When a private equity firm or corporate seller pulls a business unit out of a larger entity, the technology workstream is where the deal’s economics get won or lost. KPMG found that 71% of PE firms are actively pursuing carve-outs, and the separation of IT systems often becomes the single largest unplanned cost on the post-close P&L. A BCG analysis from June 2021 pegged one-time technology separation costs at 1–5% of the divested entity’s revenue, but that range hides enormous variance driven by TSA scope, data entanglement, and licensing. This guide puts real numbers against the line items that matter—TSA exit timing, identity and email separation, ERP and data untangling, licence renegotiation, and the contractor ramp—so you can model the workstream before you sign the purchase agreement.
If you’re running a mid-market carve-out in the US, Canada, or Australia, the numbers below are built from real engagements. We’ve seen a $40M-revenue manufacturing carve-out burn $2.8M on IT separation because the TSA clock ran out before email was cut over; we’ve also seen a $120M financial services divestiture land at $1.9M because the buyer brought in a fractional CTO who locked the architecture on Day 30. PADISO’s CTO as a Service exists for exactly that moment.
Table of Contents
- The Real Cost Framework
- Revenue-Band Benchmarks for Technology Separation
- The Three Budget Killers
- Licence Renegotiation: The Hidden Lever
- Contractor Ramp and Staffing Models
- How PADISO De-risks the Technology Workstream
- Summary and Next Steps
The Real Cost Framework
Technology separation costs fall into three buckets that Deloitte’s M&A IT cost classification captures well: one-time separation capital expenditure, transitional service agreement (TSA) run costs, and the steady-state run-rate uplift for the newly independent entity. The BCG 1–5% figure covers only the one-time separation capex. When you layer in TSA fees—which often run at a 15–30% premium over the parent’s internal chargeback—and the permanent cost of standing up a standalone IT function, the total cash outflow over the first 18 months can reach 7–12% of divested revenue for a heavily entangled carve-out.
FTI Consulting’s analysis of hidden carve-out costs underscores that buyers consistently underestimate IT separation by 40–60% during diligence. The root cause is almost always an over-optimistic TSA exit date. When the parent’s shared services team is not incentivised to prioritise your migration, the 12-month TSA you negotiated becomes 18 months, and the cost of keeping the lights on inside the seller’s data centre eats your first-year EBITDA uplift.
A Vendor Benchmark guide on carve-out IT costs breaks the one-time separation into infrastructure, applications, and data workstreams, with infrastructure typically consuming 35–45% of the budget. That’s where the hyperscaler strategy matters: if you can land the carved-out entity on AWS, Azure, or Google Cloud within the TSA window, you avoid building a replica on-prem environment and the associated capital outlay. PADISO’s Platform Design & Engineering practice specialises in exactly that—designing a cloud-native landing zone that is audit-ready from Day 1, which is why PE firms in New York and Austin call us before the deal closes.
Revenue-Band Benchmarks for Technology Separation
The table below gives a realistic range for one-time technology separation costs, based on the BCG June 2021 benchmark, cross-referenced with Umbrex’s carve-out playbook and Preferred Data’s cost estimate categories. These are total project costs—internal and external—over a 12–18 month separation window.
| Divested Revenue | Low-Entanglement Estimate | High-Entanglement Estimate | Typical TSA Monthly Fee |
|---|---|---|---|
| $10M–$25M | $400K–$800K | $800K–$1.5M | $50K–$120K |
| $25M–$75M | $800K–$1.8M | $1.8M–$3.5M | $120K–$250K |
| $75M–$150M | $1.8M–$3.5M | $3.5M–$6.5M | $250K–$500K |
| $150M–$250M | $3.5M–$5.5M | $5.5M–$9M | $500K–$800K |
Entanglement is a function of how many shared systems the carve-out depends on. A business that runs on its own instance of NetSuite with a separate Azure AD tenant is low-entanglement; a division that shares SAP S/4HANA, Active Directory, M365, and a custom data warehouse with the parent is high-entanglement. The TSA fee column assumes a 12-month agreement; extend that to 18 months and you add roughly 50% to the total programme cost.
The Three Budget Killers
Three line items consistently blow the technology separation budget. If you control these, you control the deal.
1. TSA Scope Creep and Exit Timing
The TSA is a commercial agreement, not a technical one. The seller’s IT team will deliver the minimum viable service, and every additional request—a new VPN tunnel, a firewall rule change, an extra month of support—comes with a change order. MASynergy’s carve-out best practices note that buyers who assume the TSA will cover “business as usual plus migration” routinely overspend by 25–35%.
The fix is to define exit criteria by system, not by date. For each application in the separation scope, agree with the seller: “When the carved-out entity’s instance passes user acceptance testing and production cutover is confirmed, TSA support for that system ends and the monthly fee reduces pro rata.” Without that clause, you pay for the full TSA until the last system migrates, even if 80% of systems are already live on the new stack.
2. Identity, Email, and Collaboration Separation
Email and identity are the first things users feel, and they can consume 15–20% of the separation budget if the carve-out has been on the parent’s Microsoft 365 tenant for years. The technical work itself—tenant-to-tenant migration of mailboxes, SharePoint sites, and Teams channels—is well-understood, but the hidden cost is in user downtime and helpdesk tickets. A 500-user migration typically generates 1,200–1,500 support interactions in the first two weeks post-cutover, and if you haven’t staffed for that, productivity tanks.
We always recommend a staged cutover: migrate a pilot group of 20–30 power users two weeks before the full organisation, capture every edge case (shared mailboxes, delegate permissions, mobile device profiles), and then run the main migration over a weekend with a war room staffed by both the migration vendor and the buyer’s IT team. PADISO’s case studies include a financial services carve-out where this approach kept user-impact tickets under 200 for a 400-user base, compared to an industry average of 1,000+.
3. ERP and Data Untangling
ERP separation is the most expensive single line item in a carve-out technology workstream. When the divested business shares an SAP or Oracle instance with the parent, you face a choice: lift out the relevant data and re-implement a standalone ERP, or negotiate a licence transfer and partition the existing instance. The first option is cleaner but costs more up front; the second is faster but leaves you with technical debt that compounds.
Data untangling is where the real budget overruns happen. The parent’s data warehouse almost certainly contains tables that blend carve-out and retained-business data, and untangling them requires deep knowledge of the schema. If that knowledge sits with a single DBA who is staying with the seller, you pay consulting rates for their time—and they are not prioritising your project. We’ve seen data migration estimates double because the buyer assumed they could simply “export and filter” a 50TB data lake, only to discover that referential integrity constraints required a full re-architecture of the reporting layer.
Licence Renegotiation: The Hidden Lever
Enterprise software licences do not automatically transfer in a carve-out. The parent’s ELA (Enterprise Licence Agreement) with Microsoft, Oracle, or Salesforce typically prohibits assignment, and the buyer must negotiate a new agreement under time pressure—often with the same vendor who knows you have no alternative. That’s a recipe for a 20–40% price premium over a comparable standalone deal.
The countermove is to bundle the carve-out’s licence needs with the buyer’s existing estate. If the PE firm already has a Microsoft ELA across its portfolio, adding the new entity as an affiliate can reduce per-seat costs by 25–35% compared to a standalone negotiation. Similarly, if the carve-out is moving to AWS, PADISO’s hyperscaler strategy includes negotiating enterprise discount programmes that can cut compute costs by 30–50% relative to on-demand pricing.
CTOinput’s IT carve-out guide emphasises that licence renegotiation must start during diligence, not after close. The moment the deal is announced, the incumbent vendor’s account team will classify the carve-out as “new revenue” and price accordingly. Getting a fractional CTO involved during the IOI stage—someone who has done this negotiation with Microsoft, Oracle, and Salesforce across multiple deals—pays for itself in the first ELA.
Contractor Ramp and Staffing Models
The technology separation workstream is inherently a temporary surge. You need infrastructure engineers, data migration specialists, ERP consultants, and programme managers for 12–18 months, after which the steady-state IT team is much smaller. The most common mistake is hiring full-time employees for the surge, then carrying their salaries into the steady-state P&L.
The right model for a mid-market carve-out ($25M–$150M revenue) is a core team of 3–5 full-time hires—a head of IT, a security lead, and application owners—supported by a contractor bench of 8–15 specialists. At peak, the monthly contractor burn can range from $120K to $350K depending on the ERP complexity and the number of parallel workstreams. That’s a significant cash outflow, but it’s predictable if you scope the work packages before the TSA clock starts.
PADISO’s Venture Architecture & Transformation engagement model is built for this exact staffing profile. We embed a fractional CTO who owns the separation architecture and vendor management, then bring in a curated network of cloud engineers, data migration specialists, and compliance leads who have worked together on previous carve-outs. The result is a team that ships from Week 1, without the ramp-up time that burns TSA months.
How PADISO De-risks the Technology Workstream
Kevin Kasaei founded PADISO to give mid-market buyers and PE firms the technical leadership that the Big 4 and strategy consultants reserve for $1B+ deals, at a price point that fits a $10M–$250M carve-out. Our work on technology separation spans three continents—San Francisco, Dallas, Miami, Sydney, and Melbourne—and we bring the same playbook to every engagement.
Here’s what that looks like in practice:
- Pre-close architecture lock. We join diligence in the final weeks and produce a separation architecture document that defines the target state for identity, networking, ERP, data, and security. This document becomes the scope baseline for the TSA negotiation, preventing the seller from charging for out-of-scope work.
- Vanta-powered audit readiness. Carve-outs often need SOC 2 or ISO 27001 compliance to retain enterprise customers post-separation. Our Security Audit service uses Vanta to compress the audit-readiness timeline to weeks, not months, so the new entity can pass its first audit before the TSA expires.
- Hyperscaler landing zone. We design and deploy a cloud-native foundation on AWS, Azure, or Google Cloud that meets the security and compliance requirements of the carve-out’s customer base from Day 1. This avoids the costly “lift and shift then re-platform” pattern that adds 30–40% to infrastructure costs.
- AI and automation from the start. Instead of replicating the parent’s manual processes, we inject agentic AI and workflow automation into the new entity’s operations. That might mean an AI-driven accounts payable workflow that reduces headcount requirements by 2–3 FTEs, or a customer service agent built on Claude Opus 5 or Sonnet 5 that handles Tier-1 queries from Day 1. Our AI & Agents Automation practice has shipped these solutions for multiple portfolio companies, delivering measurable EBITDA lift within the first two quarters.
Summary and Next Steps
Technology separation in a carve-out is a numbers game. The BCG benchmark of 1–5% of divested revenue sets the floor; the actual cost depends on how tightly you control TSA scope, identity migration, ERP untangling, and licence renegotiation. The three budget killers—TSA scope creep, email and collaboration separation, and data entanglement—can double your programme cost if they’re not addressed before close.
The single highest-ROI decision you can make is to bring in a fractional CTO who has done this before—someone who can write the separation architecture, negotiate the TSA exit criteria, and stand up the contractor bench without burning months on ramp-up. That’s the gap PADISO fills for PE firms and mid-market buyers across the US, Canada, and Australia.
If you’re evaluating a carve-out or already in the TSA window and feeling the clock, book a call with our team. We’ll give you a bottom-up cost model for your specific deal, not a generic benchmark. Read more on our blog for deep dives on AI strategy, platform engineering, and compliance, or browse our case studies to see how we’ve delivered separation outcomes for companies like yours.