Most B2B SaaS companies have a customer success function. Most of them are doing a version of the same thing - health scores, QBRs, check-in cadences. The ones that are genuinely driving retention and expansion are making different decisions underneath the familiar structure. Here is what actually separates good customer success from great.
Early in my CS career at Zendesk, I inherited a book of accounts where about 60% had green health scores. About a quarter of those accounts churned in the next two quarters.
The health scores were not lying. They were measuring the things they had been built to measure - login frequency, feature usage, support ticket volume. What they were not measuring was whether the product was actually embedded in the customer's workflow in a way that made it hard to leave.
Those are different things. And the gap between them is where most B2B SaaS customer success strategy gets stuck.
The visible layer of CS - QBRs, health scores, check-in cadences, NPS surveys - is table stakes at this point. Almost every company with a meaningful SaaS revenue line has these things. The difference between companies that are genuinely driving retention and expansion and the ones that are managing churn defensively is what they have built underneath that layer.
What Most B2B SaaS Customer Success Programs Actually Are
Most B2B SaaS CS programs are organized around monitoring and intervention. The monitor-and-intervene model looks like this:
- Health score goes red
- CSM schedules a call
- CSM works on recovering the account
- Health score goes green
- Repeat
This is a reactive model wearing the language of proactive customer success. It is not without value - intervention does prevent some churn. But it is fundamentally optimized for the wrong thing. (For a look at how this reactive vs. proactive dynamic plays out in other areas, see my thoughts on AI for customer service: the honest version in 2026).
It is optimized for detecting problems, not for building the kind of deep product integration that makes churn unlikely in the first place.
The companies doing customer success well are building a different model. Not instead of health scores and QBRs - alongside them. The foundation is a genuine understanding of what value the customer actually receives from the product, and a systematic program to expand that value over the relationship.
The Metric That Most CS Teams Are Not Tracking (But Should Be)
The single most useful metric for predicting long-term retention in B2B SaaS is not health score. It is the number of people inside the customer organization who have embedded the product into their daily work.
Call it user depth, or workflow embedding, or whatever term your organization uses. The concept is: not just whether someone is logging in, but whether the product has become a necessary tool for how a specific person does a specific job.
One power user who cannot imagine doing their job without your product is worth more to your retention than ten casual users who log in regularly.
Per a Gainsight study on SaaS retention drivers, accounts where three or more users had the product embedded as a daily work tool had 87% lower churn rates than accounts with similar login metrics but shallow usage. The depth of usage matters more than the breadth.
Most CS teams are measuring breadth. The ones with better retention are measuring depth.
What Great Customer Success Strategy Actually Includes
Let me be specific about what this looks like in practice, because the framework version of this advice tends to stay abstract.
A clear model of value for each customer segment. Not a generic value proposition - a specific articulation of what success looks like for a customer in this industry, at this size, with this use case. The QBR where you present product usage data and ask if everything is going well is less useful than the QBR where you say "based on what you told us in onboarding, you were trying to achieve X - here is where you are against that, here is what is in the way, here is what we are going to do about it."
That requires knowing what X is for each customer. Most CS teams do not know. The information exists in onboarding notes or sales handoff documents that nobody has systematically reviewed.
Active expansion of user depth inside the account. The CSM's job in a well-run program is not just to maintain the existing relationship. It is to expand the number of people inside the customer organization who have built the product into how they work.
This is different from upsell. It is the precondition for sustainable retention.
At Adobe, I worked on programs that tracked user depth inside enterprise accounts as a leading indicator of renewal health. The accounts that had expanded user depth - not just usage volume, but genuine workflow embedding across different teams - renewed at significantly higher rates and at larger contract sizes than accounts at similar ARR with shallower embedding.
Proactive engagement around customer business outcomes, not just product outcomes. The customer does not care that their usage is up 20%. They care about whether the product is helping them achieve the outcome they bought it for.
The CS teams that can connect their product metrics to the customer's business metrics - pipeline, cost reduction, time saved, error rate, revenue - are having fundamentally different conversations than the ones that present product usage dashboards.
A systematic early warning system for risk that is not based only on health score. Health score is useful. It is a lagged signal.
The leading signals - changes in stakeholder engagement, shifts in the types of questions being asked, changes in renewal conversation timing, new contacts appearing who did not exist in the original relationship - require relationship intelligence that most CS tools are not capturing systematically.
The Customer Success Playbook for Expansion
Expansion revenue in B2B SaaS is the difference between a CS team that pays for itself and one that is a cost center. Most CS teams know this. Fewer have a systematic approach to it.
The expansion model that works is built on genuine value expansion, not on sales pressure. Customers who expand their spend do so because they have experienced enough value that the case for expanding is obvious. The CSM's job is to create the conditions for that case to exist, not to build the sales pitch.
What that looks like in practice:
Identifying expansion-ready accounts before they know they are ready. The signals are usually visible: usage that is consistently hitting limits, users from adjacent teams asking about the product, a business objective the customer mentioned that the product could address but has not yet been deployed against.
Introducing the expansion conversation as a next step in achieving something they already said they want. Not "I want to talk to you about upgrading your contract." "You mentioned in our last QBR that you were trying to roll this out to the operations team - I think we are ready to help you do that. Want to map out what that looks like?"
Building internal champions who will advocate for expansion. The economic buyer who signed the contract is not usually the person who will push for expansion. The person who pushes for expansion is the one who has embedded the product in their daily work and whose colleagues are asking them about it.
CS teams that identify and invest in that person consistently see better expansion rates than ones focused only on the economic buyer relationship.
What Separates Great CS Teams Structurally
The tactical things matter less than most CS leaders think. Health score thresholds, QBR frequencies, success plan templates - these are execution-layer choices. The structural decisions that separate great CS teams are higher level.
How the team is sized and segmented. A CS model that tries to apply the same coverage model to a $10K ARR account and a $500K ARR account will underserve both. The customers that drive your retention economics need more investment than the long-tail accounts. (This dynamic where attempting to standardise the approach hurts performance is often what micromanagers are actually afraid of losing control over when teams segment properly).
That sounds obvious. Most CS teams are not resourced to execute it.
How CS and sales are connected at the account level. In the best-run programs I have seen, CS and sales have a shared view of each account - what the customer said in onboarding, what the CS team has observed, what expansion potential exists, and what the risks are. In most organizations, these teams operate with separate data and separate relationships and then try to coordinate at renewal.
How success is defined internally for the CS team. If the CS team is measured only on retention and NPS, they will optimize for retention and NPS. If they are also measured on expansion and on user depth metrics, they will build the behaviors that drive those outcomes.
The measurement model shapes the behavior of the team over time.
