a bunch of customers, but one has a contract that’s like four or five years out.

Dean Peters:
All right. I do want to kill it, but I’ve got this one. So, I think the first thing, like I said, is we need to talk to legal and figure out what some options are. I think it’s also not a bad time to have a conversation, maybe one level up, and say, “Listen, there are a couple of different ways we can go with this.” For example:

  • We can discount things, or throw on a subscription to make up for what we’re removing from the package they have
  • We could do some sort of rebate
  • We could offer them something new. Like I said, what we did was take who our competitor was and hand them over: “Look, you get a discount to boot, without having to redo your data”
  • Can we possibly keep it running on the cheap, just for them, until the contract expires?

The biggest answer there is: how do you make it right for them, Kenny? That’s a conversation. It’s going to involve finance. It’s going to include legal. It’s probably going to include your executives in product, and it’s also probably going to bring in the account execs who are dealing with that customer, because you want to know:

  • How big is this client?
  • How big is the opportunity?
  • What does the lifetime value of this customer look like?
  • What did it cost to acquire this customer, and have we made that break even?

All of these options are open to us, but we need to have those conversations first, and we need to have them with the right people, before we sit there and tell the client, “Sorry, it’s end of life.” And before we even announce to the public. You don’t want them calling up saying, “I’ve got a four-year contract with you guys. What the heck?” So before things go out the building, you want to have those scenarios worked out. By the way, AI is really good at running different scenarios. I think that toolkit can run you a few different scenarios, because we’ve given that toolkit several scenarios, saying, “What would be some possible ways I could make it right for the customer without running us into ruin?” So I think those are the conversations that need to happen.

Kenny Kranseler:
Dean, I had this exact problem with one of the products I was working on, where we had this big, huge customer right in the financial industry. They were doing legacy stuff. What I did, as you mentioned, Dean, is I gathered up my account manager, and we went out, with the executives as well, and had a long discussion with that client. Essentially, what we gave them was priority rights on the replacement product. We built a good understanding of the things that they might have lost in moving from the existing product to the replacement product that we had set up, and made sure that we built that functionality into the early versions of the new product, so that we anticipated their needs and built that in and secured them. Because if we’d lost that customer, it probably would have been critical to the business.

Dean Peters:
Kenny, when I was at Aprimo, we had very, very large customers: entire marketing departments for banks and insurance companies. Not small customers, and it was all on-prem. Part of the effort I was part of was taking an on-prem product that had 95 different versions, written over 15 years, a bazillion lines of code, and figuring out how we were going to move this into a sustainable SaaS model, because we were going to have to say no to several things. We got really good at what I called feature hostage negotiations and stakeholder shuttle diplomacy. We had a lot of conversations, but we had to tell our customers, “Look, it’s an existential thing. We can’t just keep doing 95, 96 bespoke different versions. It’s not good for you. It’s not good for us.” And man, they did it, and it didn’t just save the company, it helped the company survive and thrive to what it is today.

Q&A: Relaunching Failed Launches & Leveraging AI

Dean Peters & Kenny Kranseler | 00:55:08 – 00:59:40

Dean Peters:
So it happens at all different scales: moving from on-prem, moving to SaaS, and with acquisitions, how you make that happen and what stays and what doesn’t stay. I think we’ve got some other questions.

Question:
Someone has a zombie product, and the reason the product was not so successful is that it was launched poorly. It might not be a bad product, but I need to remarket it. What should I do?

Dean Peters:
I think we could take some lessons from how things worked in the world of mobile. I remember when mobile first came out, you would have an app that would go out and just get vilified. The app was fine. It just had a really sloppy launch, a really sloppy go-to-market motion, or maybe some other things were dropped along the way. So I think this is possibly an opportunity. I don’t know the product well enough; I know the company, but not the product. It’s possible that it’s a rebrand play. But if you’re going to do that, I’m going to be a real bear about the differentiation. You’d probably all agree on its benefits, but we’re going to have to have a very different set of differentiation that’s a lot more compelling than just saying we’re high quality, or we have more polite service. If you’re going to rebrand and relaunch the product, or even just say we’re going to do a relaunch, you need to be transparent, but you also need to really paint the picture of what the difference is. That’s all part of the go-to-market game as well. You know this game way better than I do, the marketing and messaging here. It probably should have happened a lot sooner, and things like differentiation should have been worked out much tighter and much more prominently. So there are obviously opportunities and success stories around that. But taking a step back, ask:

  • Do we rebrand?
  • Do we reprice?
  • Do we remarket?
  • What does the marketing and messaging look like before the product goes out?
  • How are we going to create buzz before this newer version goes out?
  • Who do we work with on this?
  • What’s the messaging going to be, and how’s it going to be different?
  • What promises and proof points are we going to make that are much more potent than what we went out with?

So there are things to be learned that way, and it’s definitely been done before.

Kenny Kranseler:
I think you hit on some of the good points there, Dean. What you’re doing is treating it as a new launch. If you have to rebrand it, rebrand it. That sometimes helps, because it lets you put your legacy behind. But you do have to make sure you’ve got compelling reasons for difference versus that launch you may have done and failed upon. It may not be new features, new capabilities, new components, but make sure that you stress the ones that you may have forgotten to stress last time.

Dean Peters:
Yeah. What makes you different, delightful, and hard to copy? It’s margin-enhancing for them, more so than the competitors. So we’ve got that. I think we’ve hit most of the key questions that I see out there, and we’re probably getting close to the top of the hour.

Kenny Kranseler:
I’ve got one question I think I want to answer.

Question:
What’s the one thing I need to be using AI for as a product manager?

Dean Peters:
I want to reframe that. First of all, there’s no one thing in product management. Kenny, I think you’ll agree, after doing it for so many decades: just like great security and great quality, product management is layered. Think of it like the dessert baklava: so many beautiful, delicious layers, but it’s really good when you put it all together. So I don’t think there’s one thing to be using AI for. It’s more: what’s part of my AI game, and where can AI help amplify, augment, even accelerate the types of things I do? We talk about that in the class quite often, the augmented intelligence play. Ethan Mollick is a really good author on that. He just came out with a new book on coexistence. His first book was Co-Intelligence. Both would be good reads to understand what that layered play is, and how to take that AI and make it your assistant, especially when dealing with zombie products.

Dean Peters & Kenny Kranseler | 00:59:40 – 01:00:00

Kenny Kranseler:
Yeah, definitely. All right. Thank you, Dean, for your time, for your attention. I want to thank the audience for hanging out and being around. We hope you found it valuable. And as I mentioned earlier, the replays will be available and will show up in your inbox soon. Enjoy.

Dean Peters:
All right. Have a great weekend, people. Keep it safe. Take care.

Webinar Panelists

Dean Peters

Known for building AI products decades before they were cool. Learn from Dean’s obsession with product and passion for AI.
LinkedIn

Kenny Kranseler

Kenny has led product teams from KFC to Amazon to Seattle-based startups. He has the range and strategic insight to differentiate what works from what doesn’t.
LinkedIn

Webinar Q&A

A product sunset strategy is a planned, phased approach to retiring a product that has stopped creating enough value, so the business can free capacity without stranding customers. It is most often needed for zombie products: offerings that keep consuming engineering time, support, compliance effort, and budget while nobody owns the decision to fund or end them. Productside principal consultant Dean Peters describes the test as whether the orange is still worth the squeeze. To spot zombie products, compare what a product earns against what it eats. Revenue is the visible tip of the iceberg, while support, maintenance, compliance, and technical debt sit below the waterline. Dean also points to opportunity cost, since maintaining an old product can block newer ones that go further on less fuel or battery. Zombie products usually survive for four reasons: inertia, the sunk cost fallacy, hope used as a strategy, and legacy commitments that pull in legal and finance. Checking every quarter, or monthly for fast-moving products, keeps a decline from quietly turning into a negative P&L.
Build a product sunset strategy as a progressive sequence, not a single announcement email. Dean Peters walks through an end-of-life lifecycle with five stages: announce internally first, send customers a notice of status change, declare end of sales, declare end of expansion with no new features, and finally reach end of service life. Each stage should be written into contract language so customers are never left hanging, and channel partners and distributors need old versus new feature comparisons so they are not left in the lurch. The goal is to lose the legacy product without losing the legacy customer. That means giving customers a clear timeline and a real path to your replacement offering rather than surprising, stranding, or bricking them. Revolv is the cautionary example: bricking lifetime users of the device triggered an FTC investigation and refunds. Sales discipline matters too. At Citrix, GoToMyPC had an end of sale announced years earlier while the product was still being bundled to sweeten deals. Kenny Kranseler adds that acquired products need an explicit path to success for existing customers, even when the old capabilities are retired.
Retire a zombie product when the data shows it no longer earns its keep, and keep it when the benefit still outweighs the cost. Dean Peters and Kenny Kranseler illustrate both outcomes. Early in his career at Kentucky Fried Chicken, Kenny was asked to kill the spork because executives thought it looked dated. After talking to operators, franchisees, and customers, he found that franchisees saved money on cheaper utensil packets and customers liked it, so the spork stayed. Dean names four good reasons to retire a product instead: the cost beats the revenue, strategy has shifted, a better replacement exists, or the market has changed and the need is gone. Deciding when to sunset a product therefore comes down to usage trends, true cost to serve, and the roadmap work the product blocks, not the loudest voice in the room. That loudest voice problem, along with losing customers on retirement, topped the audience poll. A sound product sunset strategy tests each product against these facts at least quarterly, so decisions rest on evidence instead of politics, a pet project, or an executive’s attachment.
The Retain, Retire, Revive, Replace framework is a four-move decision model that tells product leaders what to do with an aging product once its life cycle shows maturity or decline. Retain keeps a product that still fits, as with the spork. Retire brings it to end of life with no successor. Revive gives it new life through a version 2.0 or a rebrand and relaunch, which demands far stronger differentiation than claims of quality or polite service. Replace swaps it for a newer offering. Dean Peters stresses timing: start planning the extension or replacement play while a product is still maturing, because the best time to plant a tree was 20 years ago and the next best time is now. Makers of agricultural drones and cryogenic labels did this, introducing successors before the old product ended so no revenue gap opened. Productside’s Zombie Survival Kit, an agent skills library, supports the framework with skills for zombie identification, play selection, communications, and capacity transfer, and it offers expert, learner, and spitball modes. The tool recommends, but the product manager decides, which keeps any product sunset strategy a human judgment.
Get leadership buy-in by naming the freed capacity, showing the best bets it funds, and presenting the trade-offs, because retiring a product is a diplomatic and political battle as much as a technical one. Dean Peters recommends visiting legal and finance first, since contracts, bundles, and financial projections carry the most paperwork. Next come support, then sales and partners, a few key customers, and only then engineering and DevOps. For a holdout customer with a contract running four or five years, Dean suggests discussing options one level up: a discount or subscription, a rebate, an offer on something new, or keeping the product running cheaply until the contract ends. Account executives should size the client’s lifetime value and acquisition cost. Kenny Kranseler gave a major financial services customer priority rights on the replacement, and he built the functionality they feared losing into its early versions. Settle these scenarios before any public announcement, so no customer calls in confusion. A product sunset strategy that retires zombie products this way protects revenue while freeing teams to work on the next big bet.