How Corporates Can Keep a Startup Collaboration From Turning Into Free Consulting

Before you read: This is general educational information, not legal, procurement, investment, accounting, or tax advice. Contract scope, intellectual property, confidentiality, payment, and acceptance terms require case-specific review by the responsible parties.

A collaboration starts drifting when the corporate keeps receiving custom work but still cannot say what it has committed, what result it will accept, or what decision comes next.

The risk is not only that the startup works without payment. The corporate also loses decision quality: requirements keep changing, no result is conclusive, and the strongest teams become less willing to engage.

Three signals that the project needs a reset

1. Commitment remains vague

Budget is one visible form of commitment, but not the only one. A corporate can also commit an accountable owner, protected staff time, approved data access, a testing site, or a scheduled decision gate.

The warning sign is that every commitment remains personal and verbal. The sponsor is enthusiastic, but no unit owns the result and no internal process has begun.

2. Success and stop criteria are missing

Without written criteria, each review meeting can create a new definition of “good.” The startup produces another version; the corporate collects more material; neither side gets a decision.

Useful criteria identify the question, evidence, threshold, constraints, and decision date. They also define when a weak path should stop.

3. Scope grows without a new decision

Learning can justify changing scope. Informal expansion is different.

If a test for one workflow becomes a dashboard, an integration, and a second use case without new approval, timing, or commercial discussion, the project has moved beyond discovery. Both parties should pause and redesign it.

Separate discovery from delivery

A short discovery phase can be reasonable even when it is unpaid. It should be reciprocal, bounded, and designed to answer whether a formal project is worth creating.

The boundary should be visible before the startup allocates significant engineering time.

A corporate sponsor’s reset conversation

When a project is drifting, the sponsor can bring both sides back to five questions:

1. What decision are we trying to make? 2. What work is inside the current phase? 3. What evidence will be sufficient? 4. Who owns the decision inside the corporate? 5. On what date will we proceed, redesign, or stop?

If the requested work no longer fits the current phase, treat it as a scope change. The options may include reducing the request, designing a funded project, changing the evidence method, or closing the exploration.

Closing is a valid outcome. A clear “not now” with reasons is more useful than another month of ambiguous activity.

The counterview: not every free contribution is exploitation

However, a strict “never share anything without payment” rule can also block useful discovery.

A startup may choose to demonstrate an existing product, discuss a reusable architecture, or analyze a small anonymized sample because the learning and relationship are valuable. A corporate may be unable to open a procurement process until feasibility is clearer.

The test is reciprocity and boundedness. Is each side learning? Is the work reusable? Is there a limit? Is a real decision scheduled? If yes, a small exploratory contribution may be rational. If the work is custom, cumulative, and open-ended, the boundary has failed.

Protect the startup and the corporate at the same time

For the startup, the main risks are engineering distraction, disclosure, uncompensated customization, and a case that cannot be reused. For the corporate, the risks are weak evidence, unclear ownership, reputational damage, and a pilot that cannot move into operations.

The same controls help both sides:

• one accountable owner on each side;
• the minimum data and disclosure needed for the current phase;
• written success and stop criteria;
• a clear method for approving scope changes;
• a scheduled decision, including the option to stop.

A practical decision rule

Do not judge collaboration health by meeting frequency or praise. Judge it by mutual commitment, decision clarity, and controlled scope.

If two or more of those are missing, pause new work. Reset the phase, create an appropriate formal project, or close the exploration. Continuing without a boundary is not momentum.

For related paths, see How to design a corporate PoC and PoC, procurement, strategic investment, or M&A.

Sources and scope

• BCG — A Framework for Deep-Tech Collaboration
• McKinsey — Three Essentials of Successful Corporate Venture Capital

The appropriate process depends on industry, organization, risk, and applicable rules. Confirm contract, payment, procurement, data, and intellectual-property arrangements with qualified internal and external professionals.

Further reading

• Corporate & CVC learning path
• Why corporates work with startups
• How to design a corporate PoC