The M&A process is broken

Atlassian is built on the belief that there are better ways to get work done. If we find a way of working to be ineffective, we will change it. This belief is fundamental to how we build and sell our products, and now we’re applying it to how we acquire companies (aka mergers and acquisitions, or “M&A”).

M&A is a key part of our strategy – over our history, we’ve acquired more than 20 companies for approximately $1 billion, including Trello, Opsgenie, and AgileCraft. And one thing has become very clear to us about the M&A process – it’s outdated, inefficient, and unnecessarily combative, with too much time and energy spent negotiating deal terms and not enough on what matters most: building great products together and delivering more customer value.

We want to change that.

We’ve studied data from hundreds of technology acquisitions from the past 5+ years. The data makes two things clear:

1. Big buyers are bullies: Large tech companies exert negotiating power over founders and selling companies because they can, and stack the deck in their favor with buyer-favorable terms.
2. Big buyers get more protection than they need: Many of these terms are unnecessarily one-sided. For instance, buyers hold back a much higher portion of the purchase price to cover potential liabilities than they need.

Here’s why this matters: these one-sided terms are introduced at the outset of the M&A process and require a disproportionate amount of time and goodwill to resolve. This needlessly introduces friction and mistrust, making the far more important task – getting our people and products to work well together – more challenging.

We know there’s a better way to do this.

In an effort to reduce this unnecessary friction and increase trust, we’re doing something that, to our knowledge, no company has done before: we’ve crafted a new M&A term sheet and we’re making it public. The Atlassian Term Sheet is more favorable to selling companies than any we’ve seen among strategic acquirers in technology. It’s guided by our research and includes explanations to make our approach more understandable. We’ve done this because we want to be fair to our future team members and we don’t want to spend energy and goodwill on things that almost never matter. And we believe that by being fair and transparent, we can make the M&A process more efficient, human, and aligned with why we acquire companies (and the incredible founders and teams behind them) in the first place.

Why be open?

The M&A process can be confusing and intimidating, especially for founders who are selling a company for the first time. Term sheets can read like a different language, and when founders ask buyers why a term is the way it is, buyers often say things like “this is market” or “this is our form” (i.e., this is standard, it’s just how we do things). The reality is that because of the leverage that many buyers exert over sellers, certain “market” terms have evolved to buyers’ advantage, even though, based on the data, it’s simply not necessary.

We believe we can cut some of the friction by focusing on the terms that actually matter to us, not the terms we can get just because we’re bigger than the companies we acquire. The Atlassian Term Sheet has the protections Atlassian needs and reflects the values we live by. It’s comprehensive — we don’t want to hide anything during term sheet negotiations. And it includes explanations to help you understand why we care about these terms. And now, it’s public.

Why bother sharing the Atlassian Term Sheet publicly? The answer is simple: we want to approach M&A with the same open spirit with which we offer our products. We publish prices and standard terms for our products on our website, which we believe promotes trust and makes customers more comfortable with our low-touch sales model. We don’t negotiate these prices and terms, so our customers know they’re being treated fairly. This has resulted in one of the most efficient enterprise software sales models in the world, which has allowed us to invest more in R&D and deliver more customer value.

Now, with the Atlassian Term Sheet, we’re providing tremendous visibility into our approach to M&A. We hope it gives founders a sense of what to expect during initial deal discussions and confidence that we’re treating them equitably. And similar to our sales model, we believe this transparency will enable all parties to spend more time on the things that matter most: building great products together and delivering more customer value.

Our term sheet is fair and informed by data, so instead of spending time and emotional energy negotiating things like indemnity, we can focus on things like diligence and integration planning, organizational structure (including founders’ roles), and retention strategies. By beginning with terms that are more favorable to sellers and avoiding needless contention, we’ll put more effort into shaping what success looks like after the actual acquisition and laying the groundwork to get it done.

Being open also demonstrates the trust that we’re placing in you, the founder (and future leader at Atlassian). We want to do things more collaboratively and need you to help us. This is no longer about either side putting points on the board, or satisfying ego –  it’s about getting this done right for both of us and focusing on what matters most from the start.

Want to geek out on M&A deal terms and the data that drives our approach? If so, read on! Keep in mind that the next few sections assume familiarity with M&A terminology, so if it’s confusing, check out the annotations in our term sheet for more explanation.

What does the data say?

When you actually look into it, many of the key “market terms” established by strategic tech acquirers are unnecessarily protective of buyers, at the expense of founders, the selling company, and its shareholders.

Beyond key business points like purchase price, founders’ roles within Atlassian, stock vs. cash consideration, and retention equity, there are a range of other terms that are vital to understand. Some of the most important ones — and often the most heated negotiations — are around indemnification, escrow, and risk allocation, so we focused a lot on those in our research. We looked at data from hundreds of deals with a focus on technology and software. Here’s some of what we found:

Escrows funds are much bigger than necessary

• In software transactions, the median escrow is 10% of the purchase price and gets up to 20% at the higher end.
• In the data set we examined for software deals, the substantial majority of transactions had no indemnification claims made at all, and there was no single transaction where the entire escrow fund was exhausted.
• Even when claims are made, they rarely come close to exhausting the escrow fund. We found that even in the worst case scenarios, amounts returned to buyers out of escrow still amounted to an estimated 2% of the total purchase price.

What does this mean? Buyers are holding back way more escrow than they need. While there are prominent examples of M&A deals going very wrong, buyers tend to focus on these worst case scenarios or edge cases that rarely materialize.

IP and privacy don’t need to be special reps

• In nearly 60% of tech transactions, IP and privacy are  “special reps” that aren’t capped at escrow and survive for longer than the escrow period. In other words, a buyer can go after a seller’s shareholders after the closing of the deal for more than the escrow amount if the buyer suffers damages from an IP or privacy breach, and it has a longer period than the escrow period in which to do so.
• Of technology deals that have IP and privacy as a special rep, almost 50% of acquirers protect themselves with between 25-50% of the purchase price, and almost 20% protect themselves up to the full purchase price.

What does this mean? In the event of an IP or privacy claim, if the amount of the claim exceeds the escrow, then a buyer can then go after selling shareholders for that excess amount. This is a hard one for a seller to stomach, because money that has been deposited into an employee’s bank account and spent, perhaps for a down payment on a house, could be clawed back by the buyer. Buyers do this because of the leverage they exert, but the data indicates that this is absolutely unnecessary and we found no examples in the data we reviewed of buyers actually going beyond escrow to recover additional amounts for IP or privacy claims.

Our terms

Based on the data, we landed on terms that we believe in. There’s a lot more in the Atlassian Term Sheet, but we’ll highlight a few of the most important terms here. The bottom line is we’re focused on getting away from the typical nickel-and-diming and trying to cover every potential edge case. We want to approach the M&A process in the most efficient and fair way we can.

Indemnity, escrow, and insurance: Choose your own adventure*! We’ll leave the choice to you: either provide a 5% escrow for 15 months or pay for a buy-side rep and warranty insurance policy and provide a 1% escrow for 15 months (insurance will cover the remaining 4%). Either choice is far better than what other strategic acquirers are offering in today’s market. The 15-month escrow period is shorter than the market median and enough time for us to get through an audit cycle, which is when the majority of claims (if any) shake out. Most important, outside of fundamental reps and items that are within sellers’ control, such as covenants, or serious issues that we discover in diligence that result in special indemnities (a rare occurrence), you’ll have no exposure beyond the escrow. We’re taking on that incremental risk.

* As long as your transaction is over $50M. If the price is under $50M it’s not cost-effective to get insurance, and we’ll go the 5% escrow route.

IP and privacy: IP and privacy are not special reps, and are capped at escrow. This means we can’t claim more than the escrow for IP and privacy claims, and once the escrow period has passed, we can’t go after a seller at all for IP and privacy claims. Put differently, you will bear only up to 1% (or 5%, if you don’t choose the insurance route) of the cost of damages for IP and privacy rep breaches. Atlassian/insurance would be on the hook for the rest.

It’s worth emphasizing how different our approach is versus other strategic acquirers. It has become common practice among tech acquirers to have protection for IP and privacy claims above and beyond the escrow which, in our experience, can lead to a ton of heartburn for founders and endless cycles of negotiation. So we’re not asking for that because the data tells us we don’t need it, and we feel confident in our thorough diligence process.

Fraud: We’ll protect ourselves against fraud up to the purchase price. It’s unlikely that companies we work with will commit fraud, but this is really about values for us. Bottom line: don’t commit fraud.

Documentation: Regardless of your choice of 5% escrow vs. 1% escrow with insurance, the legal agreements will read essentially the same. We expect comprehensive reps & warranties along with fulsome disclosure schedules as part of the transaction process.

Let’s do this!

Selling a company is always emotional, especially for the founders who care so deeply about what they have built. We get that we won’t always see eye to eye, and getting to a result that works for everyone will continue to be challenging and require flexibility. We understand that the diligence process is grueling for founders, a reality that can’t be avoided because it’s the only way for us to get to know their company and prepare for a successful integration. But even if we can’t eliminate all the tussling and late nights that are the hallmarks of M&A, we believe our open approach can substantially improve the process and allow all of us to focus on the things that actually matter. And because this is Atlassian, we’re excited to try.