Acquihires Explained: How Big Tech Actually Buys Teams

An acquihire is a transaction in which a company acquires a startup primarily to hire its team, not to buy its product or business. It is increasingly used as a modern alternative to classic M&A: faster to close, lighter on diligence, and built for situations where the people are the asset. But acquihire is not a single legal form. It is a deal outcome that can be reached through several different transaction structures, each with its own mechanics, tax profile, and risks.

This article breaks down how acquihires are structured in practice, what founders get paid, and what to watch out for before signing.

TL;DR

• An acquihire is when a company buys a startup to hire its team, with the product being secondary or irrelevant to the deal.
• Acquihires can be structured as IP assignments paired with employment agreements, as subsidiary arrangements, or as hire-and-license deals.
• Compensation typically combines a signing bonus, equity in the acquirer, and an IP assignment fee.
• Acquihires and full M&A transactions work very differently, and treating them the same leads to bad deals.

What Is an Acquihire?

An acquihire is when a company acquires a startup primarily to bring the team on board, not to integrate the product or scale the business. The focus is on the people: their skills, their cohesion, their momentum. 

In the classic version of the deal, the product gets shelved or wound down. But that is not always the case. Sometimes the acquirer wants both the team and what they built, and the deal is structured accordingly. What makes it an acquihire rather than a standard acquisition is the logic driving it: the team is the core value of the deal.

‍How Is an Acquihire Structured?

There is no single acquihire agreement. The term describes a deal outcome, not a legal form, and in practice it is assembled from a few different transaction structures. Which one applies depends on how the company is set up, what the acquirer actually wants, and how clean the founders need the exit to be.

Structure 1: IP Assignment + Employment + Dissolution

This structure works best when the product exists but the business around it hasn’t taken off. Maybe the team is still early in development. Maybe the product serves a niche audience that was never going to scale into a standalone company. Either way, the IP is real and worth capturing, but a full acquisition doesn’t make commercial sense for the buyer.

Within this structure, the founding team assigns all relevant intellectual property to the acquirer (code, trademarks, domain, pending patents) and signs employment agreements with the acquiring company. The startup entity is then wound down and dissolved. Payment flows directly to the founders as individuals, typically split between a signing bonus and an IP assignment fee.

Why dissolution rather than keeping the entity alive? Because an empty company with no staff is a liability that stays with the founders. Outstanding debt, stale contracts, and ongoing maintenance costs like registered agent fees, filing obligations, and accounting don’t disappear when everyone leaves. Dissolution draws a hard line and gives founders a clean exit.

Structure 2: Subsidiary + Employment

A less common but tactically useful alternative, particularly when founders want separation between themselves and the transaction, or when the acquirer wants to keep the acquired entity intact for operational or branding reasons.

Here, instead of dissolving the startup, the acquirer takes it over as a wholly owned subsidiary. The team continues operating under the existing legal entity, at least for a transitional period, and payment can flow to the company rather than directly to individuals. This can simplify how compensation is categorized and reported, and in some cases gives founders more flexibility in structuring their personal tax position.

It also lets the acquirer preserve contracts, regulatory licenses, or brand equity tied to the original entity that would otherwise be lost in a dissolution.

Structure 3: Hire-and-License

This is the structure that has defined the largest acquihires of the past two years, particularly in AI, and it works quite differently from the two above.

Rather than purchasing the company or its assets outright, the acquirer hires the founders and core team and signs a non-exclusive license to the startup’s IP. The startup itself remains nominally alive. The Microsoft-Inflection deal became the template: a $620 million non-exclusive licensing fee for Inflection’s models, plus a $30 million payment to waive legal claims tied to mass hiring. Google’s Character.AI deal followed the same logic, with founders rejoining Google under DeepMind alongside roughly 30 researchers through a $2.7 billion non-exclusive technology license.

The appeal for large acquirers is regulatory. A full acquisition of a competitor triggers merger review. A licensing deal combined with direct hiring often does not, at least not yet. EU and US regulators are now openly scrutinizing these structures, especially where teams and licensed IP move together, so this window may not stay open indefinitely.

For founders, the tradeoff is real. The structure can generate a significant licensing fee paid to the company, but employees outside the core team and investors often see little from it. The Windsurf deal illustrated this clearly: Google’s acquisition focused on the founders and technology license, initially leaving the broader employee base with uncertain equity outcomes. Only a subsequent acquisition of Windsurf’s remaining assets by Cognition secured value for the rest of the team.

Real-World Case Studies

The past two years have seen a wave of high-profile AI acquihires that show these structures in action, especially the licensing-plus-hiring model.

Microsoft + Inflection AI (March 2024)

Microsoft hired nearly the entire 70-person team from Inflection, including co-founders Mustafa Suleyman (who became CEO of Microsoft AI) and Karén Simonyan. Instead of a traditional acquisition, Microsoft paid roughly $650 million ($620 million for a non-exclusive license to Inflection’s models plus about $30 million tied to hiring). The structure helped Microsoft avoid full merger review while quickly strengthening its AI efforts. Inflection’s product (the Pi chatbot) was largely sidelined. Investors reportedly achieved modest 1–1.5x returns through the licensing fee paid to the company. This deal became the blueprint for modern mega-acquihires.

Google + Character.AI (August 2024)

Google brought back co-founders Noam Shazeer and Daniel De Freitas (former DeepMind researchers) along with roughly 30 researchers. The deal involved a reported $2.7 billion non-exclusive technology license. Character.AI remained independent and continued focusing on post-training capabilities with the rest of the team. This “reverse acquihire” of ex-Google talent showed how licensing deals can efficiently move top researchers while avoiding traditional M&A scrutiny.

Google + Windsurf (July 2025)

In a deal valued at $2.4 billion, Google hired Windsurf’s CEO Varun Mohan, co-founder Douglas Chen, and key researchers to advance agentic coding at DeepMind. Google received a non-exclusive license to the technology but took no equity stake or control of the company. The remaining 200-plus employees stayed behind. Windsurf’s core business was later acquired by Cognition AI. This case clearly illustrated the founder-and-core-team focus of many acquihires: key talent and IP moved, while the broader team and the entity often did not.

Amazon + Adept (June 2024)

Amazon hired most of Adept’s technical team (about 80 percent, including CEO David Luan) and signed a licensing deal for its agentic AI technology. This followed the same licensing-plus-hiring playbook and further showed Big Tech’s preference for targeted talent grabs over full acquisitions.

These deals highlight both the power and the tradeoffs of acquihires: massive upside for founders and key talent, regulatory advantages for buyers, but often limited or delayed benefits for the broader team and early investors.

How Are Founders Compensated in an Acquihire?

The compensation package in an acquihire is usually made up of three components.

Signing Bonus

A signing bonus is a cash payment, often delivered at or shortly after closing, tied to the founders and key team members signing employment agreements and completing the IP assignment. Think of it as a transaction fee structured as a salary advance.

The size varies considerably, from a few months of salary to several years of projected compensation, depending on how much the acquirer wants the team. There is no market standard. It is a negotiation.

Equity in the Acquirer

Most acquihires include equity grants in the acquiring company through options, RSUs, or direct stock. These serve a retention function: they give the incoming team a financial stake in what they are now building.

The specifics matter more than the headline number. What is the vesting schedule? Is there a cliff? Does the equity accelerate under any conditions, like if the team is let go within the first two years? These are the questions that determine whether the equity is worth holding out for.

IP Assignment Fee

Sometimes structured separately, the IP assignment fee is a payment made specifically in consideration for the intellectual property being transferred. This matters significantly for tax purposes, since compensation income is taxed differently from proceeds on the sale of property. Getting this allocation right and properly documented protects both sides.

What Should Founders Watch Out For?

Acquihires are structurally simple by design, which can create a false sense that the legal work is minimal. It is not. Here are the three areas that most often cause problems.

Fair Vesting Terms

When founders join the acquiring company, they receive equity as part of the compensation package. What they need to check is whether the vesting terms offered are in line with market standard.

The benchmark is a 4-year vesting schedule with a 1-year cliff: nothing vests in the first year, then the remainder vests monthly or quarterly over the following three. Some deals use a shortened version, typically 2-year vesting with a 6-month cliff, particularly when the team being brought in is senior or the acquirer wants to signal commitment. Both are reasonable. Anything that deviates significantly from these structures, whether through longer vesting periods, steeper cliffs, or unusual conditions on vesting events, should be treated as a red flag and negotiated.

Also pay close attention to acceleration provisions in the equity grant from the acquirer. Single trigger means your unvested equity in the acquiring company vests automatically if that company is acquired or goes through a qualifying liquidity event. Double trigger requires two things to happen: the liquidity event and a second condition, usually the founder being terminated without cause shortly after.

Single trigger is more founder-friendly: if the acquirer gets bought or hits a liquidity event, your equity vests automatically. Double trigger is what acquirers typically push for, since it requires both the liquidity event and a termination without cause before acceleration kicks in. It protects the acquirer from a scenario where founders cash out on a sale and have no incentive to stay through the transition.

Tax Consequences

The way a payment is labeled affects how it is taxed, and in an acquihire the allocation across payment types matters more than in most deals.

For the acquired team, signing bonuses are ordinary income, taxed at the highest marginal rate. An IP assignment fee may qualify as proceeds from the sale of a capital asset, which in many jurisdictions means a significantly lower rate. If the startup has been operating for more than a year and the IP was developed over time, founders may be able to treat a meaningful portion of their total payout as a capital gain rather than income. That difference can be substantial.

For the acquirer, the structure also matters. Employment costs are generally deductible as business expenses in the year they are paid. IP transferred in a purchase transaction can often be capitalized and amortized over time, creating a long-term tax benefit that a straight hiring arrangement would not provide. In the US, Section 197 intangibles acquired in a purchase transaction can be amortized over 15 years.

There is also a broader efficiency argument. A full acquisition often triggers complex tax events: goodwill calculations, purchase price allocations across multiple asset classes, potential recognition of gain by selling shareholders. An acquihire, particularly one structured as an asset purchase limited to IP and employment, can sidestep much of that complexity while still delivering the tax benefits of a documented transfer.

The allocations have to be documented, agreed upon by both parties, and reported consistently. Mismatches between how the buyer and seller characterize the same payment are a common audit trigger. Getting a tax advisor involved before the term sheet is signed is what separates founders who maximize their take-home from those who figure out what they actually received six months later.

Acquihire vs. M&A: What Is the Difference?

Founders sometimes arrive at an acquihire negotiation expecting it to function like a merger. It does not.

The speed and simplicity that make acquihires attractive also mean that founders have less leverage and fewer legal protections than in a traditional deal. Knowing that going in lets you negotiate for what matters most: equity terms, job security, or a clean break for the company you are leaving behind.

Final Thoughts

An acquihire can be an excellent outcome for the right team, with the right acquirer, at the right moment. But the legal structure is not a formality. The choices made around transaction type, compensation packaging, vesting mechanics, and closure terms determine whether the deal is a real win or a long regret.

If you are navigating one on either side of the table, get counsel involved early. The deal may be simple. The documents rarely are.