Transcription tools. Automated health scores. Predictive churn models. AI-generated QBRs. The pitch is always the same. Buy this tool, automate the boring stuff, and your team will be free to focus on what matters.

But nobody is talking about what happens after the boring stuff is gone.

If AI handles your notes, your summaries, your meeting prep, and your follow-up emails, what are you supposed to be doing with all that freed-up time? The assumption is that CSMs will naturally become more strategic. More consultative. More valuable.

That is a dangerous assumption. Because most CSMs were never taught how to do any of that.

The skill nobody is teaching

Here is what I have learned the hard way. AI does not make you a better CSM. It makes you a more efficient one.

Efficiency without skill is just faster mediocrity.

The real upgrade is not in the tooling. It is in what comes out of your mouth on a customer call.

I started using NotebookLM recently to take deeper dives into my customer conversations. I feed it transcripts and ask it to look for pain points, technical requirements, business objectives, and anything else I can use to draft a strong mutual success plan. What it found surprised me.

NotebookLM picked up on several small, nuanced insights that came directly from the customer’s mouth. Things I completely missed while I was engaged in the conversation. A passing comment about a budget review. A hesitation when they mentioned a timeline. A technical requirement buried in minute thirty-seven that I was too deep in the flow to register.

The AI heard what I did not.

And that forced me to confront something uncomfortable. If AI is now dissecting every word my customer says, then every word I say matters more than it ever did before. My questions are the input. The quality of those questions determines the quality of everything that comes out the other side.

Discovery is the unlock

Sales teams have known this for years. They call it discovery. The structured process of asking the right questions to understand a customer’s situation, their pain, what is at stake, and how decisions get made. Sales professionals train on this. They role-play it. They build entire methodologies around it.

Most CSMs have never been trained on discovery at all.

There is a framework called SPICED, developed by Winning by Design, that breaks discovery into five layers. It was built for recurring revenue businesses, which means it applies across the entire customer lifecycle, not just the initial sale.

Situation is the background. Who is the customer and what does their world look like right now. Pain is what is actually hurting them underneath the surface-level frustration. Impact is what happens if the pain does not get solved, and this is the layer most CSMs skip entirely. Critical Event is the trigger forcing action, a renewal, a board meeting, a budget cycle. Decision is who decides, how they decide, and what criteria matter.

SPICED is not just a sales tool. It is a diagnostic framework that CS teams can use at every stage of the customer journey. Onboarding. QBRs. Renewals. Expansion conversations. Every time you sit down with a customer, you have an opportunity to re-diagnose where they are and what they need.

Where AI and discovery collide

Here is where it gets interesting. When you combine strong discovery skills with AI transcription tools, you get something that did not exist five years ago.

You can retroactively run discovery against your own calls. The customer already told you their situation, their pain, their impact. You just did not catch it because you were too busy managing the conversation. Now you can go back, feed the transcript into a tool like NotebookLM or a ChatGPT project, and pull out the SPICED information after the fact.

Retroactive discovery is a safety net, not a strategy.

The real power is in learning to ask better questions in the moment so the AI has better raw material to work with. Your transcript is only as good as the conversation you had. If you asked shallow questions, you get shallow insights. If you went deep, second layer, third layer, fourth layer, the AI gives you back gold.

I have started changing how I show up on customer calls. I am not thinking about notes anymore. I am not thinking about action items or what I promised last time. I am thinking about one thing: what is the next question that gets this customer to tell me something they have not told anyone else.

The grace period

If you are in a customer-facing role right now and you are not in sales, learn discovery now. Sales teams have been training on it for decades. Customer success is just catching up. Right now you can practice and build the muscle before it becomes a requirement, because this is where every customer-facing role is heading.

The AI is easy. Learning to actually listen is the hard part.

If you have started using AI tools in your customer conversations, I want to hear how it has changed the way you show up on calls. Hit reply and tell me.

One nearly killed a deal. The other created one worth four hundred thousand dollars.

Both happened on customer calls. Both involved people on my team. And the difference between them had nothing to do with process, playbooks, or pipeline management.

It had everything to do with reading the room.

The call that went wrong

We had just delivered the first cohort to a large customer. Barely on time. The customer was skeptical, on edge, and still working through buyer’s remorse about the original deal. We were halfway through the contract and only just starting to build real trust.

I brought our account executive into the call because I wanted them to see how delivery was going. Understand the customer. Be part of the conversation.

Instead of doubling down on the value we had just started to prove, the AE doubled down on the size of the contract. Another two million dollars. Right there on the call.

The customer was barely stomaching the first deal. And we were trying to shove a second one down his throat.

The call ended awkwardly. And the worst part was the AE did not even realize what had happened. He came off that call feeling proud. He had put the number on the table. In his mind, that was progress.

I gave us both time to settle before I said anything. When we did sit down, I walked him through where the customer actually was. Not where the pipeline said they should be. Where they actually were. Still uneasy. Still unsure. Still working through whether the first purchase was the right decision.

The customer had not had their moment of value yet. They had not hit that first real milestone that tells you the onboarding is working and the customer is ready to move into the next phase of their journey. Until that happens, any conversation about growth is going to feel like pressure, not partnership.

The AE was put off. It did not match the speed he wanted to move at. But he respected the call. We agreed to wait. Let the delivery speak for itself. Let the customer experience that first genuine moment of value where they stop questioning the purchase and start seeing what it can actually do for them.

When we brought the AE back in at the right time, the conversation was completely different. The customer was no longer defensive. They were excited. They had seen the results. They believed in the product. And they were the ones asking what else was possible.

We expanded from ten seats to over thirty. The account went from one and a half million to five point four million dollars. That is 260% growth in six months. Not because we pitched harder. Because we waited until the customer was ready to buy more.

Subscribe now

The call that went right

Different customer. Different team member. Same lesson from the opposite direction.

I had a CSM on my team who was new to the idea of expansion. Uncomfortable with anything that felt like a sales conversation. Most CSMs are. They got into this job to help people, not to sell them things. And there is this deeply ingrained belief in a lot of CS teams that revenue is someone else’s responsibility. That if you start talking about money, you are betraying the relationship.

I told him to lean in hard anyway.

He got on a call with a customer who had problems. A lot of them. At first, I could see him getting overwhelmed. The customer was doing most of the talking, unloading everything, and my CSM looked like he was in over his head. He was drowning in the customer’s frustration and did not know where to grab hold.

But he kept listening. He kept asking open-ended questions. He did not try to pitch. He did not try to redirect. He just stayed in it.

Then the magic happened.

The customer started connecting their problems to things our product could actually solve. And because this CSM had done the work to truly understand our products and services and how they deliver value, the light bulb went off. He had his moment.

I watched his ears perk up. He sat up a little straighter. He was biting his tongue, waiting for his moment to jump in. And when he did, he found this beautiful flow state with the customer. He was not selling. He was mapping their problems to our solution in real time.

What the customer needed was a bump to our simulator as a software model. That meant equipment refreshes and premium software licenses across all their enclosures, unlocking commercial features they did not have access to before. My CSM did not pitch this. He arrived at it by listening to the customer describe their problems and connecting those problems to a solution that actually existed.

The customer could feel it. My CSM could feel it. I could feel it from across the room.

That conversation turned into a four hundred thousand dollar growth opportunity.

When he came back to me afterward, he was over the moon. Not because of the number. Because he had solved a real problem for a real person and it actually meant something.

It is like riding a bike. Once you figure it out, it gets addicting.

What these two stories actually prove

Same company. Same product. Same types of customers. Two completely different outcomes. The AE read the room wrong. The CSM read the room right. And the difference was not skill or talent. It was timing and awareness.

We tracked NRR and GRR religiously — not as dashboard numbers to report up the chain, but as operational tools. GRR told us the truth about retention. NRR showed us where growth was actually happening. Together they gave us a map of which customers were primed for expansion, which were at risk, and which were quietly sitting in the middle — satisfied but not growing. That middle group is where most CS teams leave money on the table.

The moment of value is the hinge point. It is what takes a customer from questioning their purchase to being ready for more. The AE tried to skip it. The CSM waited for it. One created resistance. The other created a four hundred thousand dollar deal.

The hard part

Customer success is easy. People are hard.

The frameworks, the playbooks, the expansion models, the revenue metrics. Those are the easy part. The hard part is sitting in a room and knowing whether this is the moment to lean in or the moment to shut up. Knowing whether your customer has had their moment of value or is still working through buyer’s remorse. Knowing whether you are solving a problem or creating one.

You cannot teach that from a playbook. You learn it by being in the room. By reading the body language. By listening longer than feels comfortable. By trusting that the revenue will come if you get the timing right.

260% growth in six months. Four hundred thousand dollars from a single conversation.

Neither happened because someone followed a script. They happened because someone read the room.

If you have ever been on one of those calls, where everything clicked or where it all went sideways, I want to hear about it. Hit reply and tell me the story.

The customer signed up for a solution. What they got was sixty to eighty percent of one. The roadmap says the rest is coming. The customer does not care about your roadmap. They care about outcomes. And right now, your product is not delivering them.

So who fills the gap?

You do.

I saw a term this week that stopped me cold. Service as a Software. Not Software as a Service. The opposite. And it hit hard because I have lived this at every company where I have built CS from scratch.

The product was never ready. It was never finished. And the customers were never going to wait around for it to get there. So I became the thing that made it work.

Mapping the gap

The first thing I do when I walk into one of these situations is take the customer’s desired outcomes and map them back through the product. Not what the product could do someday. What it can do right now. Where does the software actually do the job? Where does it fall short?

That gap becomes my checklist. And that checklist becomes the real onboarding plan.

For example, maybe there is no clean way to mass upload users into the platform. The feature is on the roadmap. Probably Q3. The customer needs it now. So I facilitate it manually. I walk them through it step by step, help them get their team activated, and navigate them to the next milestone so they start seeing value even when the product is not making it intuitive.

Or maybe the reporting dashboard does not surface the metrics the customer actually cares about. So I build a manual report, pull the data myself, and walk them through it on a call until engineering catches up.

A lot of people would call this onboarding. What it really is, is a project plan built around the product’s limitations.

This is not a failure. This is the job. And if you are building CS at a company where the product is still maturing, this is the first thing you need to accept.

When the product catches up

This is the part that actually feels good.

Engineering ships the feature you have been duct-taping around for months. The thing you have been doing manually, the workaround you built, the process you held together with spreadsheets and calendar reminders. It is finally in the product.

Now you get to go back to your customers with a good news story. “Hey, remember that manual upload process we walked through together? There is a new feature that lets you do it in a fraction of the time.”

That check mark comes off your list. One less thing you are doing manually. One more thing the product handles on its own.

And here is where it gets interesting. Every time a gap closes, your role shifts. You spend less time on execution and more time on strategy. Less project management, more partnership. Less “let me walk you through this” and more “let me help you think about what is next.”

That is the direction you are always trying to move in. The checklist gets shorter. The conversations get deeper. The value you bring to the customer evolves from tactical to strategic.

When this cycle works, it is one of the most rewarding things in CS. You can literally watch your role mature alongside the product. You start as the bridge. You end up as the advisor. And the customer never had to feel the gap because you were standing in it the whole time.

When the product does not catch up

This is where it breaks.

If the company starts growing but the product stays at sixty percent, project creep sets in. What worked with ten customers becomes unsustainable with fifty. The manual processes that felt manageable when you had a small book of business start collapsing under their own weight. You start making mistakes. Outcomes get worse. Not because your team does not know what to do, but because they physically cannot keep up.

Then the burnout hits.

Your CSMs are exhausted. They are feeding product feedback into a system that feels like a black hole. They flag the same gaps every week. Engineering has different priorities. The problems that are making your customers frustrated are sitting in a backlog behind features designed to win new logos, not retain existing ones.

Your team starts to disengage. Not because they do not care, but because they care too much and nothing is changing. They know the product needs to catch up. They know the manual work is unsustainable. And they know that every new customer that comes through the door is another set of gaps they will have to fill by hand.

That is not a CS problem. That is a system failure. And it starts the moment leadership decides the product roadmap and the customer experience are two separate conversations.

The bridge has a time limit

Service as a software is not a permanent state. It is a bridge. The whole point is to hold the line long enough for the product to earn the right to stand on its own.

But if you are building CS from scratch, you need to design for it. Build your onboarding around outcomes, not features. Track the gaps you are filling manually so you can show product exactly where the software is falling short. Make that checklist visible. Share it with your product team, your leadership, your engineering partners. Turn your service layer into a feedback loop that actually drives the roadmap.

And keep shortening that checklist.

The companies that get this right build CS teams that start as the bridge and evolve into strategic partners. The ones that do not get it right burn through CSMs and wonder why nobody stays longer than a year.

If you have lived this, I want to hear from you. What does your gap list look like right now? Hit reply and tell me.

NRR. 110 to 120 percent. Post-COVID. Board was happy. Leadership was happy.

I was happy.

Two years later, I realized that number had been lying to all of us. We were losing 20 to 30 customers a quarter. Every quarter. The headline metric had been covering for a churn problem nobody wanted to name. Net new revenue was papering over it in real time.

That’s when I understood something most CS teams get completely backwards.

Every CSM owns or influences a revenue stream. Renewals. Expansion. Upsell. Churn prevention. Something. But most CSMs can’t name it. And most leaders haven’t made it clear.

So CSMs fill the gap their own way. They run QBRs because it feels strategic. They track adoption because it’s measurable. They schedule check-ins because it’s an activity they can show. But none of it ties back to money.

Then leadership wonders why CS can’t articulate its impact.

If you’re a CSM and you can’t name the revenue stream you own right now, this is for you.

Subscribe now

Start with the revenue streams

If you own renewals, your job is protecting and growing that number. That means understanding which customers are at risk, which customers can expand during renewal, and which customers have maxed out their current contract and need a conversation about growth. Not every customer gets a QBR. The ones at risk or poised for expansion do.

If you own expansion or upsell, you’re hunting for signal. Product adoption gaps. Seat saturation. New use cases emerging. Customers who bought module A but don’t know module B exists. That’s your territory. Your smallest customer might be your biggest opportunity this quarter if they’re positioned for growth.

If you’re tracking churn prevention, you’re watching for the early warning signs. Usage drops. Stakeholder turnover. Support ticket spikes. Silence. Sometimes the most dangerous customers are the ones who stop complaining, because they’ve already decided to leave.

The point is the same across all of them. You need to know your lever. You need to know which revenue stream you touch, and you need to tie every activity you do back to it.

The two metrics that tell you the truth

This is where NRR and GRR come in. If you're in CS and you don't understand these two numbers, stop what you're doing and learn them. They're not finance metrics. They're your metrics. They tell the story of whether your work is actually landing.

Net Revenue Retention (NRR)

This measures your starting revenue, plus expansion and upsell, minus churn and downsell. If you started the year with $1M in recurring revenue from your existing customers and you end the year with $1.1M from those same customers, your NRR is 110%. You grew without adding a single new logo. That’s the power of CS when it’s working.

NRR is the headline number. Leadership loves it. Investors love it. It tells a growth story. But here’s the problem. NRR can lie to you.

A big upsell in one account can mask churn bleeding out of three others. Net new expansion revenue can cover up the fact that you’re quietly losing customers every quarter. The top line looks healthy. Underneath, the foundation is cracking.

That’s exactly what happened to me. The only reason our NRR looked industry-leading was because we were annualizing revenue on initial purchases. These purchases were always significantly more than the subscription because they included cost of goods and equipment. It wasn’t truly recurring revenue, but it was annualized to present as recurring revenue. Within a $300K ticket item, only $15K was truly recurring. The rest was masking a churn problem we couldn't see.

Gross Revenue Retention (GRR)

This strips all of that away. GRR only measures your starting revenue minus churn and downsell. No expansion. No upsell. No growth to hide behind. Just: of the revenue you started with, how much did you keep?

If your GRR is 85%, that means you lost 15% of your existing revenue to churn and downsell before any growth kicked in. Your NRR might still look great at 115% because your expansion revenue covered the losses. But your GRR is telling you the truth. You have a retention problem, and you’re papering over it with new money.

That’s why GRR can be the more decisive metric. It’s harder to game. It forces you to look at the health of your base without the comfort blanket of expansion masking the cracks.

Once we cut through the noise and looked at GRR, we realized we had a serious churn problem that had been running for at least two years. So we went back to basics. We started understanding why these customers were actually churning. And when we figured that out, my team launched a reactivation campaign. In six months, we recovered 44% of the revenue we had lost over two years.

Can you imagine what happened to our NRR? It skyrocketed. But only because we decided to look at the hard thing instead of celebrating the sexy number.

What this means for your day-to-day

When you understand your revenue stream and you understand NRR and GRR, your priorities change. You stop treating every customer the same. You start segmenting by risk and opportunity. You stop running QBRs as a checkbox exercise and start running them as revenue conversations.

Your activities become intentional. Every call, every email, every touchpoint ties back to one of two things. Protecting existing revenue or creating the conditions for growth. That’s it. Everything else is noise.

This is how you stop feeling like admin. This is how you stop flying under the radar. This is how you become undeniable to leadership. Because leadership speaks money. If you can’t articulate which revenue streams you own and how you’re protecting or growing them, you’re not a strategic partner. You’re a checkbox runner.

Know your revenue stream. Own it. Measure it. Communicate it.

If you’re a CSM, which revenue stream do you actually own? Can you name it right now? Can you tie the activities you did this week back to it?

If you’re a leader, have you told your team which revenue streams they own? Or are you leaving them to figure it out on their own?

Subscribe now

I’ve walked into CS functions where the first thing leadership says is “We need playbooks.”

No customer journey mapped. No segments defined. No data. No process. But they want playbooks.

Here’s the thing about playbooks. They only work if you actually know what’s going on with your customer. Where they’re at. What they need. What stage of the journey they’re in.

And most CS teams don’t know that yet. They’re guessing.

So let’s back up.

Before you build a single playbook, you need to answer a few things first.

Do you have a customer journey? Is it mapped from start to finish? And is it one journey or multiple? Because the journey for a single user on a self-serve plan looks nothing like the journey for a large enterprise org. But if you serve both, you need to understand both. Different segments, different motions, different needs.

After that, you need to be honest about where your own CS program actually is.

Stage one is infancy. You’re putting pen to paper. You don’t have a lot of data. You don’t have a lot of process. You’re still figuring out what your CS motion even looks like. And that’s fine. But you’re not ready for playbooks. You’re ready for listening.

Stage two is early maturity. You’re building a database. You’re creating some data hygiene. You’re putting fundamental processes in place to capture customer experience and take action. But let’s be clear. This is still reactive. You’re responding to what comes at you, not anticipating what’s next.

Stage three is where playbooks actually make sense. Because by now you’ve seen the patterns. You’ve identified the repeatable issues. You’ve done the root cause work. You know what keeps showing up and you’ve built responses that actually drive impact.

Until you get to stage three, everything is reactive. And that’s not a failure. That’s the job.

What matters is what you do in those reactive moments.

Did you write it down? Did you take note of what happened and why? Did you bother to identify the root cause? Did you bring that root cause forward to drive a solution? Did you figure out whether it was a recurring issue or a one-off?

If it’s recurring, that’s playbook material. If it’s a one-off, document it somewhere, put it in a wiki, and move on.

That’s the unsexy work. That’s the work nobody puts on a slide deck. But that is how playbooks get built. They’re earned through reps, not manufactured in a workshop.

Here’s what happens when you skip it.

Three or four years from now, your company has grown. Revenue is up. Customer base is bigger. But your CS program is sitting at the same maturity level it was on day one. More customers, more complexity, same reactive team scrambling without a system.

You didn’t skip the playbook. You skipped the foundation the playbook was supposed to stand on.

So if you’re in a CS function right now and someone is asking you for playbooks, ask yourself this first. Do I actually understand my customers well enough to build one? Do I know where they are in their journey? Do I know where my own program is in its maturity?

If the answer is no, that’s not a problem. That’s your starting point.

The playbook comes later. The work starts now.

If you’re a CSM, where are you at right now? Still in the reactive stage or starting to see the patterns that turn into playbooks?

If you’re leading a CS team, have you built the foundation first or did you jump straight to playbooks and wonder why nothing stuck?

You might be wondering what the hell you’re doing here. It’s probably because you’re in the middle of something messy. Maybe you’re leading a team for the first time and nobody gave you a playbook. Maybe you’re a founder building something from scratch and realizing the people part is harder than the product part. Maybe you manage customers, manage people, or manage both, and you’re tired of advice that sounds great on a slide but falls apart on a Monday morning.

This Substack is for you.

I’m Josh. I’ve spent the last decade in customer success, but the lessons I’ve learned aren’t about customer success. They’re about people. Difficult stakeholders who won’t return your calls. Leadership teams that say they want change but won’t fund it. Teammates who are burned out but won’t tell you. Customers who smile in your QBR and ghost you before renewal. The gap between how things should work and how they actually do.

That’s the stuff I write about. The human side of doing hard work in messy organizations.

I’ve built programs from scratch with no function, no headcount, and a half-abandoned CRM. I’ve grown a single account from $1.3M to $5M in 90 days. I’ve been laid off the day after my birthday while my family was overseas for a funeral. All of it taught me something. Most of it taught me the hard way.

What I’m trying to do here is put pen to paper and share those experiences in a way that actually helps. Not frameworks pulled from a textbook. Not AI-generated advice that sounds smart but says nothing. Just what I’ve seen, what I’ve built, what I got wrong, and what I’d do differently.

If you’re in customer success, I want to be the person who makes the job make sense. If you’re leading people in any function, I want to give you something real you can use this week. If you’re a founder trying to figure out why your customers aren’t sticking around, I want to make it simple.

I’m here to be a sounding board. A thinking partner. Someone who will share what they know, admit what they don’t, and welcome a good kick in the teeth when I’m wrong. Because that’s how you learn. And I’m still learning.

I publish here twice a week, with a deeper newsletter every two weeks called The CS PRESS. If any of that sounds like something you need right now, stick around.

And if you’re building something, leading people, or just trying to figure it out as you go, I’d love to connect. Drop a comment, reply to a post, tell me what you’re working on. I read everything.

Let’s figure this out together.