The Truth About Convertible Debt at Startups and The Hidden Terms You Didn’t Understand

Posted on Sep 5, 2012 | 21 comments

The Truth About Convertible Debt at Startups and The Hidden Terms You Didn’t Understand

This article initially appeared on TechCrunch – with a minor update highlighted in red below.

Ah. We’re back to discussing convertible debt again. This time by the efforts of Adeo Ressi to introduce a new kind of structure called “convertible equity.”

I applaud all efforts by people to take on this issue and especially be Adeo who – let’s be honest – was really the first champion of trying to make the VC world more transparent by launching TheFunded, which didn’t exactly endear him to VCs initially.

My initial reaction to Adeo when we spoke was that while it may have solved some issues (debt versus equity) it didn’t solve the ones that I’ve been warning entrepreneurs about most loudly. And frankly I feel he overstates the potential harm of convertible debt, which if properly structured acts mostly as an equity instrument anyways.

To better understand the arguments for / against convertible equity I suggest you read my posts on those topics:

So let me weigh in more loudly than in the past. I can’t say it much simpler than this:

“What if I took some of the worst, most egregious terms in a standard term sheet and made them the defacto standard in most convertible debt deals?

If I could persuade you that they’re already in these documents would you consider abandoning this structure?

Because convertible debt deals often have both a ‘full ratchet’ and often have ‘multiple liquidation preferences’ “

Yup. That’s right. Those terms you fought so hard to get out of “clean term sheets” by using TheFunded, VentureHacks and the like are in your current documents.

But how?

When convertible debt first started being introduced as a “faster, cheaper way to get startups funded” they didn’t have pricing built into them. A standard entrepreneur retort I heard back then (2008-09) was “I don’t know what my company is worth now. By doing a convertible note we can delay the discussion until we figure out how big this is going to be.”

And some seed stage investors told me, “I prefer not to fight over price now. I just want into the hottest deals. That’s all that matters. They’ll get priced soon enough by a VC.”

I think both points-of-view were a bit naive.

What the entrepreneurs were really saying is, “I don’t want to take a lower valuation now, while I don’t have customers or a full team. I want to use investor money to build these things. Then those investors – the ones who took the most risks (i.e. my friends, family & former colleagues) – get to pay a higher price later when a VC comes in a prices the round.”

Sure, you gave them a 15-20% discount. But that’s hardly fair compensation when your former cube mate gave you $25,000 of money she didn’t really have to invest in you, took tons of risks with her money, and now has to pay a VC price for that money a year after she invested it.

What investor would put money into a company and then agree a price later based on success? Can you imagine investing in the stock market where your price was determined at a future date and the better that company performed the HIGHER the price you paid for that investment. In fact, most early investor work hard to help their startups get to the next level so it makes no sense for the angel investor and founders to be at odds.

Investors call Bull Cap.

Enter “the cap.” What this did is set the maximum price of the deal. As in, “your money into my company will convert at a 15-20% discount to the next round of capital I raise with a maximum price of $8 million pre-money valuation (or whatever the cap was).”

And I’ve been banging my head against the wall ever since.

Convertible debt with no cap is stupid for investors. Convertible debt WITH a cap is stupid for founders.

With a cap means that every person who wrote you a check assumed that they were going to pay the cap. So if I write you a $500,000 check into a convertible note with a $4.5 million cap I am assuming when I write the check that I will own 10% of your company. If I didn’t assume this I shouldn’t write the check because I have to get involved knowing that I might pay that price.

But entrepreneurs – convertible notes have no MINIMUM! So you’re taking all of the pricing risk. This has worked very well in the 2009-2012 time frame because the tech market has boomed in this period. But many convertible-debt companies are starting to feel that pinch now. I’m starting to hear it more often. And then the market does slow down you’re going to hear an entire generation of convertible-debt companies moan.

Let me explain it more clearly in equity terms.

In a standard VC term sheet there is a standard term called an “anti dilution provision” and they are in nearly 100% of deals. What it says in human terms is, “if you raise money after me at a cheaper price than I paid I get a discount on the investment that I’m making now.”

Full Rat-shits

The key for entrepreneurs to understand is whether it’s a “full ratchet” or a “weighted average ratchet.” In a full ratchet it means that if Investor A paid $2 per share for your stock and you later raise money at $1  per share (because that’s the only price at which you could find investors) then Investor A’s entire holding is converted to this lower price – even if it happens a year later.

So Investor A might have bought 20% of your company in 2012 and in 2013 with no addition money invested suddenly owns 40% of your company. POOF.

You rarely find full ratchets in early-stage deals any more. They are either a sign of naive founders or naive investors (because they will quickly develop a reputation in the market and because future investors in that company will all demand your terms, which therefore will include a full ratchet).

They can occur legitimately in a company whose valuation skyrockets unexpectedly as in, “your last round was at $50 million and you want to raise 9 months later at $750 million. Fine. I’ll pay the price but I want a full ratchet.”

The logic being that the current investors and founders have more inside knowledge of the company performance and dynamics than a brand new investor and thus if the new investor is going to “pay up” they shouldn’t take all of the pricing risk in the deal. I’m not arguing for full ratchets – just explaining them.

Weighted-average ratchets (broad-based or narrow) only convert a portion of Investor A’s stock to the new price and the formula that dictates this is based upon how much new money comes in at a lower price and how low that price is. It has nowhere near the same dilutive effects as a full ratchet except in extreme edge cases.

If you really want a deep dive on anti dilution better that you go read Brad Feld’s post on the topic.

How Convertible Debt with a Cap Behaves Like  Full Ratchet

Now how does this apply to convertible debt with a cap?

A new investor in your round is saying, “I’ll agree to pay the $5m, but if you raise at $2.5m I want 100% of my stock to convert at that lower price.” So my $500k doesn’t buy 10%, it buys 20%. Voila. Just like that. Convertible notes have full ratchets.

But wait! There’s more!

Hey, bud. I don’t just want a full ratchet. I want that 20% discount you promised me!

Seriously, dude?

Yup! That was part of our deal. Convertible Notes felt good when prices only went up, didn’t they? Now that it’s 2014 and prices have gone down it doesn’t feel so good.

So that $500k I took from you not only didn’t convert at $5m but it converted lower than $2.5m? So it’s worse than a full ratchet?


In fact, that’s what that 20% discount we agreed in your note was all about. You didn’t think it was only a discount if your price went up, did you?



Convertible Notes Also Can Have Multiple Liquidation Preferences

Hey, Mark. Are there any other things I should worry about in a convertible note? I mean, I had no idea that they had full ratchets. I thought we got rid of that shit in 2003?

Well, they also have a term that is one of the most hated by entrepreneurs. But since it’s not spelled out on the convertible note, most entrepreneurs don’t even understand that the note potentially has this term. Convertible notes often have multiple liquidation preferences.

That’s not possible. Nobody I know accepts multiple liquidation preferences in early-stage deals. And every seed stage VC tells me they don’t ever even ask for them.

If they do a convertible note they do. Even though many I’ve talked to don’t even realize it since it’s hidden.

But how? What the …

Well … let’s take a $500k convertible note at a $4.5m cap. So I get 10% of your company at worst. Our convertible note says that it “converts into the next round of capital and into the same security.”

So when you raise that up round on your Series A of $3 million at $12 million pre (that investor got 20% of the company and has a 1x liquidation preference) my stock converts into that same security. And since I only paid $5 million post money your lawyer is likely going to issue me with almost 3x the number of shares as if I bought them in the Series A to make up for the price difference.

Why is that a problem? I agreed to the lower price when you funded me – so I’m cool with that.

But my $500k, while only buying 10% of the company (and now diluted down to 7.5% after the new investor came in), has nearly $1.5 million of liquidation preferences. That a 3x liquidation preference in disguise. It’s what you fought about for all these years.

Damn. How come nobody ever told me?

Dunno. Maybe because it’s on small dollar investments. But I’ll be damned if convertible notes aren’t getting larger and larger.

[update: Adeo thought that I hadn’t quite explained easily enough why this creates a 3x liquidation preference so I agreed to expand. Only read this red bit if you didn’t quite follow the argument above.

If somebody gives you money under a convertible debt note at a $2.5m valuation and another person funds you with convertible debt at $5m valuation (high resolution financing) and your equity round finally closes at a $10 million valuation … what technically happens?

The most straightforward way to do the deal and what most people do is to issue the first investor 4 times more shares than the ultimate equity investor to adjust for the 4x discount in price (ie if I give you 4x the shares it’s the same as though you paid 25% of the price for the shares). The second investor gets 2x more shares (50% discount). So in the end they all end up with Series A stock priced at the exact same price (say, $1.5 per share) but investor 1 & 2 have more stock than investor 3. And if I weren’t simplifying the math, I’d show you that the first investors get even more shares since they have a discount to the next round.

So as your initial investor at a $2.5m they now have 4x the stock and thus 4x the liquidation preferences (since each share has liquidation preferences on it).

The alternative is to give investors 1,2 & 3 the exact same amount of preferred Series A stock and give investors 1 & 2 more common stock (which doesn’t have liquidation preferences) to adjust for the discount. But investors 1 & 2 won’t be happy with this because when they bought the convertible note they were expected to get preferred stock. So they’ll feel cheated.

The third option is the one I listed below, which is what I have done.]

Is there way around this?

Yeah. I’m sure there are several. What I’ve done in the past is convert the notes into their own classes that only have 1x liquidation preferences. I’ve done deals that had many different price points (since the company brought in convertible notes over time “high resolution financing” and price kept going up with the new notes) and so we’ve had a class for each note.

So we ended up funding the Series A. But there is a Series AA-1, Series AA-2, Series AA-3 and so on. We all vote together as a single class of preferred stock but each Series has its own price in order to prevent multiple liquidation preferences.

Damn. That’s complicated.

Yup. And what a joy it was for me to have to explain to the many convertible note investors that came before me.

And it makes it harder to explain to future investors but I’m sure you’ll get over that if your business does well.  Those legal fees you saved in your last round suddenly went up in the Series A round. But at least you had more money to pay them.

But what about this debt thing? Adeo says that debt is bad for me.

It might be. I’m not an expert in that but I don’t personally see it as a bad thing in the way he does. I don’t see it as a ticking time bomb.  For the most part convertible debt doesn’t act like debt it acts like equity. A good convertible debt round round answers the questions:

1. What happens at maturity?
I recommend that startups agree the “conversion price” at maturity. As in, you’re giving me $500k at a 20% discount and a $5m cap. If we don’t raise a bone fide round of capital (say, $1.5 million from new investors) within one year, your money will convert at a $3m pre-money (i.e. something lower than the cap – this amount to be negotiated).

But won’t that signal to new investors my price? No. No more than the cap will. When you talk to new investors about the mandatory conversion at the end of the debt’s life you say, “that was the downside, worst-case provision I agreed to investors in case we weren’t successful raising money.”

2. What happens in a sale or acqui-hire?
I recommend that you agree a multiple of investment to be paid to investors in this case. I normally recommend 2x but this is negotiable and up to you. This feature in a note says that if I gave you $25,000 and you sold the company without raising more money or before maturity (i.e. my note didn’t convert) then I get $50,000 out of the proceeds of the sale.

At a minimum every angel investor should ask for this. Otherwise it’s a sucker investment. You give money and if the company sells you get your investment back? That would suck.

And I recommend that companies give this to investors in a convertible note because it’s the fair thing to do.

It basically IS a 2x liquidation preference but that investor doesn’t own equity in your company so they deserve a return and the liquidation preference automatically goes away when you do your equity financing round.

Are there any advantages to a convertible debt deal? Surely there must be some !!?!!

Yes. Of course. But not the ones you’ve been sold on. Mostly you were told that they are: Cheaper, faster and you can more easily price people at different prices.

They are not necessarily cheaper. With the many “series seed” documents that exist now you can often get a law firm to do your docs for $5k – even when it’s equity. They have to believe you’re going to be successful because they won’t make money at this price but they will do it.

They are not necessarily faster. I’ve done series seed, early-stage deals in 8 days. So you can’t tell me debt is faster.

And I’ve explained why the multiple price points causes problems on liquidation preferences. But equity can also have multiple price points either be giving investors warrants or by doing the Series AA-1, Series AA-2, etc. structure I described above.

So to me the real advantages of convertible debt are:

1. They often don’t have control provisions. With equity often investors want “blocking rights on a sale or future finaning” that they often don’t get in a convertible debt deal. If this is the reason you’re doing it, then perhaps talk to investors about whether they’d be willing to give up that right in a Series Seed equity deal.

2. They often don’t have board seats attached to them. Again, this should be negotiable with a Series Seed.

There may be more (feel free to add in the comments section if I missed some) but those are the main ones.

So if you’re obsessed with having convertible notes I’ve been telling people,

“Don’t do convertible debt with a cap, do convertible debt with a price.”

Think about it. If you’re going to agree the maximum price you might as well agree a minimum price, too. Which means you’re agreeing the actual price. That gets rid of the ratchet. At that poing you might as well have done equity but if for reasons of speed, cost and control you prefer not to – at a minimum I’d ask for price security.

Some of the smartest early-stage lawyers have weighed in and to my reading don’t seem to be too far off from my position:

Ted Wang (Fenwick & West) – here.

Yokum Taku (Wilson Sonsini) – here.

And Dave Young from Cooley who read a draft and made sure I wasn’t completely off base in my views. He assures me I’m not. He agrees with my theses and tells me that these issues are all the chatter amongst law firms who work with early-stage tech companies.

  • terrycojones

    Hey Mark.

    Thanks for taking the time to write all this up – I learned some things, so much appreciated.

    I know you’re not fussed about typos etc., but a couple that would make things clearer (I think):
    s/that investor got 25% of the company/that investor got 20% of the company/
    s/but investor A & B have more stock than investor B/but investor A & B have more stock than investor C/

  • chrissheehan

    Mark, one other issue to put on the table – pro rata rights. In most con debt deals there are no pro rata rights (ive seen them occasionally but its cumbersome). For larger VC funds investing in the note, this is not such a big deal as they will have the inside running on the A round. But for angel investors that do follow-ons (many do; some don’t) not having a pro rata right places them in a weaker negotiating position in the A round, particularly if its a competitive deal where VC funds want to own as much as possible. This can put the founder in a difficult position.

  • Jess Bachman

    Holy crap Mark, if there was ever a post that warranted an infographic… I may just make one using your numbers.

  • msuster

    If you want to do an infographic I can help you with actual numbers

  • msuster

    Thanks for the addition. Yes, that’s an important point.

  • Antone Johnson

    Great post, Mark. I guess my take on this is that, with apologies to Churchill, “Convertible notes are the worst form of seed-stage financing for deals under $1MM — except for all the other forms.” Your observations are astute and take the analysis to a whole new level, yet many of the terms suck for either investors or entrepreneurs, or pose risk for one or the other in one scenario or another. (To take one example, I would argue that submitting startups to a full ratchet on the downside, which sucks for entrepreneurs, is more than offset by a valuation cap on the upside, which penalizes them for achieving spectacular business success that third-party investors recognize in the form of dramatically higher valuation. IMHO, valuation caps should be unnecessary if the conversion discount is set high enough.) That’s pretty much the case for any type of financing structure.

    I hate to be such a cynic, but “convertible equity” (really a contractual right to acquire equity) strikes me as a solution in search of a problem. So do Series Seed and its ilk, except the problem there is real (large law firms doubling their fees over the past decade for delivering the same product), but the cure can cause more problems than the disease. I don’t think pushing everyone into priced equity rounds is a solution. Convertible notes are a quick-and-dirty hack that ends up being “good enough” in most cases for the earliest stage companies doing the smallest deals.

  • Joe Wallin

    Mark, great post. One thing that the convertible security remedies or fixes that you didn’t talk about was the tax problem of accruing interest. If the investors don’t really want debt–they don’t want to be on top of the equity as debt, they want equity but don’t know how to price it yet, so they are doing a convertible note until the pricing can be figured out–if they take convertible debt they typically wind up with being taxed on interest income when they never really wanted debt in the first place. (Unlike the principal amount, the interest cannot be converted to stock tax free; this might sound like a trifle, but I’ve seen the interest income pile up, and if you have foreign investors there can be mandatory tax withholding…). The convertible security problem solves that problem. Still, it strikes me that many investors might still want to have convertible debt because their rights as debt holders are more defined than convertible security holders.

  • brgardner

    You forgot the tax benefits to Founders for doing a convertible debt. As a founder if I do an 83b election, then I only pay taxes on a very low valuation of my founders shares at the time I am issued my founder shares. You will be able to justify your low price to the IRS, if they audit, early on assuming you have little to no revenue. However, if I do an equity round shortly after the issuance of founders stock, and then declare my stock price much higher, the IRS is going to flag. And who really wants to be flagged by the IRS. For me a big reason for founders to do convertible note early on is if they think their company is going to be hugely successfull and wants to take the risk of paying taxes on their issued stock early on.

  • LD Eakman

    A good one. Even LPs should understand the securities held in underlying funds. I don’t understand not pricing it now, too much conflict can occur later. It’s bound to make one side feel the fool.

  • Matt Bartus

    A well-advised company will have issued the founder stock well in excess of an equity round, so this is not typically an issue.

  • Joe Wallin

    I agree with you on this Matt. Usually this timing issue does not arise.

  • brgardner

    Not necessarily. It all depends on how you value the company.

  • Mike Prozan

    Mark’s analysis is excellent and holds depending on the faith one puts in the conversion at term price.

    I’m with you, Antone, though I take it a step further. The whole point here is that the round is not priced. Given that the investor has full ratchet, the investor gets full protection on the downside, save for total failure. Seems to me, the debt without out cap is just the mirror image of that. The number they pick for conversion at term is really just a guess. Mark seems to put a lot of faith in it.

    Likewise with liquidation preference. If selling above the conversion at term price, capped or uncapped, begets an enhanced liquidation preference, then it seems that selling below the mandatory conversion price reduces the liquidation preference.

  • Yokum

    Hi Mark – great post. One tweak to convertible debt that is floating out there is the concept that the note be converted into both preferred stock and common stock. The original principal amount (and simple interest) gets converted into preferred at the same price as the cash investors in the round. The remaining number of shares that needs to be issued (or the discount element) is converted into common stock. This was a tweak that is in the form of convertible equity documents that Adeo released and is in the current form of note that YC companies use.

    My previous thoughts about this type of tweak are in this post.

  • msuster

    thanks, yokum. yes, I’ve seen that done as well, which I mentioned in the red section of the post above. that solution works. provided that the note holder agrees to take common as their discount. certainly not what they thought they were getting when they invested, though.

  • Matt C.

    This was my thought as well – an infographic would be worth a thousand words here.

  • Hector

    Mark, long time not seen! I see you are doing great (from an old-time once competitor). Convertible vs Priced is a local US-fad, truly non-global, care to check convertible vs priced elsewhere than US? Latter wins by a landslide. Eventually will become a non-issue.

  • Peter Cranstone

    Here’s the truth.

    The guy with the gold makes the rules. Nothing else matters. Convertible notes are no more egregious than VC term sheets. Each is an instrument to reward the investor. Some investors are smarter than others, some not so.
    In the end the only thing that counts is that your build a sustainable business which rewards ALL the stakeholders.
    I’ve read every document i can on raising money, what works and what doesn’t. Ultimately it boils down to execution and a commitment by ALL to succeed. Along the way there will be bumps, some big (beware the ‘Hell’s Angel or VC’) it’s up to everyone to work through those issues.
    After reading this post i’m surprised entrepreneurs even bother getting out of bed. All that effort and you still have to deal with the VC (if you can even get a meeting with them).
    Remember if VC’s and Angels had all the answers then their IRR’s would be a whole lot better than they are. In the end think of them merely as one of the parts of the vehicle to move your product from concept to reality.
    Because in the end the only thing that counts is the customer, not the Angel, not the Investor or the deal that ‘You’ negotiated.

  • Mark Essel

    This would be a great supporting tool, please do Jess.

  • Bo Sartain

    What is wrong with using warrants instead of a conversion
    discount? Warrants have fallen out of
    favor, but why? The numbers work out
    better for the issuer with warrants, better for the investor with a conversion

    Example: take a $100K investment. Let’s say the next round price is $1.00 per
    share. If there is a 20% conversion
    discount, investor gets 125,000 shares.
    If there is 20% warrant coverage, investor gets 20,000 warrant shares, so 120,000 total. (these calcs exclude the interest on the note).

    Use of warrants can also help avoid the multiple liquidation
    preference problem. The note converts dollar for
    dollar into the equity sold in the qualified financing. The company shouldn’t have to create multiple
    tiers of preferred to accommodate the note-holders. Investors have to pay (or net exercise) for
    their warrant shares, which will normally be preferred.

    I’m not sure why we don’t see more convertible note and
    warrant deals anymore.

  • Joe Wallin

    Bo, I think one reason warrants have fallen out of favor is that if you issue a note and a warrant together they are considered together one investment unit–and under the federal income tax law you have to allocate some of the purchase price to the warrant based on Black-Scholes or some binomial pricing method–and then you have OID in the note…and the OID has to be reported to the investor has income. It is a mess. You can read more about this issue here: