Definition of an Earnout
An earnout means that if you are a key employee in a company being acquired, some of the total purchase price in the deal gets paid to you over time based on your continued work on the product post-acquisition.  The premise is that some people are passive shareholders who put in cash and don’t have a key role going forward, so they just get paid out at closing.  On the other hand, some team members may be necessary after closing to keep development moving on the product, so their compensation in the deal is linked to continued success of the selling company’s product post-closing.

Keeping the Product Going
Implicit in this formula is the idea that the buyer will keep the seller’s product alive post-closing so that both sides can measure success and the sellers have confidence in their ability to achieve the earnout.  The WSJ gives Disney’s purchase of Club Penguin as an example of this- Disney bought Club Penguin and has kept it as an independent product.  This makes it easy(ier) to measure Club Penguin’s future results and calculate whether earnout metrics have been met.

Compare this with Google’s acquisition of Grand Central.  The product essentially went dark for about two years and then re-emerged as the largely free Google Voice.  I don’t know if the Grand Central acquisition involved an earnout.  If it did, it would certainly be much more difficult to work out success metrics than in the Club Penguin situation.  Google buys a lot of companies for the teams or for product sets that can be worked into larger Google suites, so I can see why they would have trouble measuring earnouts.

Key Considerations
When an earnout is involved in a deal there are several important factors that buyers and sellers need to consider.

-The first is the most objective metrics the deal will permit.  Product revenue is ideal.  It is a pretty easy number to track and works well if the product will be independent post-closing.  If the product will be rolled into a suite of buyer’s products post-closing, or if the seller’s product is free then this metric doesn’t work as well.  On top of that, even if revenue is the main goal we need to think about what would happen if the buyer decides to close down the product line.  Would the earnout accelerate in that situation? Partially accelerate?

-Second is freedom to manage the business.  This usually comes down to “best efforts” language so that the seller has something in writing to say the buyer will put enough resources behind the product to allow the earnout to be met.  It is very hard to make an ironclad promise out of this, since the buyer usually wants freedom to change its business plans and goals as the market requires.

-The third factor is flexibility.  It is impossible to predict what will happen to a product or business, so an ideal earnout will provide enough wiggle room so that if the buyer’s business plans change, the seller can still claim right to payment of some or all of the earnout.

-The last factor is trust.  When a big part of the deal hinges on future performance, buyer and seller both need to have faith that everyone will behave going forward.  The buyer wants to know that the seller personnel won’t make decisions that guarantee their earnout at the expense of the larger business, and the seller needs to believe that the buyer will keep the product alive and manage it in a way that lets the seller achieve the earnout.

Earnouts are usually the most contentious part of M&A deals.  We always do our best to understand the other side’s needs and objectives in the deal, but there is inevitably a leap of faith on both sides that the deal will work out to everyone’s benefit.

To do that, a client and I recently adopted a strategy that I have used informally for a long time, which is to develop a playbook of preferred terms that we incorporate in our form documents so that when I sit down to read a contract I know exactly what I am looking for.

I start my review by comparing the contract against my playbook and making notes on any differences.  After I finish that I usually have a good sense of how the contract is structured, and I can then read through to put the sections together.

On top of the contract-review benefits, having a formal playbook makes it easier to coordinate contract strategy with clients, and also to maintain consistency over time.  When we have a clear sense of what “normal” is, we can develop a set of arguments to support our preferred terms, and also keep track of which deals required us to deviate from our standards.

A contract playbook is a great tool to read difficult contracts quickly, carefully and comprehensively.  I recommend it to anyone who needs to review a lot of documents.

Either way, developers of these products sometimes look for customers among big companies.  It is really common then that the customer’s contracts manager doesn’t quite get the nuances and sends over some kind of Master Services Agreement to document the deal.  Those agreements are generally based on the idea that the vendor is providing a specific, custom deliverable and don’t fit the situation very well.  I have been through this drill a lot and have a few observations about some of the important differences between a custom services and a SaaS deal.

1)  Work for hire language.  A Master Services Agreement will almost always say that the customer owns all technology and works created by the SaaS vendor during the course of performance.  This would be true if the vendor was writing custom software, but is the opposite of what the vendor wants in a SaaS situation.  This should be replaced with something that says the vendor owns all the software and anything developed during the term of the contract, and that the customer has a license to use all of it (subject to payment of all fees).

2)  Service levels.  The Master Services Agreement may not have any service level terms, such as an uptime guarantee or detailed procedures for responding to service errors.  This can work out well for the vendor since it allows more flexibility and avoids difficult conversations about discounts if the service goes down.

3)  Source code escrow.  Some companies feel strongly that if an important vendor goes out of business they should be able to take over the source code to preserve continuity.  With SaaS products these terms are especially inappropriate because the service is hosted- a customer would have to take over the entire service rather than just getting the source code to maintain an installation at its own facility.

4) On-site services.  Master Services Agreements can use a lot of ink on issues like vendor behavior on the customer’s premises and the customer’s ability to replace vendor personnel.  This comes out of the idea that vendor personnel will be in the customer’s data center regularly, which is not correct in a SaaS relationship.  I watch out for language that is really egregious here, but mostly try to leave this stuff alone since it just doesn’t apply very often.

The point of this post is that a Master Services Agreement is the wrong tool for the job in a SaaS deal.  From the vendor’s side, some terms definitely need to be changed, while others are not applicable but can be left mostly alone.  I always try to make the minimum set of changes that will let everyone sign the deal and get the relationship underway, while making sure there is nothing in the agreement that can come back to bite my clients later.

The “moonlighting” problem is pernicious for startup entrepreneurs.  It can take months or years to nurture an idea to the point where it can pay a salary, so it is natural for many people to chip away at their plans while earning a paycheck somewhere else.  The problem is that if their idea is closely related to their day job activities, and if they are not extraordinarily careful to avoid using work resources for the side project, then their employer can claim ownership of the new business ideas.

Jill has a great example up on her blog based on the Bratz dolls.  You should click through to read the whole thing (and subscribe to the blog), but some key facts include:

* $1 billion in claimed damages

* $100 million in legal fees for defendants MGA Entertainment, Inc. and Carter Bryant.

Ouch.

Pricing
This is a fairly obvious one, but will your site’s model be free, freemium or full-fare to users?  Will there be a trial period?  You would get extremely different pictures of what is standard for a news website depending on whether you used WSJ.com or NYTimes.com as a model.

Use and Content Licensing
This point is critical.  Will your users principally browse content on your site?  Will they use it to filter content from other sites?  Will they post a lot of original content? Reblogged content?  Will their information all be publicly available or will users maintain private information?  There are a lot of variables here and good terms of use will tailor themselves to your site’s particular needs.

For example, if your site is social like Flickr, your terms will reflect a higher degree of concern for ownership rights and you will want your users to affirm that they own or have rights to the content they post in very clear terms.  You also want unequivocal rights to display posted content publicly.

At the other end of the spectrum is a site like Mint, where users import private information on their banking and financial records.  You would want your users to grant you a license to incorporate their financial information within their accounts, but you would probably not want a lot of language about public display lest you cause confusion about what information you can/will make public.

Social Networking
Will your users interact on the site?  If so terms regarding a code of conduct are warranted.  Nearly all new web businesses I talk to these days envision some kinds of user interaction, so this is pretty close to “boilerplate” but I beef it up or trim it down depending on the site’s exact model.

Third Parties
If your site relies on advertising, sponsorship or other types of promotional content from third parties you need to explain to users what types of interaction to expect, what types of user information you can give to third parties and when you can/will give it.  This is a touchy area since sites need to develop partnerships and mine value from their user bases, but users are very wary of being spammed.  There is a delicate balance to be found depending on a site’s model.

Governing Law/Disputes
This one varies a bit less from site to site.  The important thing to know is that website TOU are enforceable in court, but fairness is a strong consideration.  If your TOU require users to come to a specific location to bring action against you, be aware that a court may well decide to throw out the term if the value users get is low.  Having a bunch of onerous provisions in your TOU may well lead a judge to look unfavorably on the whole thing as well.  There hasn’t been that much litigation over TOU in the last 15 years so it is hard to draw strict rules here, but the rule of thumb is to be reasonable.

Notice of Changes
The last important point is- what happens when you need to change your terms?  Most TOU say that changes are effective when posted to the site, and there is also case law saying that it is not reasonable to require users to check up on the current TOU continually.  Opinions vary widely on how to manage this, but my own view is that (i) non-material changes are ok to simply post, but (ii) if you need to add significant provisions or change material terms, you should let your users know by email as well.  That is why my practice with TOU is to tailor terms as carefully as possible to a client’s needs, and also leave room for the model to evolve so that we don’t need to revisit the TOU frequently and/or send out notices to users except in rare situations.

This is a complex topic and I have certainly not tried to be comprehensive.  Proper treatment of this subject would also include key terms in a Privacy Policy, since privacy policies and TOU go hand in hand.  This post is long enough already, though.  I will come back to privacy another day.

Percentage Ownership
The founders have figured out an ownership ratio that makes sense to them.  Let’s say there are 3 co-founders, all of whom will be active day-to-day.  One founder will invest $100,000 in seed capital and the others will invest only nominal cash.  The founders have agreed that each of them should get 33.33% of the initial shares.  For simplicity let’s say that each founder gets 1,000,000 shares.

Stock Price
We always want to keep the price of common stock low so that as new employees, co-founders or others come along they can buy stock (or get stock options) at a low price.  I usually like to start with a founder stock price of $0.001 per share.  Stock should always be bought for cash, so we immediately have a problem matching the 1/3-1/3-1/3 ownership ratio with the varying amounts of cash being invested.

Using our hypothetical numbers, Founders A and B are getting 1,000,000 shares at $0.001 per share, which means they need to put in $1,000 each.  If Founder C is buying the same type of stock, also at $0.001, his $100,000 will buy him 100,000,000 shares; he will own 99.99% of the company.

Preferred Stock to the Rescue
My recommendation here is to treat Founder C an investor as well as a sweat equity founder.  By this I mean that we can issue some of his shares as common stock like the other founders, and some of it as preferred stock, which lets us set a higher per share stock price.

Preferred stock is “worth” more because it has rights preferential to common stock.  The rights can vary a lot, and in this case I would provide only a nominal step-up in rights compared to the common stock, so that if our company gets sold Founder C would get his money back before any money is distributed among the common stock holders.  If the company is sold in an extreme fire sale, it is possible that Founder C would be the only one to get any money out, but with luck we will be able to sell this company for more than $100,000.

Stock Repurchase Right
The last important piece here is that all founders should have their sweat equity shares subject to a company repurchase right.  The stock “vests” so that if a founder leaves after a year or two, she only gets to keep the equity she has earned through service.  In my view, all of the common stock should be subject to the repurchase right, but since the preferred stock is purchased for a cash investment, it should not have a repurchase right attached.

In this example, 100% of Founder A and B’s shares are subject to repurchase, but 90% of Founder C’s are not.  This might be the right outcome- or we could adjust the Preferred Stock price and the relative amounts invested for common stock and preferred stock so that Founder C owns more common stock, and has more stock subject to repurchase.  There are a no “right” answers here and it is just a matter of finding the set of conditions that best represents the founders’ relationships.

(Convertible debt means that instead of purchasing stock in the company at $0.xxx per share, an investor buys a promissory note with the intent that when a larger financing happens in the future, the promissory note and accrued interest will convert into the same kind of stock sold in the large financing).

I agree 100% with everything Scott says about convertible debt.  It is simpler to document and it avoids having to put a valuation on the company.   I also know a lot of angels who know how to evaluate great technology, but have no idea how to figure out how much it is worth.  Convertible debt helps get money to entrepreneurs more quickly and with a bit less discussion of valuation.

There are a few gotchas, though.  The biggest one is that if the later financing never happens, or doesn’t raise the minimum amount needed to convert the notes, the company is stuck with a bunch of debt it generally can’t repay.  This becomes awkward.  The investors came on board expecting to end up with equity and instead hold a bunch of near-worthless promissory notes.  The notes themselves come ahead of the founders’ stock in line for repayment, so until the company can find a new source of cash to pay them down, management’s stock is completely worthless and the founders are working 100% for the investors.

Investors would end up with $0 if they foreclosed on their notes in this situation so they tend to be very accommodating, but entrepreneurs still spend a lot of time managing the relationships.

I have seen convertible debt work best as a bridge, where everyone knows the large financing is coming and one of the investors in that deal puts up a little short-term cash to see a company through a tight spot.  In a pure startup situation, taking on convertible debt is a gamble that the big money will present itself.

I talk to clients about their early-stage financing options all the time.  Here’s how I break it down to them in a nutshell:

Debt
Pro: Convertible debt is useful to document small investments quickly and without having to place a valuation on the company in its earliest days. 
Con: If the later financing doesn’t come through, or if it takes longer than expected, entrepreneurs can spend a lot more time managing the company structure than if they had sold stock at the outset.  In the worst case, note holders could force a company to sell and shut down before the founders are ready.

Equity
Con:  Preferred stock requires more paperwork to document.  It also requires that the founders commit to sell __% of the company to the investors, and the valuation can cause trickle-down issues for stock options and other equity incentive compensation, including the dreaded 409A rules. 
Pro: Once done, it’s done.  If founders and investors can agree on the valuation (and the investors can agree to do a stripped-down set of financing terms), I generally recommend going the equity route since then the investors’ situation is locked down and less likely to require time spent managing the equity relationship.

Good post, Scott.

First, those who don’t follow legal news should be aware that 2008-09 has been brutal for the legal profession. The best summary I can find is this chart, showing well over 100,000 jobs lost at big law firms since the start of 2008.

(image courtesy lawshucks.com)

Most recently, the entire US economy lost 11,000 jobs in November, of which 2,900 were in the legal profession.  As a friend put it, the legal industry is “shrinking faster than any that I can think of“.

Into this environment steps Jason Calacanis, who last week announced plans for his Open Angel Forum.  The forum is a great idea and I certainly wish him well with it.  Here is the part that rubs me the wrong way, though:

The organization is charging a small group of service providers $1,500 each to attend.

I certainly understand the concept here.  My old firm used to run an event we called Angel Law Forum for entrepreneurs and service providers would routinely outnumber entrepreneurs and investors.

To say that lawyers can easily afford it, though, is comical for two reasons:

1)  Easily is the wrong word. Lawyers (and other service providers) are struggling along with everyone else this year;

2) Jason doesn’t seem to have worked through his own math.  The only way any service provider can afford a $1,500 seat at a networking dinner is by charging high rates to clients.

As I said in my comment on the OAF’s announcement (awaiting approval as of this writing), I hope the event goes superbly and I look forward to hearing great things about it. In the meantime, I will be attending cheapie networking events so that I can keep my rates low.

The Good – Voting Control
This type of structure allows founders to retain a lot of control over a company even as it grows enormously.  To use a very simple example, as founder of a company I could issue 1,000,000 shares to myself.  If the company grows rapidly and takes on a lot of employees I might want to issue 3,000,000 shares to them.

When I do that, I incur economic dilution and voting dilution.  This blog won’t speak to the economic side at all and will just assume the money all works out.  On the voting side, though, you can see that I went from controlling 100% of the stock to a mere 25%.  I have lost most of my control over the company.

I could issue more shares to myself to bring my percentage ownership back up, but there are a bunch of tax and logistical issues that make that hard to do.

Instead, I could convert my shares into supervoting stock.  If my shares all vote 10:1 and the other 3M shares are 1:1, then I keep control of the company because my 1M shares have 10M votes compared to everyone else’s 3M.

Google has said it adopted this system before going public because it wanted to think long-term and the founders did not want to risk being outvoted on shareholder matters.  B corporations can use similar structures to ensure that their companies’ core social missions can not be easily stripped away.

The Bad -Investors Hate it
My experience with these structures is that investors have a hard time getting comfortable.  Professional investors want to know that they can influence major decisions by the company, so it takes a while to get folks comfortable with the idea that founders have super-special voting rights.

My advice to entrepreneurs is not to rock the boat.  There are lots of perfectly good, non-investor-threatening reasons to create supervoting stock, but if the setup makes it harder to raise money then it’s a big gamble for a startup.

These things work for Facebook and Google because those are well-established, already successful companies.  They have negotiating leverage on their side.  Most startups are not so fortunate.

Related articles by Zemanta

• Facebook introduces dual-class structure for stock (financialpost.com)
• A New IPO-Style Stock Structure for Facebook, But No Official IPO Plans (insidefacebook.com)

A few times a year I help people unwind difficult situations as well.  These are much harder deals to do- emotions run high and frequently the two sides don’t trust each other at all.  There is a tendency to look at these disputes as zero-sum situations: one side needs to lose something for the other to benefit.

Occasionally that is right.  I worked on a transaction a while ago where there was only 1 substantive issue between the two sides- one guy thought he was owed a bunch of money and the other thought the amount was much lower.  In the end neither side got what they really wanted out of it, which was unfortunate.  The deal later fell apart and I believe it was mostly because neither side got enough value from the agreement.

Most disputes don’t work that way.  Usually each side has something it doesn’t really need but the other side wants, and vice versa.  My job as counsel is (i) to help my clients figure out what the other side really wants and how to give it to them (this was a great quote from Hiten Shah- if only I could find the tweet), and (ii) help my clients figure out what they can give in order to get the deal done, move on and stop talking to me as much.   Some of the “give” items are material; frequently there are intangible items as well like reputational benefit and helping someone exit a bad situation gracefully.

So how do I help clients figure these things out? This post has been stuck in draft form all week because this is the hardest thing to do.  I talk to my clients a lot about all the facts in the situation, we use legal arguments as a backstop to think about what the result would be in litigation, and we spend a bunch of time assessing the other side’s needs.

We then make a series of offers intended to reflect our needs, the value we place on the things we want from the other side and the items we think the other side wants.  If we are lucky, the other side has done the same thing and we can find a small patch of common ground in short order.  From there, the thoughtful back-and-forth process lets us feel each other out and eventually find the key terms to avoid a zero-sum equation.

In other words, the alternative to a zero sum equation is probably something like: 40% part pragmatics (esp. how much money everyone is willing to spend on lawyers), 30% legal knowledge, 25% psychology, 10% intuition.  The rest is luck.