{"site":{"name":"Koji","description":"AI-native customer research platform that helps teams conduct, analyze, and synthesize customer interviews at scale.","url":"https://www.koji.so","contentTypes":["blog","documentation"],"lastUpdated":"2026-11-12T08:04:11.677Z"},"content":[{"type":"blog","id":"281a280d-6004-4672-bbd6-acf72ac98dc3","slug":"carve-out-customer-interviews-pe-playbook","title":"Carve-Out Customer Interviews: A PE Operating Partner Playbook for Keeping Customers Through Day 1","url":"https://www.koji.so/blog/carve-out-customer-interviews-pe-playbook","summary":"A playbook for private equity operating partners on interviewing the customers of a carve-out before Day 1. It cites KPMG's 2026 Global M&A Outlook (71% of PE dealmakers open to separations, with no customer item among the challenges they rate as material), sets a sampling rule by revenue, gives an eight-week timeline counted back from Day 1, introduces the Three Ties test (parent, people, product) for sorting accounts, includes an eight-question interview guide, and shows the setup in Koji with guardrails, personal-detail removal and personal links.","content":"Carve-out customer interviews are the conversations you hold with the customers of a business you are buying out of a larger company, before Day 1, to find out which of them stay because of the product and which only bought because of the parent. Every carve-out plan has a workstream for IT, finance and people, and almost none has one for customers, even though they signed their contracts with someone else. Run as AI-moderated interviews, you can hear dozens of accounts in about three weeks, sorted by revenue, while there is still time to change the Day 1 message. By the end of this playbook you will have the sampling rule, a timeline counted back from Day 1, a test for sorting accounts, an interview guide to paste, and the setup in Koji.\n\n## Why carve-out customer interviews matter now\n\nCarve-outs have moved from an occasional deal type to a steady part of private equity's pipeline, and the planning tools have not caught up on the customer side.\n\n**More sponsors are buying carve-outs.** KPMG's 2026 Global M&A Outlook, published in March 2026 from a survey of 700 dealmakers (181 of them in private equity), reports that 71% of PE dealmakers are open to or actively pursuing portfolio separations, and 55% already have carve-outs under consideration, against 21% of corporate dealmakers ([KPMG, March 2026](https://www.kpmg.com/xx/en/media/press-releases/2026/03/kpmg-survey-of-global-dealmakers-reveals-rising-m-and-a-expectations.html)). KPMG calls 2026 the year of the carve-out.\n\n**Deals are more numerous and smaller.** S&P Global Market Intelligence counted 421 private equity carve-out deals in 2025, about 10% more than the year before, while their total value fell 13.1% to $65.2 billion ([S&P Global Market Intelligence, April 2026](https://www.spglobal.com/market-intelligence/en/news-insights/articles/2026/4/global-private-equity-carve-out-deals-rise-values-fall-in-2025-100137147)). In Europe, White & Case data reported by Private Equity Wire put carve-outs at 90 private equity deals in the first quarter of 2026, 7% of deal volume and 13% of deal value ([Private Equity Wire, 2026](https://www.privateequitywire.co.uk/?p=159578)). Smaller carve-outs mean smaller deal teams and a mid-market operating partner running the separation with no consultants on the customer side.\n\n**The challenges dealmakers name leave customers out.** Asked what makes carve-outs hard, KPMG's respondents named operational disentanglement (52%), valuation complexity (43%), IT and data separation (40%), talent retention (32%) and regulatory hurdles (25%). Customer retention is not among the challenges KPMG reports.\n\n![Bar chart of carve-out challenges dealmakers rate as material, led by operational disentanglement at 52%, with no customer item on the list](/visuals/pe-carve-out-challenges-dealmakers-name/pe-carve-out-challenges-dealmakers-name.png \"Customers are missing from the list. That is the gap this playbook fills. · Source: KPMG, 2026 Global M&A Outlook, March 2026\")\n\n**Yet customers are where the value sits.** Two partners at the advisory firm Stax put it plainly: \"Sales continuity is often the single most important driver of value in a carve-out\" ([Stax, in Financier Worldwide, November 2025](https://www.stax.com/insights/featured-in-financier-worldwide-accelerating-value-in-pe-carve-outs-strategy-and-execution)). Their advice is to secure revenue first and layer in synergies after.\n\n**A change of owner moves customers.** There is no good public number for customer loss in corporate carve-outs, so look at the closest hard evidence. Gallup found that customers of acquired banks left at 8% a year against an industry average of 5%, rising to 10% when the new owner engaged customers less well than the old one ([Gallup, 2015](https://news.gallup.com/businessjournal/183443/bank-mergers-may-chase-away-customers.aspx)). Curinos found that three months after Legal Day 1 of a bank merger, the worst quarter of deals lose more than three and a half times the balances of the best quarter ([Curinos, February 2025](https://www.curinos.com/insights/according-to-the-data-more-m-and-a-horizon-moderate-damage)). Across industries, a study in the Journal of Marketing found customer satisfaction 1.14% lower a year after a merger or acquisition than at comparable firms, with firm value 2.43% lower ([Umashankar, Bahadir and Bharadwaj, Journal of Marketing, 2021](https://phys.org/news/2021-08-firm-mergers-acquisitions.pdf)). Banks are not software or industrial suppliers, but the pattern holds: the same product under a new owner, and the customers who feel the change least are the ones who stay.\n\n**The separation takes a long time.** Forvis Mazars notes that a separation commonly takes six to 18 months depending on IT complexity and the length of the transition services agreement ([Forvis Mazars, February 2026](https://www.forvismazars.us/forsights/2026/02/common-carve-out-transaction-pitfalls-how-to-avoid-them)). Every month of that is a month in which a customer's invoices, support desk or account manager might change.\n\nThe cost of not asking: a Day 1 letter written for the customers you imagine, a top account that quietly moves to the parent's preferred supplier, and a first renewal season that writes down the equity story you just paid for.\n\n## What good looks like\n\nMost carve-out communication plans are written inside the deal team and checked by legal. The best ones are written after someone has asked the customers what they think they are buying and from whom. Here is how experienced operators run it.\n\n![A small deal team around a meeting table with papers and laptops, planning which customers to call before a carve-out closes](https://images.unsplash.com/photo-1714974528693-f77f6fcc56af?crop=entropy&cs=tinysrgb&fit=max&fm=jpg&ixid=M3w5NDUzMDN8MHwxfHNlYXJjaHw2fHxidXNpbmVzcyUyMHRlYW0lMjBtZWV0aW5nJTIwdGFibGUlMjBkb2N1bWVudHN8ZW58MXwwfHx8MTc5NDQ2OTgxMHww&ixlib=rb-4.1.0&q=80&w=1080)\n\n_Photo by [Vitaly Gariev](https://unsplash.com/@silverkblack?utm_source=koji&utm_medium=referral) on [Unsplash](https://unsplash.com/?utm_source=koji&utm_medium=referral)_\n\n**1. Get permission before you get the list.** Before close, you contact the business's customers only with the seller's written agreement, under the clean-team and contact rules your counsel sets. Agree three things with the parent: which customers you may contact, who sends the invitation, and what it may say. Shared customers, who also buy from the parent, need a joint message.\n\n**2. Sample by revenue, then report by revenue.** Fewer than 25 customers: invite all, and have the incoming CEO call the top five personally. 25 to 250 customers: aim for 30 to 50 completed interviews, with every top-20 account invited and at least two contacts at each (the buyer and a daily user). More than 250: 50 to 80 completes, stratified by revenue band and by whether the customer also buys from the parent. Always report each theme twice: by count (\"12 of 41\") and by share of revenue.\n\n**3. Ask about the contract, not the brand.** \"How do you feel about the change?\" gets politeness. \"Who signs your renewal, and what framework or master agreement does this sit under?\" gets facts. Many customers bought through a parent procurement framework they will have to re-qualify for.\n\n**4. Interview before the Day 1 message is final.** The output is not a report for its own sake. It is the evidence behind the Day 1 letter, the account-owner list, and the TSA scope for support and billing.\n\n**5. Run it twice.** Once before Day 1, and again before the transition services agreement ends, when invoices, support and logins actually move. Use the same guide so you can compare.\n\n**Common mistakes**\n\n- Letting the parent's sales team pick who to interview. You hear from the friendly accounts.\n- Announcing the deal in the interview. If the customer has not been told, the interview is not where they find out.\n- Interviewing only the buyer. The users who log in every day decide whether the product feels the same after Day 1.\n- Asking about prices. Keep pricing for a later, separate study, once the new owner has a pricing plan.\n- Reporting only counts. Ten small accounts saying \"fine\" can hide one top-five account planning to leave with the parent.\n\n## The Three Ties test\n\nTo turn interviews into a sort you can act on, listen for what ties each customer to the business. This is Koji's working method for carve-outs, built from standard switching-cost thinking, not an industry standard.\n\n![Framework of the Three Ties test for carve-out customers: the parent tie, the people tie and the product tie, and the three account groups](/visuals/carve-out-three-ties-customer-test/carve-out-three-ties-customer-test.png \"Sort every account by what holds it, before you write the Day 1 letter\")\n\n1. **The parent tie.** They buy because of the parent: a bundle, a master agreement, a procurement framework, the parent's brand, or a single invoice. This tie breaks on Day 1.\n2. **The people tie.** They stay because of named people: an account manager, a support engineer, a sales lead. This tie breaks if those people stay with the parent.\n3. **The product tie.** They stay because the product does a job they cannot easily replace. This tie survives a change of owner.\n\nNote whether each tie was **named unprompted** or only agreed to when asked. Then sort accounts:\n\n- **Product-held:** names the product tie unprompted. Tell them what stays the same.\n- **People-held:** relies on named people. Confirm those people move with the business, and introduce any new owner before Day 1.\n- **Parent-held:** relies mainly on the parent tie. These need a commercial plan (a new agreement, framework re-qualification, a transition offer) before Day 1, not a letter.\n\nThe share of revenue that is parent-held is the number your investment committee should see.\n\n## The timeline, counted back from Day 1\n\n![Process showing the carve-out customer interview timeline from eight weeks before Day 1 to a rerun before the TSA ends](/visuals/carve-out-customer-interview-timeline/carve-out-customer-interview-timeline.png \"Eight weeks back from Day 1, then one more round before TSA exit\")\n\n- **T-8 weeks:** Agree contact rules with the parent. Get the customer list with revenue, products, contract owner, and a flag for customers who also buy from the parent.\n- **T-7 to T-6 weeks:** Write the decisions the study feeds (Day 1 letter, account owners, TSA scope). Build the guide, preview it, run 5 pilot interviews.\n- **T-5 weeks:** Wave 1: every top-20 account and every shared customer, two contacts each, with a personal link.\n- **T-3 weeks:** Wave 2: the rest by revenue band, one reminder each. Close fieldwork at T-2 weeks.\n- **T-2 to T-1 weeks:** Read the report. Sort accounts with the Three Ties test. Cut by count and by revenue.\n- **Day 1 to T+30 days:** Send the Day 1 message built from what customers said. Re-interview a sample of parent-held accounts after it lands.\n- **Six weeks before TSA exit:** Rerun the same guide before billing, support and systems move off the parent.\n\n## How to set it up in Koji, step by step\n\nKoji has PE and VC templates in the dashboard's template gallery. The closest for this job is **Value Creation Monitoring**; Portfolio Company NPS Interviews also works. Start from either and replace the brief with the one in the guide below, or type your research question into \"What do you want to research?\".\n\n1. **Start a study.** Open the dashboard and pick the template, or start from your research question.\n2. **Shape the research brief.** State the decisions (Day 1 letter, account owners, TSA scope), the audience (current customers by revenue band, flagged when they also buy from the parent) and what to avoid. Edit the brief yourself or ask the research assistant to change it.\n3. **Add structured questions where you need a number or a choice.** Koji supports open-ended, scale, single choice, multiple choice, ranking and yes/no questions. Use a single choice for \"who do you think you buy from today\" and a 1 to 10 scale for \"how disruptive would a new invoice and support desk be\", so you can compare accounts and rerun them at TSA exit.\n4. **Set guardrails.** List what the interviewer must never do: mention the sale before the customer does, quote prices, name other customers, or promise anything about the roadmap.\n\n![Koji guardrails settings listing topics the AI interviewer must avoid, such as quoting prices, naming other customers and discussing a pending acquisition](/email/shots/shot-guardrails.jpg \"Guardrails keep the interviewer off the deal, prices and other customers\")\n\n5. **Turn on personal-detail removal.** In the study's privacy settings, remove personal details so names, emails, phone numbers, addresses and employers come out of each transcript when the interview ends. That makes it easier to share findings with the deal team and the parent.\n6. **Choose voice or text, then preview.** Run the preview yourself. Offer both: executives often prefer ten minutes of voice, daily users prefer to type. Participants can pick the interview language on the welcome page, which matters when the carve-out serves several countries.\n7. **Invite the customers.** Import the list from a CSV so each person gets a personal link and you know which account answered. Set a response limit so one band cannot fill the sample. Send it from the address the parent and you agreed on.\n8. **Read the report.** Findings show how many participants raised each theme, with the quotes behind it. Download it as slides or a PDF for the steering committee.\n\n**Plan and cost.** Every step above works on every plan; credits are what you spend. A text interview costs 1 credit and a voice interview 3, and only interviews that pass the quality check use credits. The 10 credits you get at signup cover a pilot. Forty text interviews use 40 credits, forty by voice 120, so a full study needs a paid plan or a credit pack; check the pricing page for current plans. Sending invitations and reminders by email from Koji, and paying participant incentives, need a paid plan. Bring-your-own-key and other data controls are on Enterprise plans.\n\n## Copy this interview guide\n\nPaste this into any tool. In Koji, paste the brief into \"What do you want to research?\" and edit.\n\n**Brief (one line):** _[Business] is becoming a standalone company. We want to understand what each customer relies on us for, what ties them to us today, and what would make them stay or leave when invoices, support and contacts change, so we can plan the move around them._\n\n**Opening line:** \"Thanks for making the time. We are planning some changes to how we work with customers and want to get them right for you. We report themes across customers, not your answers by name. It takes about 10 minutes.\"\n\n| #   | Question                                                          | Type                                                | Follow-up probe                           |\n| --- | ----------------------------------------------------------------- | --------------------------------------------------- | ----------------------------------------- |\n| 1   | What does our product do for your team in a normal week?          | Open                                                | Who else depends on that?                 |\n| 2   | When you think about who you buy this from, who comes to mind?    | Single choice: [Business], [Parent], both, not sure | What makes you say that?                  |\n| 3   | How did you first come to buy it?                                 | Open                                                | Was it part of a wider agreement?         |\n| 4   | Who signs the renewal, and what agreement does it sit under?      | Open                                                | Would that agreement need to change?      |\n| 5   | Who do you contact when something goes wrong?                     | Open                                                | What would happen if that person changed? |\n| 6   | How disruptive would a new invoice, contract and support desk be? | 1 to 10 scale                                       | What would make it a 3 instead?           |\n| 7   | If you had to replace the product in a year, what would you use?  | Open                                                | What would stop you?                      |\n| 8   | What would you need to hear from us to feel confident?            | Open                                                | From whom, and by when?                   |\n\n**Closing question:** \"Is there anything we should know about how you work with us that we have not asked?\"\n\n![Question bank of eight carve-out customer interview questions in three groups: the work today, what ties you to us, what would change your mind](/visuals/carve-out-customer-interview-question-bank/carve-out-customer-interview-question-bank.png \"Eight questions to reuse before Day 1 and again before TSA exit\")\n\n## What you get\n\nA report with themes ranked by how many participants raised them, the quotes behind each theme, and the scale and choice answers as distributions. Each finding links back to the interviews that support it, so when a steering committee member asks \"which accounts said that?\", you can show them.\n\nHow to use it the next day:\n\n- **The account sort.** Every account as product-held, people-held or parent-held, with revenue next to each. This is the page the steering committee spends its time on.\n- **The Day 1 letter.** Written in the customers' words, answering the questions they raised (who to call, what changes on the invoice, what stays the same).\n- **The account-owner list.** People-held accounts get their named contact confirmed, or a personal introduction to the new one, before Day 1.\n- **The TSA scope.** If customers name support and billing as their worry, those services stay under the agreement until the standalone versions are ready.\n- **The commercial plan.** Parent-held accounts get a new agreement or a transition offer, negotiated before Day 1.\n\n## The same playbook elsewhere\n\nThe logic carries across roles. Right after a normal acquisition, an operating partner runs the [100-day customer baseline](/blog/100-day-customer-baseline-pe-playbook); a carve-out is the same idea, started earlier because the customer's supplier is changing too. Before signing, a deal team tests the same risk from outside in [change-of-control customer interviews](/blog/change-of-control-customer-interviews-due-diligence-playbook). A customer success team runs [renewal-risk interviews](/blog/renewal-risk-interviews-customer-success-playbook) one account at a time, which is the right follow-up for any account the Three Ties test marks parent-held. For the integration side of a deal, see the guide to [post-merger integration customer research](/docs/post-merger-integration-customer-research). Later in the hold, the [AI substitution interviews](/blog/ai-substitution-customer-interviews-pe-playbook) reuse the same sample frame.\n\n## Why Koji for carve-out customer interviews\n\n![Table comparing Koji with consultant-led interviews, a customer survey and account team calls for carve-out customer research](/visuals/carve-out-customer-research-methods-compared/carve-out-customer-research-methods-compared.png \"Four ways to hear carve-out customers, compared on what a separation needs\")\n\n| Criterion         | Koji                              | Consultant-led interviews | Customer survey        | Account team calls                 |\n| ----------------- | --------------------------------- | ------------------------- | ---------------------- | ---------------------------------- |\n| Depth             | Open conversation with follow-ups | Deep, trained interviewer | Shallow, fixed answers | Varies by person                   |\n| Time to results   | Days to a few weeks               | Several weeks             | Days                   | Weeks, unrecorded                  |\n| Scale             | Dozens to hundreds                | Tens                      | Hundreds               | As many as the team has time for   |\n| Candour           | Not their account manager         | Independent               | Anonymous but brief    | Low: they are talking to their rep |\n| Consistency       | Same guide for every account      | Varies by interviewer     | Identical              | Different every call               |\n| Rerun at TSA exit | Same guide, same questions        | New engagement            | Easy                   | Hard to compare                    |\n| Analysis effort   | Themes and quotes generated       | Done by the firm          | Manual for open text   | Notes in the CRM, if any           |\n\nA consultant-led study is the right choice for the five or ten accounts that make up most of revenue, where a senior human interviewer adds judgement. A survey is fine for a quick read on awareness. Account team calls matter for the relationship, but they are not research. For the middle ground, dozens of accounts that each need follow-up questions in a three-week window, Koji is built for the job:\n\n1. **Follow-ups on every answer.** \"We'll stay\" gets a \"what would change that?\" every time, which is how parent ties come out.\n2. **Numbers and stories in one interview.** The \"who do you buy from\" choice and the disruption scale become distributions you can rerun at TSA exit.\n3. **Personal links from a CSV.** You know which account answered, so you can weight by revenue and chase thin bands.\n4. **Guardrails.** The interviewer will not mention the deal, prices or other customers.\n5. **Evidence behind every finding.** Each theme shows how many participants raised it, with the quotes. See [tracing report findings to evidence](/docs/tracing-report-findings-to-evidence).\n\n## Questions you might have about AI interviews\n\n**Will customers talk to an AI in the middle of a separation?** Customers answer the business they already work with. The invitation comes from that business (or jointly with the parent), says how long it takes, and offers voice or text. The interviewer asks follow-up questions and listens; it does not sell. See [how AI interviewers work](/docs/how-ai-interviewers-work).\n\n**Can we contact customers before close at all?** Only with the seller's agreement and within the contact rules your counsel sets. Many teams run wave 1 between signing and close with the parent's consent, and the rest after close. Guardrails keep the interviewer off the deal until the customer raises it.\n\n**Where does the data go, and who sees the quotes?** Turn on personal-detail removal so transcripts carry labels instead of names, and agree with the parent who sees the report. Read [AI interview data privacy and security](/docs/ai-interview-data-privacy-security). Teams that need their own model keys can use bring-your-own-key on Enterprise plans ([docs](/docs/bring-your-own-key)). Interview quality is checked, and only conversations that pass use credits ([how the quality gate works](/docs/how-the-quality-gate-works)).\n\n## When Koji may not be the right fit\n\nIf the carve-out has fewer than about 15 customers that matter, call them yourself: each one deserves a senior conversation, and quotas make no sense. If the parent will not agree to any customer contact before close, start the study on Day 1 and use the timeline for the TSA exit instead. And if the customers are public bodies or regulated buyers whose rules require a named human interviewer, use one.\n\n## Metrics to track\n\n- Completion rate by revenue band, against quota\n- Share of top-20 and shared accounts reached, with two contacts each\n- Share of revenue that is product-held, people-held and parent-held\n- Share of customers who think they buy from the parent\n- Median disruption score (1 to 10), before Day 1 and before TSA exit\n- Gross revenue retention for parent-held accounts at the first renewal after Day 1\n\n## FAQ\n\n**How many customers should I interview before a carve-out closes?** Fewer than 25 customers: invite all. 25 to 250: 30 to 50 completes, with every top-20 account and two contacts each. More than 250: 50 to 80, stratified by revenue band and by whether they also buy from the parent.\n\n**When should carve-out customer interviews start?** Eight weeks before Day 1, once the parent has agreed the contact rules. Fieldwork takes about three weeks in two waves, leaving time to change the Day 1 message.\n\n**Should the interview mention the sale?** Not unless the customer has been told. Ask about how they work with the business and what would make them stay; guardrails stop the interviewer raising the deal.\n\n**What is the Three Ties test?** Koji's method for sorting carve-out customers by what holds them: the parent tie, the people tie or the product tie. Parent-held accounts need a commercial plan before Day 1.\n\n**What does it cost?** Text interviews cost 1 credit and voice interviews 3. The signup credits cover a pilot; a full study needs a paid plan or a credit pack.\n\n## Start before the Day 1 letter is written\n\nStart from the Value Creation Monitoring template in the dashboard, or paste the brief above into \"What do you want to research?\". See [Koji for PE and VC](/for/pe-vc) or browse [every PE and VC playbook](/use-cases/pe-vc).\n","category":"Use Cases","lastModified":"2026-11-12T07:50:10.94326+00:00","metaTitle":"Carve-Out Customer Interviews: PE Playbook for Day 1","metaDescription":"Interview carve-out customers before Day 1: who to ask, an 8-week timeline, the Three Ties test and a question guide for PE operating partners.","keywords":["carve-out customer interviews","carve-out customer retention","private equity carve-out Day 1","transition services agreement customers","portfolio separation customers","operating partner playbook"],"aiSummary":"A playbook for private equity operating partners on interviewing the customers of a carve-out before Day 1. It cites KPMG's 2026 Global M&A Outlook (71% of PE dealmakers open to separations, with no customer item among the challenges they rate as material), sets a sampling rule by revenue, gives an eight-week timeline counted back from Day 1, introduces the Three Ties test (parent, people, product) for sorting accounts, includes an eight-question interview guide, and shows the setup in Koji with guardrails, personal-detail removal and personal links.","aiKeywords":["carve-out customer interviews","carve-out Day 1 customers","Three Ties test","private equity operating partner","transition services agreement","customer retention after separation","AI-moderated interviews"],"aiContentType":"guide","faqItems":[{"answer":"Fewer than 25 customers: invite all. 25 to 250: 30 to 50 completes, with every top-20 account and two contacts each. More than 250: 50 to 80, stratified by revenue band and by whether they also buy from the parent.","question":"How many customers should I interview before a carve-out closes?"},{"answer":"Eight weeks before Day 1, once the parent has agreed the contact rules. Fieldwork takes about three weeks in two waves, leaving time to change the Day 1 message.","question":"When should carve-out customer interviews start?"},{"answer":"Not unless the customer has been told. Ask about how they work with the business and what would make them stay; guardrails stop the interviewer raising the deal.","question":"Should the interview mention the sale?"},{"answer":"Koji's method for sorting carve-out customers by what holds them: the parent tie, the people tie or the product tie. Parent-held accounts need a commercial plan before Day 1.","question":"What is the Three Ties test?"},{"answer":"Text interviews cost 1 credit and voice interviews 3. The signup credits cover a pilot; a full study needs a paid plan or a credit pack.","question":"What does it cost?"}],"relatedTopics":["carve-out customer interviews","PE operating partner","portfolio company customer research","Day 1 readiness"]}],"pagination":{"total":1,"returned":1,"offset":0}}