The forecast that survived the deal model rarely survives the first two quarters after close. When two companies combine and their CRM and ERP systems stay separate, the operating partner loses the one thing the investment thesis depends on: a clean line of sight from pipeline to booked revenue to cash. Sales runs on one system, finance closes the books in another, and no one can reconcile what was sold against what was invoiced without a spreadsheet and a week of manual work. That gap is not an IT inconvenience. It is a forecast reliability problem, a working-capital problem, and eventually a board-meeting problem.
This guide is for the operating partner, portfolio CFO, or CRO who now owns the decision on CRM and ERP integration post merger. It assumes budget, authority, and a timeline. It skips the definitions and gets to what has to be decided, in what order, and how to tell whether the work is being done well or being padded.
1. Frame the decision as an enterprise-value problem, not a systems project
The mistake that burns the first 90 days is treating integration as a technical migration that IT can scope on its own. It is a revenue-visibility decision with a technical implementation attached. Bain & Company’s annual Global Private Equity Report has consistently tracked how much of the value-creation plan now rests on operational improvement rather than multiple expansion, and management visibility is the precondition for most of it. If leadership cannot see actual versus plan across the combined entity, every other lever gets harder to pull.
Before scoping anything, name the commercial outcomes the integration must produce. Common ones after a merger or add-on:
- One trusted revenue number that finance and sales both stand behind.
- Order-to-cash cycle time that does not degrade during the transition.
- Cross-sell visibility across the combined customer base, which is often part of the thesis.
- Clean handoff from the Transition Services Agreement to a standalone stack before the TSA clock runs out.
If the integration plan cannot be traced back to one of those, it is activity, not outcome. McKinsey’s research on private equity and principal investors and BCG’s work on value creation both keep returning to the same theme: the value is in operating discipline, and discipline starts with measurable objectives, not tool selection.
2. Pick the system of record before anyone touches an integration tool
Every downstream argument dissolves once one question is answered: which system owns which object. The CRM owns the account, contact, opportunity, and forecast. The ERP owns the invoice, the general ledger, inventory, and cash application. The trouble starts where they overlap, on the customer record, the product catalog, and the closed-won-to-order boundary.
Decide the direction of truth for shared objects
For each shared object, one system is authoritative and the other subscribes. Document it in plain language: the CRM creates the customer, the ERP enriches it with billing terms and credit status; the CRM owns the opportunity through closed-won, the ERP owns everything after the order is booked. This is a decision the CFO and CRO make together, not a default the integration vendor picks for them. Get it wrong and you spend the next year resolving duplicate customers by hand.
Do not standardize on a platform mid-merger by accident
Two companies often arrive with two CRMs and two ERPs. The temptation is to consolidate everything at once. Resist it unless the TSA forces the timing. Sequencing the integration, connect first, consolidate later, keeps the revenue engine running while you decide the target architecture. The seven downsides of running a business without a functioning CRM, covered in this breakdown of operating without a CRM, apply doubly during a transition when data is already fragmented.
3. Choose an integration pattern and hold the vendor to it
There are a small number of real patterns, and the right one depends on data volume, latency tolerance, and how long the combined stack has to live.
- Native connector. If both platforms are mainstream (a major CRM and a major ERP), a vetted native or marketplace connector is often the fastest path. Judge it on which objects it actually syncs bidirectionally, not on the logo count on its website.
- Integration platform (iPaaS). When mappings are complex or a third system is involved, a middleware layer buys flexibility. It also buys recurring cost and a new dependency, so make sure that trade is deliberate.
- Point-to-point custom build. Sometimes unavoidable for legacy ERPs, but it becomes the thing no one can maintain after the consultant leaves. Insist on documentation and a support handover as acceptance criteria.
The evaluation discipline here mirrors any technology buy in a portfolio company. The same lens used to buy outsourced RevOps without inheriting a mess applies to integration vendors: scope the deliverable to an outcome, not to hours, and make the acceptance test a business result you can see.
4. Fix the data before you fix the pipes
Connecting two systems that hold dirty data just moves the mess faster. The customer master is where mergers create the most damage, because both entities have their own versions of the same accounts, duplicated, mis-formatted, and coded to different schemes.
Run a data reconciliation before go-live
Match customer records across both systems, resolve duplicates, and agree a single coding scheme for products and revenue categories. This is unglamorous work that the deal team should have flagged during technology due diligence, and if it did, the risk register already tells you how bad it is. If it did not, budget for discovery before you commit to a timeline.
Set the rules for ongoing data governance
Integration is not a one-time event. Decide who owns data quality after go-live, what validation runs on new records, and how conflicts get resolved. Without an owner, the reconciled data drifts back into chaos within a quarter. The AICPA and CIMA have published extensively on the controls finance needs around combined-entity reporting, and those controls only hold if the underlying data is governed.
5. Sequence the work against the real triggers
The calendar that matters is not the project plan, it is the deal calendar. Anchor the integration milestones to events that already carry consequences.
- Day 1 to Day 100. Establish the system of record and a basic sync so revenue reporting works. The broader agenda for this window is laid out in the first 100 days playbook, and integration should be one line item within it, not a standalone effort.
- TSA exit date. If the target ran on the seller’s systems, the standalone CRM and ERP must be live and connected before the agreement ends. This is a hard deadline with penalty clauses, and it drives everything upstream. The Harvard Law School Forum on Corporate Governance has published widely on how TSA scope and duration shape post-close risk.
- First full board reporting cycle. The first board pack after close should show combined-entity numbers from the integrated stack, not a manually stitched deck. Miss this and the board’s confidence in the operating team takes a hit that is hard to recover.
PitchBook and S&P Global Market Intelligence data on deal timelines both underline how compressed the post-close window has become, which is exactly why sequencing against triggers, not against an ideal-world Gantt chart, is the discipline that holds.
6. Judge the delivery on outcomes, not activity
Once the work is underway, the operating partner needs a small set of questions that separate real progress from motion. Ask them at every checkpoint.
- Can finance and sales agree on last month’s revenue number without a manual reconciliation? If not, the integration has not delivered its core purpose regardless of how many objects sync.
- What is the order-to-cash cycle time now versus before the merger? A degradation here signals the integration broke a working process.
- How many duplicate or unresolved customer records remain? Trend this weekly. It should fall, not plateau.
- Is there a named owner for data quality after handover, or does the whole thing depend on the vendor staying?
These are the same instincts that separate a real revenue-operations engagement from a billable one. The framing in what operating partners should actually decide about revenue investments transfers directly: buy the outcome, verify the outcome, and do not confuse a busy dashboard with a working process. Standard CRM measures still apply during and after the transition, and the core CRM metrics worth tracking give you a baseline to hold the combined entity against.
Harvard Business Review‘s body of work on mergers and acquisitions returns repeatedly to integration execution as the deciding factor between deals that hit the model and deals that miss it. The systems layer is where that execution succeeds or quietly fails.
7. A short checklist before you sign off
Run this before approving the integration plan and again before accepting go-live.
- The system of record for each shared object is documented and signed by the CFO and CRO.
- The integration pattern is chosen deliberately, with maintenance and handover as acceptance criteria.
- Customer master data is reconciled before, not after, the systems are connected.
- A named owner holds data governance after the vendor leaves.
- Milestones are pinned to Day 100, the TSA exit date, and the first board cycle.
- Acceptance is tested against one question: can finance and sales agree on the revenue number without manual work.
Getting the systems layer right is one piece of a wider post-close operating agenda. For how CRM and ERP integration sits inside the full value-creation plan, and how to scope it as part of a broader marketing and revenue operations mandate, review the approach to what to buy and how to judge it for portfolio companies.
When the merger integration and revenue-stack work needs an execution partner who scopes to enterprise-value outcomes rather than tickets, review the private equity operating and PMI offer built for portfolio companies.