Key Takeaways
- Customer operations is simultaneously used to mean CS operations, cross-functional post-sale infrastructure, and strategic GTM operations—most companies confuse these and design the wrong team
- 54% of companies had a dedicated CS operations role in 2024, down from 61% in 2023—an 11% decline that signals smaller companies are cutting ops headcount exactly when they need it most
- The operational backbone (systems, data, playbooks) is invisible until it's missing—it's the difference between "our CS team can't scale" and "our CS team doubled their book without new headcount"
- A single customer operations manager can support a team of six CSMs when the function is scoped correctly
- Customer operations lives or dies on integrations, not features—a CS platform that doesn't sync cleanly with your CRM and support system creates more work than it saves
You're hiring for customer operations. You post the req. Three different people forward you three different job descriptions. One looks like a data analyst embedded in customer success. Another reads like a support operations manager who also owns onboarding. The third is a strategic role reporting to the Chief Revenue Officer with a mandate to "own the post-sale customer journey."
All three are called Customer Operations Manager. All three exist at real companies. None of them do the same work.
The reason customer operations is hard to define is that it isn't one job—it's three overlapping functions that companies confuse, collapse into one title, and then wonder why the hire doesn't work. The first model treats customer operations as an enablement layer inside customer success. The second treats it as the operational backbone spanning support, success, and onboarding. The third treats it as a strategic peer to sales operations and marketing operations, owning systems and process across the entire post-sale motion.
You don't have a customer operations problem. You have a definitional problem. And until you name which model you're building, you'll keep hiring the wrong person into the wrong org chart with the wrong success metrics.
What is customer operations—and why does everyone define it differently?
Search "what is customer operations" and you'll find six authoritative definitions that contradict each other. Salesforce says it's "everything a company does to assist customers." Front says it's "strategic management of behind-the-scenes systems." Zendesk narrows it to "the crew helping customer success managers excel." Helpdesk.com widens it to "all activities from first contact to farewell."
They're all describing different jobs using the same title.
The confusion exists because customer operations emerged organically across B2B SaaS companies over the last decade—not as a designed function with a clear charter, but as a response to the same pain showing up in three different places. Some companies felt it in customer success when CSMs couldn't scale. Others felt it in support when ticket volume exploded. Others felt it cross-functionally when handoffs between sales, onboarding, success, and support kept breaking.
Same pain. Different organizational responses. Different definitions.
The three meanings of "customer operations" used in practice today
Here's what "customer operations" actually means when companies use the term:
Definition 1: Customer Success Operations (CS Ops). A role embedded inside the customer success organization that owns the systems, data, and playbooks CSMs use. Typical deliverables: health score models, QBR templates, Gainsight configuration, Salesforce hygiene rules, CSM onboarding plans. Reports to VP Customer Success or Chief Customer Officer. Exists to make CSMs more efficient.
Definition 2: Customer Operations (cross-functional infrastructure). A team that owns operational systems and processes across all post-sale functions—support operations, CS operations, onboarding operations, sometimes knowledge operations. Typical deliverables: unified customer data model, cross-team routing rules, escalation SLAs, shared reporting dashboards, post-sale tool stack rationalization. Reports to Chief Customer Officer, VP Operations, or sometimes directly to CEO. Exists to eliminate silos between support, success, and implementation.
Definition 3: Customer Operations (strategic GTM peer). A strategic function at the same organizational level as sales operations and marketing operations, owning the entire post-sale revenue motion as a system. Typical deliverables: retention and expansion playbooks, post-sale forecasting models, cross-GTM process design, customer lifecycle analytics, tool stack strategy across the full customer journey. Reports to Chief Revenue Officer or COO. Exists to treat post-sale as a repeatable revenue engine, not a cost center.
The first model is the most common. The second is the one that actually solves the operational pain most Directors of Support and VPs of Customer Experience feel. The third emerges only at companies that have decided post-sale revenue is a first-class GTM motion.
How the definition you choose determines your org chart (and your budget)
The model you pick isn't cosmetic. It determines who reports where, what gets funded, and whether customer operations has the authority to fix the problems it's hired to solve.
| Model | Reporting Line | Headcount Trigger | What It Doesn't Own |
|---|
| CS Ops (Definition 1) | VP Customer Success | 15–25 CSMs | Support ops, onboarding ops, knowledge management, cross-GTM process |
| Unified Customer Ops (Definition 2) | Chief Customer Officer or VP Ops | Support + CS + Onboarding combined headcount reaches 40–60 | Sales process, marketing automation, product ops |
| Strategic Customer Ops (Definition 3) | Chief Revenue Officer or COO | Post-sale ARR exceeds $50M or company treats retention/expansion as peer to new sales | Product roadmap, engineering delivery |
Here's what changes based on which model you choose:
If you build CS Ops (Definition 1), you get efficiency gains inside customer success, but support and onboarding continue running their own conflicting systems. Your CS platform doesn't sync with your support platform. Customer data lives in three places. Handoffs still break.
If you build Unified Customer Operations (Definition 2), you eliminate the silos—one customer data model, one routing engine, one set of escalation rules across support, success, and onboarding. But you don't get strategic influence over the revenue motion. You're still optimizing the back office while sales and marketing make the promises customer operations has to keep.
If you build Strategic Customer Operations (Definition 3), you get a seat at the revenue table. Post-sale becomes a forecasted, optimized, repeatable motion. But you need executive sponsorship, real budget authority, and a company mature enough to treat retention and expansion as peer priorities to new business.
Most companies default to Definition 1 because "customer success operations" is a known job title with a Radford compensation band. But Definition 2 is the structure that solves the real problem—fragmented post-sale operations across multiple teams.
What customer operations teams actually do (when they're set up correctly)
The value of customer operations is invisible until it's missing. When it's working, customer-facing teams operate smoothly. Data flows between systems. Playbooks exist and get used. New hires ramp in weeks instead of months. Health scores update automatically. Escalations route to the right person with full context.
When it's not working, your Director of Support is manually exporting ticket data into spreadsheets every Monday morning. Your CSMs are guessing which accounts are at risk because the health score model broke six months ago and nobody fixed it. Your onboarding team is reinventing the implementation plan for every new customer because nothing is templated.
Customer operations is the difference between "our CS team can't scale" and "our CS team just doubled their book of business without new headcount."
The operational backbone: systems, data, and process that keep customer-facing teams from drowning
Customer operations owns the how. Frontline teams own the who and the what.
Here's the dividing line: a CSM decides which accounts get a QBR this quarter (the what) and conducts the QBR with the customer (the who). Customer operations builds the QBR template, ensures the data feeding the QBR is current, configures the calendar automation that schedules it, and tracks completion rates across the team (the how).
A support manager decides escalation priority for a VIP ticket (the what) and personally handles the customer conversation (the who). Customer operations owns the routing rule that flagged the ticket as VIP in the first place, the SLA that governs response time, and the dashboard showing whether SLAs are being met (the how).
An implementation specialist runs a customer onboarding project (the who and what). Customer operations owns the milestone template, the health check that triggers if a milestone is overdue, and the handoff protocol that moves the customer from onboarding to ongoing success (the how).
The operational backbone is systems, data, and process. Systems: the tools customer-facing teams use—CRM, CS platform, support platform, communication tools, analytics. Data: customer records, usage data, health scores, interaction history, escalation logs. Process: playbooks, workflows, routing rules, escalation paths, reporting cadences.
When customer operations is absent or underpowered, every frontline team builds its own version of these. Support has its own customer data in Zendesk. Success has its own customer data in Gainsight. Onboarding has its own customer data in a project management tool. None of them match. Every handoff requires manual reconciliation.
What customer operations owns vs. what customer success/support/onboarding teams own
The ownership line gets fuzzy in practice because the work overlaps. A well-run customer operations function owns:
- CRM data quality rules and hygiene protocols (who updates what fields, when, and how conflicts get resolved)
- Health score model design and maintenance (what inputs feed the score, how they're weighted, when the model needs recalibration)
- Tool stack integration and administration (connecting Salesforce to Gainsight to Zendesk, managing user permissions, troubleshooting sync failures)
- Playbook templates and workflows (QBR decks, renewal motion sequences, escalation paths, onboarding milestone plans)
- Reporting dashboards and analytics (what metrics matter, how they're calculated, who sees which reports)
- Routing and assignment logic (which tickets go to which agents, which accounts go to which CSMs, how workload gets balanced)
- Process documentation and team enablement (how to use the tools, how to execute the playbooks, where institutional knowledge lives)
- Forecasting models for post-sale revenue (renewal risk scoring, expansion pipeline tracking, churn prediction)
Customer success, support, and onboarding teams own the actual customer interactions, the judgment calls that require context and relationship, and the outcomes those interactions produce. Customer operations owns the infrastructure that makes those interactions repeatable, measurable, and scalable.
The work no one sees: playbook design, CRM hygiene, health score architecture, and routing rules
Here's what a customer operations manager actually does in a given week:
Monday morning: Salesforce audit reveals 140 customer records missing renewal dates. Runs a data quality report, identifies the gap pattern (records created before a certain date when the field wasn't mandatory), writes a bulk update script, validates with finance that the dates match contract records, executes the fix, adds the field to required-field validation rules so it doesn't happen again.
Tuesday afternoon: CS team reports health scores aren't updating. Investigates. Discovers the Gainsight sync broke when IT updated the Salesforce API version last week. Reopens the integration, re-authenticates, tests on a sample account, confirms scores are flowing again, documents the fix in the runbook for next time.
Wednesday: Onboarding team escalates that three customers are stuck at the same milestone with no clear owner. Reviews the milestone definitions, realizes the handoff trigger from onboarding to success isn't defined. Schedules a 30-minute working session with the onboarding lead and the CS lead, writes a draft handoff protocol (completion criteria, notification rules, who does what), gets sign-off, configures the automation in the project tool, ships it.
Thursday: Builds a new QBR template because the old one is 18 months out of date and CSMs are customizing it inconsistently. Pulls the five best recent QBRs, identifies the common structure, standardizes the format, adds dynamic fields that auto-populate customer data from Gainsight, adds a slide on product usage trends, shares draft with three senior CSMs for feedback, incorporates feedback, publishes to the team playbook library, runs a 15-minute demo on the Friday team call.
Friday: Voice support costs $9–$16 per contact in 2026, live chat costs $5–$9 per contact. Runs a cost-per-contact analysis to see where the support team is tracking. Discovers chat volume is up 40% quarter-over-quarter but chat headcount is flat—agents are underwater. Pulls the top 50 chat topics by volume, identifies three topics that could be deflected with better help center content, flags them for the knowledge team, estimates deflection could save $12K/month if successful, writes it up in a one-pager for the Director of Support.
None of this work is visible to customers. None of it shows up in a QBR deck. But without it, the CSMs can't run QBRs, the support team drowns in chat volume, and onboarding customers get stuck at handoffs.
The three customer operations models—and how to choose yours
Most companies don't choose a model deliberately. They inherit one based on where the pain showed up first and who got budget to solve it.
If your VP of Customer Success hired the first ops person, you built CS Ops (Model 1). If your Chief Customer Officer hired someone to "fix the mess across support and success," you built Unified Customer Operations (Model 2). If your Chief Revenue Officer decided post-sale needed the same operational rigor as sales, you built Strategic Customer Operations (Model 3).
Here's when each model works—and when it doesn't.
Model 1: Customer Success Operations (the most common starting point)
Customer Success Operations is what most B2B SaaS companies mean when they say "customer operations." It's a function embedded inside customer success that owns the systems, data, and playbooks CSMs use to do their jobs.
Typical org structure: 1 CS Ops Manager or Specialist reporting to VP Customer Success. At scale (100+ CSMs), expands to a small team—CS Ops Manager, 1–2 CS Ops Analysts, sometimes a CS Systems Admin.
What it solves: CSM efficiency, data consistency within the CS org, playbook standardization, health score accuracy, forecasting for renewals and expansions managed by the CS team.
What it doesn't solve: Support operations, onboarding operations, cross-functional handoffs, customer data fragmentation across teams, tool sprawl.
A single customer operations manager can support a team of six CSMs when the scope is purely CS enablement. The ratio breaks when the CS team grows past 25–30 people or when complexity increases (multiple products, multiple segments, geographic expansion). At that point, you need a second hire.
When Model 1 works: You have a customer success team of 15–50 people, and the primary operational pain is inside that team—inconsistent CSM execution, unreliable health scores, no standardized playbooks, poor renewal forecasting. Support and onboarding are separate functions with their own ops (or no ops), and that's acceptable for now.
When Model 1 fails: Support, onboarding, and success all report to the same Chief Customer Officer, but they run on conflicting systems with duplicate customer data. Handoffs between teams break constantly. You hired a CS Ops person, but the real problem is cross-functional—and CS Ops has no authority to fix it.
Model 2: Unified Customer Operations (support + success + onboarding ops under one leader)
Unified Customer Operations treats the entire post-sale organization—support, success, onboarding, education—as one operational system with one shared backbone. It's the model that eliminates silos.
Typical org structure: Director or VP of Customer Operations reporting to Chief Customer Officer, VP of Operations, or CEO. Owns operations across support, CS, and onboarding. Team size scales with combined headcount—starts at 1–2 people when combined post-sale team hits 40–60, grows to 5–8 at enterprise scale.
What it solves: Customer data fragmentation, tool sprawl, handoff failures between support/onboarding/CS, inconsistent routing and escalation logic, inability to report on the full post-sale customer journey, operational inefficiency caused by each team running its own systems.
What it builds: One customer data model (typically in the CRM, synchronized to specialized tools). One routing engine (tickets, onboarding projects, CSM assignments all follow consistent logic). One set of escalation rules across all teams. Shared operational dashboards. Unified tool stack strategy (one support platform, one CS platform, one knowledge base—all integrated).
When Model 2 works: You have 40+ people across support, success, and onboarding combined. These teams currently operate independently with separate tools, separate data, and constant handoff friction. The Chief Customer Officer (or equivalent) has the authority and budget to consolidate operations under one leader. The company is mature enough to treat post-sale as a unified function, not three separate cost centers.
When Model 2 fails: The teams resist consolidation because they protect their autonomy. Or the company hasn't committed to unified tooling—support wants to keep Zendesk, CS wants to keep Gainsight, onboarding wants to keep Monday, and nobody has the authority to force convergence. Model 2 requires executive sponsorship and real decision-making power. Without it, Customer Operations becomes a coordinator with no leverage.
Model 3: Customer Operations as a revenue operations peer (strategic, cross-GTM)
Model 3 treats Customer Operations as a peer to Sales Operations and Marketing Operations—a strategic function that owns the post-sale revenue motion as a repeatable, forecasted, optimized system.
Typical org structure: VP Customer Operations reporting to Chief Revenue Officer or COO. Owns strategy, systems, and process across the full post-sale lifecycle—onboarding, adoption, renewal, expansion, advocacy. Collaborates with Sales Ops and Marketing Ops on cross-GTM process (handoff protocols, account planning, revenue forecasting). Team size: 3–10 depending on ARR scale.
What it solves: Post-sale treated as a cost center instead of a revenue engine. No visibility into renewal or expansion pipeline until 30 days before contract end. Retention and expansion playbooks that exist on paper but aren't systematized. Misalignment between what Sales sells and what post-sale teams can actually deliver. Inability to forecast post-sale revenue with the same rigor as new business.
What it builds: Retention and expansion revenue forecasting models. Cross-functional playbooks (how Sales, CS, and Support collaborate on at-risk renewals or expansion opportunities). Customer lifecycle analytics that connect product usage, support interactions, and revenue outcomes. Tool stack strategy across the entire GTM motion, not just post-sale. Operational frameworks that treat customer operations as a strategic capability, not back-office support.
When Model 3 works: Post-sale ARR exceeds $50M or represents a meaningful percentage of total revenue growth (30%+ of new ARR comes from expansions). The company has decided retention and expansion are strategic priorities at the same level as new customer acquisition. Leadership treats Customer Operations as a revenue function, not a cost function. The CRO or COO sponsors the investment.
When Model 3 fails: The company talks about retention and expansion but doesn't fund it accordingly. Customer Operations gets the strategic title but not the budget, headcount, or authority to execute. Or the function gets built before the company is ready—post-sale revenue is too small to justify the investment, and the VP Customer Operations ends up doing CS Ops work with an inflated title.
Real team structures: roles, ratios, and when to hire
The founding hire: Customer Operations Manager or Specialist (and what they build in year one)
The first customer operations hire is almost always made too late. It happens after the third escalation from a CS manager about "our data is a complete mess," or after the support lead manually reroutes 200 tickets in a week because the assignment rules broke, or after the CFO asks "why did we miss renewal forecast by 18% again?"
The trigger signals are measurable. CRM data quality below 60% (missing fields, duplicate records, stale information). CS team managing more than 30 accounts per CSM without segmentation by tier, ARR, or health score. Support ticket routing breaking at 500+ tickets/month because nobody owns the logic. Onboarding taking 90+ days with no consistent milestone tracking. Product usage data living in one tool, support history in another, CS notes in a third—and nobody can answer "is this customer healthy?" without pulling three reports and reconciling them manually.
When those signals appear, the founding hire is either a Customer Operations Manager (if the company can afford $90K–$120K and needs someone who can design systems from scratch) or a Customer Operations Specialist (if budget is tighter and the focus is execution—average $26.99/hour per ZipRecruiter).
What they build in year one: A clean customer data model in the CRM with mandatory fields enforced. Health score logic (product usage + support ticket volume + contract size + CSM sentiment = red/yellow/green) that updates automatically. Playbook templates for the three most common CS motions (onboarding, QBR, renewal). Support ticket routing rules that don't require manual intervention. A dashboard showing renewal pipeline by quarter with enough lead time to act. Integration between the CRM and whatever CS platform or support tool the company uses, so data flows without export-import cycles.
The work is unglamorous. It's Salesforce field mapping. It's writing routing logic in Zendesk. It's documenting the onboarding workflow so it's repeatable instead of reinvented per customer. But when it's done right, CS managers stop spending 8 hours a week on CRM cleanup, support leads stop manually assigning tickets, and renewal forecast accuracy improves from 65% to 85%.
When to add your second and third ops hire (and what changes)
The second ops hire happens when the first hire is underwater. Typically 12–18 months after the founding hire, or when combined post-sale headcount crosses 50 people. The trigger: backlog of operational requests (new playbooks, new integrations, new reporting) that can't be cleared in the current week, or the first hire is spending more than 30% of their time on firefighting (broken integrations, data cleanup, ad-hoc analysis for leadership) instead of building systems.
The second hire is usually a specialist in the area causing the most pain. If CRM data quality is still the constraint, hire a Salesforce Administrator or Data Operations Specialist. If CS playbooks exist but aren't being followed, hire a CS Enablement Manager. If support ticket routing and knowledge base management are consuming too much time, hire a Support Operations Specialist.
The third hire (usually 24–36 months in, or when post-sale headcount exceeds 80) splits the function into specialization. One person owns data and systems. One person owns playbooks and enablement. One person owns reporting and analytics. At this point the founding hire typically becomes Director of Customer Operations or Head of Customer Operations, managing the team instead of doing all the execution.
What changes with a three-person ops team: The function shifts from reactive (fixing what's broken) to proactive (designing what's next). Strategic projects that were perpetually backlogged—like building a multi-touch attribution model for renewals, or creating segment-specific onboarding tracks, or implementing AI-assisted ticket routing—actually ship. Cross-functional projects (aligning Sales Ops and Customer Ops on handoff protocols, or building shared dashboards for the executive team) become feasible because someone has the bandwidth to lead them.
Typical reporting structures—and why "where CustOps reports" signals its real scope
Where Customer Operations reports tells you what the company thinks the function does.
Reports to VP Customer Success: The function is scoped as CS Operations. It exists to enable the CS team—playbooks, health scores, renewal dashboards, CSM onboarding. It does not own support operations or onboarding operations. This is Model 1. Works when CS is the dominant post-sale function and support/onboarding are smaller or outsourced.
Reports to Chief Customer Officer or VP Customer Experience: The function is scoped as unified Customer Operations (Model 2). It owns systems and process across CS, support, and onboarding. The CCO or VP CX has authority over all three teams, so ops can drive consolidation. Works when the company has committed to treating post-sale as one function.
Reports to Chief Revenue Officer or COO: The function is scoped as a strategic peer to Sales Ops and Marketing Ops (Model 3). It owns retention and expansion as revenue motions, not just service delivery. Works when post-sale revenue is material ($50M+ ARR or 30%+ of new ARR growth from expansions) and leadership treats Customer Operations as a revenue function.
Reports to VP Operations or Chief of Staff: The function is probably under-resourced or misunderstood. It's been slotted into "general operations" because the company doesn't know where else to put it. This structure works only if the VP Ops or CoS has deep customer operations expertise and real authority over post-sale teams. Otherwise, Customer Operations becomes a coordinator with no leverage—asked to solve cross-functional problems without the authority to change how teams operate.
The ChurnZero 2024 study found 54% of companies had a dedicated CS operations role, down from 61% in 2023—an 11% decline in one year. That drop is concentrated in smaller companies cutting headcount. Larger companies (500+ employees) are increasing investment in Customer Operations while smaller companies are cutting it. The decline is a mistake being made in real-time. Companies are eliminating ops roles exactly when they need them most—during the transition from founder-led customer management to repeatable process. They will pay for it in churn and missed renewals 12–18 months later.
Benchmarks that matter: how to know if your customer operations function is working
Efficiency metrics: what ops work unlocks for frontline teams
Most Customer Operations teams are measured on lagging indicators they don't control. Net Revenue Retention. CSAT. Renewal rate. Customer Lifetime Value. These are outcomes produced by the frontline teams (CS, support, onboarding)—not by the ops function that enables them.
The operational metrics Customer Operations actually controls:
CRM data completeness: Percentage of customer records with all required fields populated (industry, segment, ARR, health score, assigned CSM, contract end date, product usage tier). Target: 95%+. If it's below 90%, the ops function doesn't have data governance under control, and every downstream system (health scores, forecasting, segmentation) is built on bad inputs.
Time to onboard a new CSM or support agent: Days from hire date to first productive customer interaction. Target for CSMs: 30 days. Target for support agents: 14 days. If it's longer, playbooks and training materials don't exist or aren't accessible. Ops owns this—not the hiring manager.
Playbook adherence rate: Percentage of customer interactions (onboarding kickoffs, QBRs, renewal discussions, support escalations) that follow documented playbooks instead of being improvised. Target: 80%+. Measured by spot-checking activity logs in the CRM or CS platform. If adherence is below 70%, the playbooks are either too complex, not embedded in workflow, or disconnected from how the work actually happens.
Average time to resolve a ticket routing conflict: When a support ticket gets assigned to the wrong team or person, how long does it take to fix? Target: under 2 hours. If it's longer, routing rules are broken and nobody owns them. This is pure operational waste—customer waits, agent waits, manager intervenes manually. Ops owns the routing logic and should know when it breaks.
Integration uptime and sync latency: How often do integrations between CRM, CS platform, support platform, and product analytics break? When data syncs, how long does it take? Target: 99.5% uptime, sync latency under 15 minutes. If customer data updated in Salesforce takes 4 hours to appear in Gainsight, the CSM is working from stale information. Ops owns the integration layer.
Forecast accuracy for renewals: At 90 days before contract end, how accurate is the renewal forecast (will renew / at risk / will churn)? Target: 85%+ accuracy. If it's below 75%, the health score model is broken or the data feeding it is incomplete. Ops owns the health score logic.
If your Customer Operations team can't produce these numbers on demand, it's not doing ops work—it's doing project management or firefighting. The function exists to make these metrics visible and improve them quarter over quarter.
Data health and system adoption (the leading indicators everyone ignores)
Data health predicts operational performance 60–90 days out. If CRM data quality drops in Q1, renewal forecast accuracy falls in Q2. If playbook adherence drops in March, CSAT drops in May. These are leading indicators—measurable now, predictive of outcomes later—but most companies don't track them until the lagging indicators (churn, NRR, CSAT) are already declining.
CRM field coverage by customer segment: Enterprise customers should have 100% field coverage (all required fields populated). Mid-market: 95%+. SMB: 90%+. If SMB data quality is at 60%, you're not managing that segment—you're guessing. Ops should run this report monthly and flag deterioration before it impacts forecast accuracy.
CS platform login frequency: How many CSMs log into the CS platform (Gainsight, ChurnZero, Planhat) daily? Target: 90%+. If it's below 70%, the platform isn't embedded in their workflow—they're managing customers in email and spreadsheets. Ops owns system adoption. Low login rates mean the tools aren't delivering value or aren't integrated properly.
Ticket deflection rate by knowledge base article: Which support articles are actually deflecting tickets (customer finds the answer, closes the page, doesn't contact support)? Which articles get viewed but don't deflect (customer reads it, still submits a ticket)? Target: 40%+ deflection rate on high-traffic articles. If articles are being viewed but not deflecting, the content is incomplete or unclear. Ops should flag low-deflection articles to the support team for rewrite.
Average touches required to complete onboarding: How many meetings, emails, or support tickets does it take to get a customer from contract signature to first value milestone? Track by segment and product. If enterprise onboarding requires 18 touches and SMB requires 22, the SMB motion is broken—it should be more automated, not more manual. Ops owns this metric and should be driving it down quarter over quarter.
The cost-per-contact and cost-per-resolution benchmarks for 2026
Customer Operations doesn't set headcount budgets or compensation—but it owns the process and tooling that determine cost efficiency. Every support interaction, every CS touch, every onboarding milestone has a cost. If ops improves routing, deflection, and automation, costs drop. If ops fails, costs scale linearly with volume.
The 2026 benchmarks for cost-per-contact by channel:
- Voice support: $9–$16 per contact
- Live chat / messaging: $5–$9 per contact
- Email / ticketing: $6–$11 per case
- Self-service (knowledge base, chatbot): $0.10–$2 per deflected contact
What this means for Customer Operations: If your support team handles 10,000 tickets per month and 80% are via email at $8/ticket average, monthly support cost is $64,000. If ops can shift 30% of those tickets to self-service deflection at $1/contact average, monthly cost drops to $47,800—a $16,200/month reduction, or $194,400 annually. The ops team didn't reduce headcount. It improved routing, built better knowledge base articles, and deployed an AI assistant that deflects tier-1 questions before they reach an agent.
First Contact Resolution (FCR) benchmark: 70% average, 85% for top-performing teams. FCR is the percentage of support cases resolved in the first interaction—no follow-up required. Low FCR (below 60%) means agents lack the information, tools, or authority to resolve issues without escalation. Ops owns this. If knowledge base coverage is incomplete, if routing sends tickets to the wrong team, if agents can't see full customer history in one screen—FCR stays low and cost-per-resolution climbs.
Cost-per-CSM (total CS team cost divided by number of CSMs) varies widely by company, but a useful benchmark: $120K–$150K fully loaded cost per CSM (salary + benefits + tools + overhead). If each CSM manages 30 accounts, cost-per-account is $4K–$5K annually. If ops improves health score accuracy and playbook efficiency so each CSM can manage 40 accounts without quality drop, cost-per-account falls to $3K–$3.75K. That's $30K–$50K in annual savings per CSM—or the ability to grow ARR 33% without adding headcount.
Customer Operations doesn't get credit for cost reduction in most companies. Finance sees lower support costs or higher CS account loads and attributes it to "the team working harder." Ops should own these metrics, track them monthly, and report them to leadership as operational performance—not hidden efficiency gains.
The tools and platform architecture (and why "best-in-class" usually means "doesn't talk to anything")
The core stack: CRM, CS platform, support platform, and the integration layer between them
Customer Operations lives or dies on integrations, not features. A CS platform with powerful health scoring, beautiful dashboards, and robust playbook automation is worthless if it can't sync cleanly with Salesforce and pull support ticket history from Zendesk. A "best-in-class" support platform that requires manual export-import to feed data into the CS platform creates more work than it saves.
The core stack has four layers:
Layer 1: CRM (system of record for customer data). Salesforce dominates at enterprise scale. HubSpot dominates at SMB and early-stage. The CRM holds account records, contact records, contract data, opportunity history, and product usage data (either natively or synced from product analytics). Everything else integrates with the CRM—it's the single source of truth.
Layer 2: Customer Success platform. Gainsight (enterprise, $1,500+/user/year), ChurnZero (mid-market, $12K–$50K/year depending on scale), Planhat (growth-stage SaaS, flexible pricing), Totango (mid-market, usage-based pricing). The CS platform sits on top of the CRM and adds health scoring, playbook automation, renewal forecasting, CSM activity tracking, and customer journey orchestration. It pulls data from the CRM, product analytics, and support platform, combines it into a unified customer view, and surfaces insights the CRM can't.
Layer 3: Support platform. Zendesk (most common at scale), Freshdesk (mid-market alternative), Intercom (product-led companies), Front (email-centric support). The support platform manages ticket lifecycle, agent assignment, SLA tracking, and knowledge base. It integrates with the CRM so customer data flows in and ticket history flows back out. Support teams work in this tool; CS teams pull ticket data from it.
Layer 4: Integration and workflow layer. Zapier, Make (formerly Integromat), Workato, native platform APIs. This is where most ops work happens—building the connections between CRM, CS platform, support platform, product analytics, billing system, and any other tools in the stack. A customer record updates in Salesforce → triggers a health score recalculation in Gainsight → creates a task for the CSM if health score drops below threshold → logs a note in Zendesk if there's an open support ticket.
The integration layer is invisible to frontline teams, but it's what makes the stack function as one system instead of five disconnected tools. Customer Operations owns this layer. When integrations break, data stops flowing, and every team reverts to manual workarounds. When integrations are well-designed, teams don't think about them—data is just there when they need it.
What Gainsight, ChurnZero, Totango, and Planhat actually do (and when each fits)
Gainsight: Enterprise-grade CS platform. Deep Salesforce integration. Comprehensive feature set (health scoring, success plans, playbooks, timeline view, customer journey analytics, forecasting, reporting). Expensive—$1,500+/user/year, often $100K+ annual commitment for mid-sized teams. Complex to implement (3–6 months to full adoption). When it fits: post-sale ARR above $50M, CS team of 20+ people, enterprise customers with complex onboarding and multi-year contracts, company committed to treating Customer Success as a strategic revenue function. Positioned as a Leader in Forrester Wave for Customer Success Platforms, Q4 2025.
ChurnZero: Mid-market CS platform. Easier to implement than Gainsight (30–60 days). Lower cost ($12K–$50K/year depending on scale). Good for SaaS companies with product usage data to track. Real-time alerts when customer behavior changes. In-app engagement (NPS surveys, feature announcements, onboarding checklists delivered inside the product). When it fits: ARR $5M–$50M, CS team of 5–20 people, product-led or hybrid motion, need to act on usage signals quickly.
Planhat: Flexible CS platform popular with European SaaS companies. Modern UI, fast implementation, usage-based pricing (pay for active customers, not seats). Strong product analytics integration. Good API for custom workflows. When it fits: growth-stage SaaS ($10M–$100M ARR), technical CS team comfortable building custom integrations, want a lighter-weight alternative to Gainsight without sacrificing power.
Totango: Mid-market CS platform, modular pricing (pay for the features you use). Pre-built integrations with common SaaS tools. SuccessBLOCs (packaged playbooks for common CS motions). When it fits: SMB and mid-market SaaS, team wants a structured out-of-the-box experience instead of building everything custom, budget-conscious but need more than spreadsheets.
The platform choice depends on company stage, budget, and how much customization the ops team can handle. Gainsight is powerful but requires dedicated admin headcount to manage. ChurnZero and Planhat are faster to value but less comprehensive. Totango is modular but feature depth varies. The wrong choice isn't "bad platform"—it's "platform that doesn't match company maturity." Buying Gainsight at $10M ARR is overkill and will sit underutilized. Trying to scale ChurnZero to support 100 CSMs managing $200M ARR hits limits.
The build vs. buy decision for smaller teams
Early-stage companies (under $5M ARR, CS team under 5 people) should not buy an enterprise CS platform. The cost doesn't justify the value, and the team doesn't have the operational maturity to use it. Instead, build lightweight Customer Operations in the tools you already own.
The lightweight stack for early-stage:
- HubSpot CRM (free tier or Sales Starter at $50/month)
- Intercom or Zendesk for support (Intercom starts at $74/month, Zendesk at $55/agent/month)
- Airtable or Google Sheets for manual health scoring and renewal tracking
- Zapier (Starter plan at $29.99/month) for basic integrations
- Slack for internal communication and alerts
What you can build with this stack: A customer record in HubSpot syncs to an Airtable base. Every week, a Zapier workflow pulls product usage data (from your product analytics tool), support ticket count (from Intercom or Zendesk), and contract data (from HubSpot), calculates a simple health score (green if usage is above threshold and tickets below threshold, red otherwise), and posts a Slack alert if any customer drops to red. CSMs manage renewals in HubSpot deals. The ops person (probably the first CS hire wearing two hats) maintains the Airtable base and Zapier workflows.
This is not scalable past 200 customers or 5 CSMs. But it's $200/month in tools instead of $30K/year for a CS platform the team won't fully use. Spend the first 12–24 months proving the CS motion works—retention rate above 90%, expansion ARR growing, CSMs following repeatable playbooks. Then buy the real platform when the manual process breaks.
The mistake early-stage companies make: buying Gainsight at Series A because a board member or advisor says "you need a CS platform." The platform gets implemented, nobody uses it because the workflows don't match how the team actually works, and 18 months later it's a $60K sunk cost with 20% login rate. Build first. Prove the motion. Buy the platform when manual process can't scale—not before.
What's changing in 2025–2026 (and what it means for how you staff this function)
CS Ops headcount is declining at smaller companies—but growing at mature ones
The ChurnZero 2024 Customer Success Leadership Study shows 54% of companies had a dedicated CS operations role in 2024, down from 61% in 2023—an 11% year-over-year decline. The drop is concentrated in companies under 500 employees. Larger companies (500+ employees) are increasing CS ops investment while smaller companies are cutting it.
What's happening: Smaller companies faced budget pressure in 2023–2024 and cut "non-revenue" roles. Customer Operations was categorized as overhead, not strategic investment. The cuts happened during the exact period when these companies needed ops most—transitioning from founder-led customer management to repeatable, scalable process.
The consequence will show up 12–18 months later. Without ops, CRM data quality deteriorates. Playbooks don't get documented. Health scores don't get built. Renewal forecasting stays manual and inaccurate. CS teams can't scale because the operational foundation was never built. Churn climbs. NRR drops. Leadership asks "why is CS underperforming?" The answer: you eliminated the function that would have prevented this.
Meanwhile, mature companies (those that survived the 2023–2024 downturn) are doubling down on Customer Operations. They've learned that retention is cheaper than acquisition, that expansion is the most efficient growth lever, and that neither scales without operational rigor. They're hiring Director and VP-level Customer Operations leaders, building 5–10 person ops teams, and treating the function as strategic infrastructure.
What this means for how you staff: If you're under 500 employees and don't have a dedicated Customer Operations person, you're making the same mistake 46% of companies are making. Hire earlier than the benchmark suggests. The founding ops hire should happen when combined post-sale headcount hits 15–20, not 40–50. At 15–20 people, one ops person can build the foundation (clean data model, health score, playbooks, integrations) before the team outgrows manual process. At 40–50 people without ops, you're already in crisis mode—backfilling what should have been built 18 months ago.
AI is eating tier-1 support work, and ops teams are the ones rewriting the workflows
AI isn't replacing customer-facing teams yet—but it's shifting what they spend time on. Tier-1 support questions (password resets, billing inquiries, product FAQs, simple troubleshooting) are being deflected by AI chatbots and knowledge base search. Tier-2 and tier-3 work (complex troubleshooting, escalations, customer-specific configurations) still requires human judgment.
The operational work this creates falls on Customer Operations. Someone has to build the knowledge base that AI retrieves from. Someone has to write the routing rules that determine when AI escalates to a human. Someone has to monitor AI performance (deflection rate, accuracy, customer satisfaction with AI interactions) and iterate when it underperforms. Someone has to retrain agents whose role just shifted from answering tier-1 questions to handling only complex cases.
This is ops work—not CS work, not support work. It's process design, system configuration, workflow automation, and performance monitoring. If Customer Operations doesn't own it, it doesn't get done consistently, and AI deployments plateau at 20% deflection instead of reaching 50%+.
What this means for how you staff: The Customer Operations role is expanding to include AI workflow design and optimization. The job description for a Customer Operations Manager in 2026 should include: "Design and optimize AI-assisted workflows for support and onboarding. Monitor AI deflection rates and accuracy. Collaborate with product and engineering on AI agent improvements. Retrain knowledge base content to improve AI retrieval." If your ops hire doesn't have these skills (or the ability to learn them quickly), they won't be effective in the next 24 months.
The compensation misalignment: CS owns revenue goals but isn't paid for them (and ops is stuck in the middle)
The same ChurnZero study found that only 44% of CS teams were compensated for expansions in 2024, down from 57% in 2023—a 13-point drop in one year. Meanwhile, CS teams are increasingly expected to drive expansion revenue as a core part of their charter.
The misalignment: CS is measured on Net Revenue Retention and expansion ARR, but comp plans don't reflect it. CSMs are paid like account managers (flat salary, small bonus tied to retention) instead of like sellers (meaningful variable comp tied to growth). The result: CS teams are accountable for revenue outcomes they're not incentivized to achieve.
Customer Operations is stuck in the middle. Ops builds the systems (health scores, playbook automation, expansion opportunity identification) that should drive revenue growth, but the frontline team isn't compensated to act on them. Ops flags 30 expansion opportunities per quarter. CSMs don't pursue them because it's not in their comp plan. The opportunities sit untouched, and 6 months later leadership asks "why didn't we hit expansion targets?"
What this means for how you staff: Customer Operations can't fix a compensation problem—but it can surface it. Ops should track expansion opportunity identification rate (how many qualified expansion opportunities does the system surface?) and expansion opportunity conversion rate (how many get pursued and closed?). If conversion rate is below 30%, the problem isn't the system—it's that CSMs aren't incentivized to act. Ops should report this to leadership as a strategic misalignment, not an execution failure.
The broader implication: As CS evolves from reactive support to proactive revenue generation, the ops function has to shift too. It's not enough to build playbooks and health scores. Ops has to build systems that connect customer data to revenue outcomes—expansion opportunity scoring, churn risk quantification, renewal forecasting with enough lead time to intervene. These are revenue operations capabilities, not just CS ops. If your ops team is still focused only on CRM hygiene and playbook templates, it's 24 months behind where it needs to be.
How to build your customer operations function in the next 90 days
Week 1–4: Audit what you have (systems, data, process gaps)
Week 1: Map every customer-facing tool and who owns it. Create a spreadsheet. Column 1: tool name. Column 2: what it does. Column 3: who administers it. Column 4: who uses it. Column 5: what customer data it contains. Column 6: does it integrate with the CRM (yes/no). You'll discover 8–12 tools nobody realized were in the stack. Three of them will be owned by someone who left the company 9 months ago. Two of them will be duplicating the same function.
Week 2: Run a CRM data quality audit. Pull a report of all active customer records. Check: How many have missing fields (industry, segment, ARR, health score, assigned owner, contract end date)? How many have duplicate records? How many have stale data (last updated more than 90 days ago)? Calculate the percentage of records with complete, current data. If it's below 70%, you have a data governance problem. If it's below 50%, you have a data crisis.
Week 3: Document the top 5 process gaps causing escalations. Interview CS managers, support leads, and onboarding managers. Ask: "What process broke in the last 30 days that required manual intervention?" Common answers: Ticket routing sent a case to the wrong team. CSM didn't know a customer had 3 open support tickets until the customer complained. Onboarding milestone was marked complete but the work wasn't done. Customer data in Salesforce didn't match what was in the CS platform. Contract renewal date was wrong and the customer almost churned because nobody reached out.
For each process gap, document: What broke? Why did it break? What's the root cause (bad data, missing integration, no ownership, unclear process)? What's the impact (customer at risk, team wasted time, revenue at risk)? These 5 gaps are your ops backlog for the next 90 days.
Week 4: Write a 1-page summary of findings. What percentage of CRM data is complete and current? How many tools are in the stack, and which ones aren't integrated? What are the top 5 process gaps, and what's the revenue or efficiency impact of each? What would it cost to fix these gaps (headcount, tools, time)? Present this to leadership. The goal: get agreement that Customer Operations is a strategic investment, not overhead.
Week 5–8: Define scope and make the first hire (or formalize the work someone is already doing)
Week 5: Write a scope document. One page. Header: "What Customer Operations Owns." Three sections:
- Systems: CRM administration, CS platform administration (if you have one), support platform configuration, integration layer between tools, data governance (field definitions, required fields, duplicate detection).
- Process: Playbook design and maintenance (onboarding, QBR, renewal, expansion, churn intervention). Routing rules (support tickets, CSM assignments, escalations). Reporting and dashboards (renewal pipeline, health score distribution, team activity).
- Enablement: New hire onboarding for CS and support teams. Tool training. Documentation of standard operating procedures.
What Customer Operations does NOT own: Customer-facing work (that's CS, support, onboarding). Product decisions (that's Product). Pricing or packaging (that's Finance and Product). Hiring or comp decisions for CS/support teams (that's the functional leader).
Week 6: Decide whether to hire or formalize. Two options. Option 1: Hire a net-new Customer Operations Manager or Specialist. Write a job description using the scope document. Post it. Budget: $70K–$120K depending on experience level and location. Option 2: Formalize the work someone is already doing. If a CS Manager or Support Lead is already spending 50%+ of their time on ops work (CRM cleanup, playbook creation, integration fixes), promote them to Customer Operations Manager, backfill their old role, and give them the scope document as their new charter.
Option 2 is faster and often better. The person already understands the business, the tools, and the pain points. They don't need 60 days to ramp—they can start shipping value in week 1.
Week 7: Set success metrics for the first 90 days. Three metrics. Metric 1: CRM data quality improves from X% to Y% (choose a realistic target—if you're at 60%, aim for 75% in 90 days). Metric 2: Top 3 process gaps from the audit are resolved (document what "resolved" means for each—ticket routing works without manual intervention, health score updates automatically, onboarding milestones are tracked in the CRM). Metric 3: One new operational asset is shipped (playbook template, dashboard, integration, knowledge base structure). Publish these metrics. Review them weekly.
Week 8: Onboard the ops hire (or the newly-promoted ops lead). Give them the audit findings, the scope document, the success metrics, and admin access to every tool in the stack. Set a weekly 1:1 to review progress and unblock issues. Protect them from ad-hoc requests for the first 30 days—they need focused time to build the foundation, not firefight.
Week 9–12: Build the first three operational assets that prove the function's value
Asset 1: A CS playbook template (pick one: onboarding kickoff, quarterly business review, or renewal discussion). Document the step-by-step process, the timeline, the required inputs (customer data, product usage, support history), the outputs (customer deliverable, CRM updates), and the success criteria. Make it a fillable template in Google Docs or Notion. Pilot it with 2–3 CSMs. Iterate based on their feedback. Roll it out to the full team. Measure adherence: how many CS interactions follow the playbook vs. improvise? Target: 70%+ adherence within 60 days of rollout.
Asset 2: A health score model in your CS platform (or in Airtable/spreadsheet if you don't have a CS platform yet). Define the inputs: product usage (logins per week, feature adoption, time in product), support activity (open tickets, ticket volume trend), contract data (ARR, contract end date, payment status), CSM sentiment (manually updated by CSM based on customer conversations). Define the scoring logic: green = all signals positive, yellow = one or more signals declining, red = multiple signals negative or critical event (contract up for renewal in 60 days, CEO change, major support escalation). Automate the calculation (Zapier workflow, CS platform automation, or weekly manual update if necessary). Surface the score in the CRM and CS platform. Measure: do CSMs trust the score? Are they acting on red accounts? Is forecast accuracy improving?
Asset 3: A Salesforce → support platform integration (or CRM → CS platform sync if that's the bigger gap). Choose the most painful data sync issue from the audit. Common examples: customer record created in Salesforce, doesn't appear in Zendesk for 24 hours. Support ticket closed in Zendesk, CSM doesn't know until they ask. Contract renewal date updated in Salesforce, CS platform still shows old date. Build the integration using Zapier, Make, or the platforms' native sync features. Test it with 10 customer records. Monitor for errors. Roll it out. Measure sync latency: how long does it take for data to flow from one system to the other? Target: under 15 minutes.
If you ship all three assets in 90 days, the function has proven its value. CRM data is cleaner. Playbooks exist and are being followed. Health scores are visible and actionable. Data flows between systems automatically instead of requiring manual export-import. CS and support teams see the difference in their daily work. Leadership sees the impact in forecast accuracy and operational efficiency.
If you can't ship all three in 90 days, one of three things went wrong. You scoped the role incorrectly (too broad, too many competing priorities). You hired the wrong person (not technical enough, not process-oriented enough, not comfortable with ambiguity). Or leadership didn't commit to the function (the ops hire got pulled into firefighting instead of building systems). Fix the root cause before the end of Q1. Don't let the function drift into undefined project management work. Customer Operations either builds repeatable systems or it's not Customer Operations—it's just someone doing everyone else's backlog.
The entire customer lifecycle—from first login to renewal to expansion—runs on a foundation of data, process, and systems that most companies don't build intentionally. They accumulate tools, patch workflows, and hope the frontline teams can make it work. Customer Operations is the function that builds the foundation deliberately. When it's done right, CS scales without linear headcount growth, support costs decline while volume increases, and retention becomes predictable instead of reactive. When it's missing, every team works harder to produce the same outcomes—and eventually, they can't keep up.
MatrixFlows is the customer operations platform for high tech—one foundation for the entire customer lifecycle: the knowledge that enables, the work and projects that deliver, and every request and submission from any channel, with AI and automation running on all of it. It deploys on the content and data you already have, works alongside your existing stack, and costs one flat price with unlimited users—no per-seat fees. And because the foundation is shared, the same platform extends to the partner and employee lifecycles the moment you're ready.
Create a Free Workspace →
Start building the operational foundation your customer teams need to scale. No credit card required.
.svg)
.svg)
.svg)
