Six portcos, six admins, six definitions of "customer." Here is how sponsors should actually staff portfolio RevOps: what each model costs, where each one breaks, and a calculator to pick yours.
The short answer: for five or more similar B2B portcos, a hybrid usually wins: a small sponsor-level team owns the standards (metric dictionary, match key, core fields, monthly pack) and each portco keeps one named local owner. A full shared team beats it when companies are near-identical and small; one team per portco only makes sense for large, unrelated companies held for a long time.
Outsourcing to a portfolio partner is the right call when the post-close window is short, the hold has under ~24 months left, or the fund will not carry headcount, as long as someone internal still owns the definitions.
Every sponsor ends up in one of four models. The cost gap is real, but the bigger gap is whether you can produce one comparable number across the portfolio.
| Criteria | One team per portco | Shared sponsor team | Hybrid (central standards + local owner) | Outsourced portfolio partner |
|---|---|---|---|---|
| Who owns definitions | Each portco, separately | Sponsor team | Sponsor team | Partner drafts, an internal owner signs off |
| Illustrative annual cost, 6 portcos* | ~$1.29M | ~$0.49M | ~$1.20M | ~$0.72M |
| Time to one comparable portfolio number | 6–9 months (hiring first) | 3–4 months | 3–4 months | 2–3 months |
| Depth inside each portco | High | Thin past ~8 portcos per operator | Medium-high | Depends on scope purchased |
| Single-point-of-failure risk | High: one person per company | Medium | Low | Low (team, not a person) |
| Best when | Large, unrelated companies, long hold | Near-identical small companies | 5+ similar B2B portcos, add-ons coming | Short window, specialized work, no headcount |
| Breaks when | You need a portfolio answer fast | The loudest portco takes all the capacity | Local owners aren't given time for it | Nobody internal owns the standards |
*Calculator defaults: $215K loaded cost per RevOps hire (from Glassdoor's 2026 US RevOps manager pay range of $134K–$231K, plus benefits and overhead), local owners at 55% of that, an outsourced retainer of $10K per company per month (JayOh's read of 2026 market ranges, not a quote). Change every input in the calculator.
The meme is about one cleanup estimate. The real problem is that every portco discovers the same mess on its own.
The cleanup estimate said 10 items. Then someone counted.
That's how most 100-day plans meet the CRM.
The scope gets written from memory, in a meeting, by people who use the system but never audited it.
The real number lives where nobody looks: unused properties, duplicate workflows, reports nobody has opened in years, picklist values with one record each.
So count before you scope.
The tools already exist: 🔸 HubSpot: Settings > Properties, then Monitor > Usage for fill rate and where it's used 🔸 Salesforce: Fields & Relationships, then "Where is this used?" on each field 🔸 Org Check from Salesforce Labs for a full-org scan 🔸 A Drive search for "forecast," "pipeline" and "FINAL" to find the shadow systems
That last one finds the dependencies a CRM audit misses.
Then sort the count. Most items have zero references and can go in waves of about 50, once you confirm nothing reads them.
The ones that matter are the few feeding forecast, comp or the board deck. Those get an owner, a plan and a date.
A big number isn't bad news. It means the real work is finally visible, and you can scope it honestly.
You can't plan a cleanup you haven't counted.
#RevOps #CRM #PrivateEquity #HubSpot #Salesforce
Portfolio RevOps rarely fails because of the model you picked. It fails because nobody owns what's shared between companies.
Each company writes its own cleanup scope from memory, defines "customer" its own way and builds its own board pack. The tenth estimate turns into 1,784 at every company, one at a time.
Add-ons were 75.9% of US buyout deals in Q2 2025 (PitchBook via Cherry Bekaert). Each one arrives with its own CRM, fields and definitions, and the cross-sell thesis needs one list nobody can produce.
In KPMG's October 2025 value creation survey, two-thirds of operating-partner respondents covered five or more portcos. Without a system, their time goes to the loudest company, not the biggest gap.
A lone admin builds the routing, the reports and the forecast. When they resign, the system walks out the door with them, usually right before a board meeting.
Board pressure pushes every add-on into the platform CRM in the first quarter. Without a shared key and shared definitions first, the migration imports the mess, and the second board meeting gets three customer counts and an apology.
Grade your portfolio against these. If you can't fill in the first two rows, that is the finding.
| Metric | Weak | Acceptable | Strong | Source |
|---|---|---|---|---|
| Time to one comparable customer count after close | More than 2 quarters | 1 quarter | 60 days or less | JayOh delivery experience |
| Portcos reporting from one metric dictionary | Under 50% | 50–90% | 100% | JayOh delivery experience |
| Portcos per centralized operator | 9 or more | 6–8 | 4–5 | Press & Associates, 2026 (centralized: ~4–8; embedded: 1–3) |
| Portcos with a named RevOps owner (internal or partner) | Some have none | All, but one person each | All, with a backup | JayOh delivery experience |
| US RevOps manager total pay (cost of the per-portco model) | $134K–$231K range, median ~$174K, before benefits, tools and recruiting | Glassdoor, 2026 | ||
| Operating partners covering 5+ portcos | About two-thirds of operating-partner respondents | KPMG global PE value creation survey, Oct 2025 | ||
| GPs ranking operational improvement as the #1 value lever | 72% | S&P Global Market Intelligence PE survey, Apr 2026 | ||
Sources: Press & Associates, Structuring the Value Creation Function (Feb 2026); Glassdoor RevOps manager salaries (2026); KPMG and S&P figures as reported by Not Very Private Equity (Aug 2026); add-on share from CapitalPad's roll-up statistics (Oct 2026). Weak / acceptable / strong bands without a third-party source come from JayOh's portfolio work and are labeled that way.
Pick by portfolio shape and time left in the hold, not by what the last fund did.
Portcos above roughly $50M ARR, different buyers and motions, 3+ years left in the hold. Each company needs depth more than comparability.
Fair warning: still standardize the board metrics, or you'll pay for six teams and still get six answers.
Six or more small B2B companies with the same ICP, the same CRM and the same motion. The economics are best here, as long as no operator covers more than ~8 companies.
Fair warning: write capacity rules down per quarter.
A small central team owns the dictionary, key, core fields and the pack. Each portco keeps one local owner for daily admin and requests. It costs more than a pure shared team and buys the depth that model lacks.
Post-close stabilization, a migration, a warehouse or attribution build, under ~24 months left, or no appetite for headcount. Use one portfolio-level agreement, not a vendor per portco.
Fair warning: keep one internal owner of definitions.
PE-backed platform with two fresh add-ons? Start hybrid even if you outsource the build: one dictionary, one match key, one cross-reference table, and one customer count reported while the three CRMs keep running. Consolidation can follow on a plan. Our guide to post-acquisition CRM consolidation covers that sequence.
These steps work under any of the four models. Only the owner column changes.
Settings > Account Management > Account & Billing for seats and editions, Connected apps for integrations. Salesforce: Setup > Company Information for licenses, Installed Packages and Connected Apps for integrations.Settings > Properties, then the usage view for fill rate and dependencies. Salesforce: Object Manager > Fields & Relationships and "Where is this used?", or Org Check from Salesforce Labs for a full-org scan.domain property. Salesforce: normalize Account.Website into an External ID field such as Portfolio_Account_Key__c. Land both in a warehouse or a governed sheet.StageName picklist, a Sales Process per record type, and validation rules for exit criteria.Standards before platforms, owners before tools, counts before scopes.
| Mistake | What it looks like | What it costs |
|---|---|---|
| Forcing every add-on onto the platform CRM in 90 days | Migration starts before definitions exist | Imported duplicates and broken reporting at the first board meeting |
| Hiring a junior admin per portco and calling it RevOps | Tickets close, nothing gets standardized | Six salaries, zero comparability, high turnover |
| A shared team with no capacity rules | The loudest CEO gets every hour | Quiet portcos with the biggest gaps never get fixed |
| Outsourcing with no internal owner | The vendor writes the definitions | Metrics that change when the vendor changes |
| Letting each portco pick its own tools | Three enrichment vendors, two routing tools | Duplicate spend and no shared playbooks |
| Measuring activity instead of comparability | "We closed 140 tickets this quarter" | No idea whether the portfolio is more predictable |
Enter your portfolio. The calculator scores all four models on fit, prices each one, and estimates time to one comparable portfolio number.
| Model | Fit | Annual cost | Per portco | Months to one number |
|---|
Cost math: one per portco = N × hire cost. Shared = (1 leader + N/6 operators, rounded up) × hire cost × 1.15. Hybrid = (1 leader + N/8, rounded up) × hire cost × 1.15 + N local owners at 55% of hire cost. Outsourced = N × retainer × 12 × 0.85 (portfolio agreement) + half an internal owner. Fit scores start at 40 and add or subtract points for portfolio size, similarity, ARR, CRM sprawl, hold time, add-ons and headcount appetite. These are planning estimates, not quotes.
Whether you hire, share or outsource, run every option through the same questions.
PORTFOLIO REVOPS DECISION CHECKLIST (JayOh) 1. INVENTORY [ ] Every portco listed: CRM + edition, seats, renewal date [ ] Named RevOps owner per portco (and a backup) [ ] Months left in the hold per portco [ ] Real count per portco: fields, workflows, reports, duplicates 2. STANDARDS (any model) [ ] Metric dictionary signed by the operating partner [ ] Match key: domain (accounts), email (contacts) [ ] Cross-reference table built before any migration [ ] Portfolio core field set, identical internal names [ ] Stage names + exit criteria standardized [ ] One monthly portfolio pack, same report spec 3. QUESTIONS TO ASK ANY OPTION (hire, shared team, partner) [ ] Who owns the definitions, and who signs off on changes? [ ] How many portcos per operator? (healthy: 4 to 8) [ ] How is capacity split across portcos each quarter? [ ] Time to one comparable customer count? [ ] What happens when the key person leaves? [ ] Can you show a portfolio pack you have run before? 4. RED FLAGS [ ] "We'll migrate everyone to one CRM first" [ ] No written capacity rules for a shared team [ ] Partner proposes its own metric definitions with no sign-off [ ] Each portco buying its own RevOps vendor [ ] Success measured in tickets closed 5. DECISION Model chosen: ____________ Owner of standards: ____________ Review date (quarterly): ____________
portfolio-revops-model-planner
Claude web or desktop: Settings > Capabilities, make sure code execution and file creation is on, then under Skills click Upload skill, choose the .zip and toggle it on. Claude Code: save the file as ~/.claude/skills/portfolio-revops-model-planner/SKILL.md, or in .claude/skills/ inside a repo.
Skills are available on Claude Pro, Max, Team and Enterprise plans. New to skills? Read how to use skills in Claude.
--- name: portfolio-revops-model-planner description: Recommends how a PE firm should staff RevOps across its portfolio (one team per portco, a shared sponsor team, a hybrid, or an outsourced portfolio partner) from the user's own portfolio inventory, then drafts the standards list and a 90-day rollout. Use when an operating partner, portfolio ops lead or platform CEO asks who should own RevOps across portfolio companies, after an add-on closes, or when planning portfolio-wide GTM reporting. --- # Portfolio RevOps Model Planner Apply the JayOh portfolio RevOps framework to the user's real portfolio. Work only from what the user provides. Never invent numbers, company names or tools. ## Inputs to collect Ask for anything missing, in one message, then proceed with what you have: 1. Each portfolio company: name or code, ARR band, CRM and edition, number of sellers, current RevOps or admin headcount, months left in the hold. 2. How similar the companies are: same ICP and sales motion, mixed, or unrelated. 3. Whether add-ons are expected in the next 12 months. 4. Budget posture: can the fund or platform carry central headcount, yes or no. 5. Loaded annual cost of a RevOps hire in their market (default 215,000 USD if they do not know; say that you used the default). ## Step 1: Inventory table Build one table: company, ARR band, CRM, sellers, ops headcount, hold months left. Flag every company with zero ops owners as a single point of failure risk, and every company with a different CRM from the platform. ## Step 2: Score the four models Score each model from 0 to 100 using these rules, and show the points for each rule: - One team per portco: +25 if average ARR is above 50M; +20 if companies are unrelated; +15 if hold months left are above 36; minus 20 if there are 6 or more companies. - Shared sponsor team: +25 if there are 6 or more companies; +25 if companies are similar; +15 if 3 or more CRMs; minus 20 if fewer than 4 companies. - Hybrid (central standards plus a local owner per company): +20 if there are 4 or more companies; +20 if companies are similar or mixed; +15 if average ARR is between 10M and 75M; +10 if add-ons are expected. - Outsourced portfolio partner: +25 if hold months left are under 24; +20 if add-ons are expected; +15 if 3 or more CRMs; +15 if the fund cannot carry central headcount. Start every model at 40 before applying rules. Cap at 100, floor at 0. ## Step 3: Annual cost per model Use the loaded hire cost (H) and the company count (N): - One team per portco: N times H. - Shared sponsor team: (one leader plus N divided by 6, rounded up) times H times 1.15. - Hybrid: (one leader plus N divided by 8, rounded up) times H times 1.15, plus N times 0.55 times H for local owners. - Outsourced: N times the monthly retainer the user gives times 12 times 0.85, plus 0.5 times H for an internal owner. If the user gives no retainer figure, show the cost as a range using 6,000 to 20,000 USD per company per month and label it a market range, not a quote. Show cost per company for each model. ## Step 4: Recommendation Name the highest scoring model. In 3 to 5 bullets, explain why using the user's own inputs as evidence (quote them). Name the runner-up and the one condition that would flip the decision. ## Step 5: Standards and 90-day rollout Regardless of model, list what must be standard across the portfolio: 1. One metric dictionary: customer (active and churned), ARR, new logo, pipeline, stage definitions, signed by the operating partner. 2. One match key: domain for accounts, email for contacts, plus a cross-reference table. 3. A portfolio core field set with identical internal names in every CRM. 4. Standard stage exit criteria. 5. One monthly portfolio pack built from the same report spec. Then draft a 90-day plan in three 30-day blocks with an owner for each item, using the chosen model's roles. ## Output format 1. Inventory table 2. Model scorecard table (model, score, annual cost, cost per company) 3. Recommendation and runner-up 4. Standards list 5. 90-day plan table (week range, item, owner, done-when) 6. Missing evidence: list every input you had to assume or that is missing ## Guardrails - Quote the user's inputs as evidence. If a number was not provided, say so in Missing evidence instead of guessing. - Do not recommend migrating every add-on onto the platform CRM in the first 90 days. Shared definitions and a shared key come first. - Do not name or recommend specific consulting firms or agencies. - Keep cost figures labeled as estimates from the user's inputs. Built by JayOh (jayoh.io). Framework and benchmarks: https://jayoh.io/portfolio-revops/
Straight answers to the questions operating partners bring to us most.
If you hold five or more B2B companies with a similar buyer and sales motion, a shared team or a hybrid usually wins: one set of definitions, one portfolio pack, and no single point of failure at each company. Let each company hire its own when the companies are large (roughly 50M ARR and up), unrelated to each other, and held for a long time.
Press and Associates (2026) puts centralized operators at about 4 to 8 companies each, and embedded operators at 1 to 3. Past eight, depth suffers: the shared team ends up serving whichever company is loudest.
Glassdoor (2026) puts US RevOps manager total pay between 134K and 231K USD, with a median near 174K. Add benefits, taxes, tools and recruiting and one loaded hire lands around 160K to 300K a year. Multiply by the number of companies and the per-company model is usually the most expensive option for six or more portcos.
No. Agree on one customer definition and one match key first, build a cross-reference table, and report one customer count while the separate CRMs keep running. Then consolidate on a planned sequence instead of rushing it before the first board meeting.
Outsource when the post-close window is short, when the work is specialized (migrations, warehouse builds, attribution), when the hold has less than about 24 months left, or when the fund will not carry central headcount. Keep one internal owner of the standards so the vendor does not end up defining your metrics.
Standardize the metric dictionary, the match key, a core set of CRM fields, stage exit criteria and the monthly portfolio pack. Keep daily admin, rep enablement and company-specific workflows local.
The fight over "shared team or one per portco" usually hides the real gap: nobody owns what sits between the companies. Fix that first. Get one dictionary, one key and one pack in place, and any of the four models can work. Skip it, and none of them will.
We've done this from both sides, standardizing RevOps across five portfolio companies and stabilizing new add-ons in the post-close window. If you want a second set of eyes on your portfolio, or a partner to run the standards while your local owners run the day-to-day, here's how we work with PE firms. Weighing a partner against hiring for a single company instead? Read RevOps agency vs. in-house vs. fractional or see what a RevOps agency costs in 2026.
Bring your portco list. We'll map it to a model in one call.
We engineer growth systems. JayOh builds the RevOps architecture, GTM execution and data infrastructure that turn messy revenue operations into predictable pipeline for B2B SaaS and PE-backed teams.
