The Cap Table That Paid Everyone but the Staff

What happened

Good Technology raised round after round of preferred stock between 2009 and 2013, growing revenue from $85.3 million to $160.4 million and growing losses alongside it. It filed publicly for an IPO in May 2014, by which time the preference stack owed to investors ahead of common shareholders had reached $267.4 million. Employees held 56.6 million options at a weighted-average exercise price of $2.169, against a last disclosed common value of $3.34. The IPO never priced, an outside appraisal later put common at $0.88 while staff were still buying at $3.34, and in September 2015 BlackBerry bought the company for $425 million in cash.

The case, start to finish

Every dollar of preferred stock raised to cover a loss was another dollar that would be paid out before the staff saw anything.

A company that was working

Good Technology sold mobile security software to large organizations and governments, and by every measure a customer would care about, it was working. Revenue went from $85.3M in 2011 to $160.4M in 2013, and the first nine months of 2014 alone brought $152.7M. Real product, real contracts, real growth.

The losses grew alongside it. $41.0M, then $90.4M, then $118.4M, then another $74.0M in nine months, piling into an accumulated deficit of $682.8M by September 2014. Each of those years was paid for by selling more preferred stock, and here is the part that decides the ending: a dollar of preferred did not merely dilute the common stock employees were being paid in. It stepped in front of it. The queue ahead of common measured $202.8M in December 2013 and $267.4M nine months later, spread across five series.

What it looked like from the inside, in 2014

Picture the offer letter. You are joining a growing company. You are granted options at a weighted-average strike price of $2.169, and the board's most recent valuation of common is $3.88 a share. There are 56.6 million of those options outstanding. Multiply it out and the staff are holding something north of a hundred million dollars, on paper. In May 2014 the company files publicly for an IPO, which is the event everyone has been told converts paper into money.

Nothing in that picture was dishonest, and nothing in it was informative either. The board's fair value is one number. The headline valuation the press repeats is another. Neither is the number that decides what an employee keeps, because that number lives in the cap table, and specifically in the order the cap table pays. Good's own filings showed $38.9M of cash at September 30, 2014 sitting behind $267.4M of preference. Anyone reading those two figures together could see that a sale at almost any plausible price would pay the preferred and then run out.

The clause, and the day someone read it

On September 4, 2015 BlackBerry agreed to buy Good for $425M in cash and lent it $30M under a bridge note the same day. Twelve months after reporting $38.9M in the bank, the company could not reach its own closing without borrowing from the buyer, which is a fair measure of how little room was left. A further $65M of the price went into escrow.

Then the merger agreement did its arithmetic. Series B-1, B-2 and B-3 took a flat $1.00, $4.92 and $4.84 a share. Series C-1 and C-2 took $4.105 a share and then, in the contract's own words, $4.105 plus the per-share amount paid to common. That single word is the whole case. It let the C series draw its investment out first and then queue again alongside the very shares it had just outranked.

Do the subtraction and about $267M of the $425M never travels as far as the common at all. Whatever survives is shared with the C series, then reduced again by transaction expenses and the bridge note. Common was reported at about $0.44 a share, against a board value of $3.88 in June 2014. Set a $2.169 average strike against a residual that thin and the options do not simply lose value. There is nothing left for them to be worth, and an employee who had exercised early and paid tax on a paper gain ended the deal out of pocket.

What to take from it if you are being paid in equity

The transferable idea is not about unicorns. It is about what a claim's position in a queue does to everything standing behind it. Revenue does not change that order. A contract signed at the Series C outranks a business that triples afterward, and by the time anyone wants to renegotiate it, the leverage is gone.

So the questions are small and answerable, and they are worth asking before you sign anything. How much preference stands in front of these shares today? Does that preference also share in whatever is left, or must it choose between taking its money back and converting? What happens if the company sells for less than the last round implied? Ask before you exercise in particular, because exercising turns a free option into a purchase, with a tax bill attached to a gain that may never arrive.

And if you are the founder writing the terms rather than receiving them, notice the asymmetry the September 2015 file makes plain. The people who negotiated the structure were not the people who paid for it, and the structure worked exactly as drafted for everyone who had read it.

Timeline

• 2009–2013 Good raises round after round of redeemable convertible preferred. Revenue grows from $85.3M (2011) to $160.4M (2013). So do the losses: $41.0M, $90.4M, $118.4M. By December 2013 the stated liquidation preference standing in front of common is $202.8M.
• May 2014 Good files publicly for an IPO. Its own filings show why the clock mattered: by September 30, 2014 the preference stack has reached $267.4M across five series, the accumulated deficit is $682.8M, and cash is $38.9M.
• Late 2014 Employees hold 56.6 million options at a weighted-average exercise price of $2.169. The board’s last disclosed common fair value is $3.88 (June 2014). On paper, staff equity is worth well over a hundred million dollars.
• 2015 The IPO never prices. Reporting later described an outside appraisal putting common at $0.88 in June, while employees were still buying stock at $3.34 in August.
• September 4, 2015 BlackBerry agrees to buy Good for $425M cash, and lends it $30M the same day under a bridge note. A company that had $38.9M of cash a year earlier needed a loan from its own acquirer to reach the closing. $65M of the price goes into escrow. The deal closes that autumn.
• Aftermath Common was reported at about $0.44 a share, roughly a tenth of what it had been carried at a year earlier, while the preferred, by contract, took $1.00 to $4.105 a share off the top before common saw a cent, with the C series then sharing in the remainder too. Former stockholders sued in Delaware (In re Good Technology Corporation Stockholder Litigation, C.A. No. 11580-VCL); the claims were resolved privately, through settlement and arbitration.

You're in the owner's chair

You are raising a Series C. One term sheet carries a headline valuation near $1 billion, a 1× preference that also participates in whatever is left, and a discount conversion if you ever list below a set share price. The other offers a visibly lower headline with a plain 1× non-participating preference. Recruiting, press and your own team all run on the headline. What do you sign?

• Take the billion-dollar headline — structure only matters if things go badly
• Take the billion-dollar headline, then refresh the option pool so employees own more shares
• Take the lower headline with the clean, non-participating 1×

What one Good Technology common share was said to be worth

• Feb 2013: board-set fair value (S-1): 4.13 $/share
• Jun 2014: board-set fair value (S-1): 3.88 $/share
• Jun 2015: outside appraisal (reported): 0.88 $/share
• Aug 2015: employees still buying at (reported): 3.34 $/share
• Sep 2015: the merger (reported): 0.44 $/share

The first two figures are from Good’s own SEC filings. The last three were reported by the New York Times in December 2015 and are labelled here as reported, not filed. The August bar is the one to sit with: the price employees were paying, two months after an appraisal said the shares were worth a quarter of it.

Business model

Enterprise mobile security software sold to large organisations and governments, on a mix of recurring subscriptions, perpetual licences, IP licensing and services. It was a real business with real customers and real growth, which is exactly why the ending is instructive. Nothing about the product explains what happened to the people who built it.

Revenue model

Revenue rose every year: $85.3M (2011), $116.6M (2012), $160.4M (2013), $152.7M in the first nine months of 2014 alone. None of it determined what an employee’s share was worth. The certificate of incorporation did that, and it had been written years earlier.

Cost structure

Research and development plus sales and marketing exceeded gross profit in every year disclosed. The losses ($41.0M, $90.4M, $118.4M, then $74.0M in nine months) rolled into an accumulated deficit of $682.8M by September 2014. Every one of those losses was funded by issuing preferred stock, and every dollar of preferred issued added a dollar to the queue standing in front of the common shares employees were being paid in.

Strategic challenge

The number everyone priced their lives on was the wrong number. A headline valuation is the price of the most recent preferred share, which carries rights common does not have. It is not a valuation of the company’s common stock, and at Good the two diverged until they were separated by a factor of ten.

Key decision

The decisive terms were signed at the Series C, years before anyone was thinking about an exit. The merger agreement spells out what they meant. Series B-1, B-2 and B-3 take a flat $1.00, $4.92 and $4.84 a share. Series C-1 and C-2 take $4.105 a share, and then, in the contract’s own words, the per-share amount for each is $4.105 plus the per-share amount paid to common. They collect their money back, and then collect alongside the people whose money they went in front of.

What worked

The preferred stock did exactly what it had been drafted to do, on schedule, without anyone breaking a rule. That is the uncomfortable part. There was no fraud required for this outcome, only a document nobody outside the finance function had read closely, doing arithmetic it had always been going to do.

What failed

Employee equity, both economically and informationally. Run the merger agreement’s own formula (common per share equals the price minus the aggregate liquidation preference, divided by the participating shares) with the last public share counts, and roughly $267M of the $425M is gone before common is reached. What is left gets split with the C series, then reduced by transaction expenses and the $30M bridge note. Options struck at a weighted-average $2.169 against that residual are not worth less; they are worth nothing, and anyone who had already exercised and paid tax on a paper gain finished worse than nothing.

Risk factors

Preference stacking with every round; participation rights that let preferred collect twice; conversion ratchets that make a soft listing worse for common rather than better; a “valuation” that describes one share class; illiquidity that leaves employees unable to price or sell; and tax bills triggered by exercising options on paper gains that later evaporate.

Lesson summary

A startup’s valuation is the price of the last preferred share, not the value of a common one. What decides an employee’s outcome is the waterfall: how much preference sits in front, whether that preference also participates in what is left, and what happens on a down exit. Read the certificate of incorporation before you read the headline, because revenue growth cannot outrun a claim that is senior to you.

Key data

• $267.4M across five series Stated liquidation preference (Sept 30, 2014)
• $425M cash Sale price
• $65M Held in escrow at closing
• $30M Bridge note from the buyer on signing day
• 56.6M at a $2.169 weighted-average strike Employee options outstanding
• $3.88 (Jun 2014) → ~$0.44 (reported) Common per share: board value → reported merger price

Sources & basis

The company here is real and named, and nothing about it was invented to make the story land. The list below is where each fact came from — public filings, court records, published reporting — so you can open a source and check it against the sentence that used it.

1. Good Technology Corporation, Form S-1/A (SEC EDGAR, December 2014) — aggregate liquidation preference of $267,384 thousand at September 30, 2014, series liquidation prices, the participation and IPO-conversion terms, 56,641,595 options at a $2.169 weighted-average exercise price, board-set common fair values, revenue and net loss View source ↗
2. BlackBerry Limited, Form 6-K (SEC EDGAR, 2015) — material change report confirming the US$425,000,000 cash purchase price, and the Agreement and Plan of Merger dated September 4, 2015 defining Merger Consideration, Aggregate Liquidation Preference, Participating Shares, the $65,000,000 Escrow Amount and the $30,000,000 Bridge Note View source ↗
3. In re Good Technology Corporation Stockholder Litigation, C.A. No. 11580-VCL (Delaware, October 27, 2017) — post-merger stockholder litigation, resolved through private settlement and arbitration View source ↗
4. Contemporaneous reporting on the per-share outcome ($4.32 a year earlier, $0.88 June appraisal, $3.34 August employee purchases, $0.44 at the merger), summarising New York Times reporting by Katie Benner, December 2015 View source ↗