CHAPTER 27

DoorDash, San Francisco, December 8, 2020

If the pandemic undercut Airbnb’s plans to pursue a nontraditional listing, it added fuel to those of DoorDash, another company in the gig economy that was pursuing its IPO in parallel.

Unlike Airbnb, DoorDash was perfectly positioned to seize on the surge in orders from diners too fearful of infection to venture to local grocery stores or restaurants, or prevented from doing so by local ordinances. Diners provided a tailwind to the company’s growth prospects. DoorDash would have to convince investors that the growth rates it exhibited during the pandemic months would persist even after the widespread adoption of vaccines made it safe again to dine at restaurants.

Founded in 2013 by Stanford University students Tony Xu, Stanley Tang, Andy Fang, and Evan Moore to deliver restaurant food around Palo Alto, before the pandemic, DoorDash got locked in a fierce battle with competitors like Grubhub, Postmates, and Uber Eats in what was a capital-intensive business.

DoorDash had found a key advantage: the U.S. suburbs, where higher growth markets sat untapped and competition was less intense. The strategy was initially lampooned because urban markets were thought to offer an easier path to profitability due to greater density in both customers and drivers, which the company called “Dashers.” Xu disregarded common wisdom and went ahead with his own idiosyncratic plan, signing exclusive deals with national chains, and pushing through an effort to broaden its reach in second- and third-tier cities across America and Canada. Even before the pandemic, DoorDash’s growth had surged.

With growth came an almost insatiable demand for cash. Over one mind-bending stretch from March 2018 through November 2019, DoorDash raised almost $2 billion from venture capitalists and other crossover investors who dabbled in both public and private markets. Its valuation soared during that period from $1.4 billion to $13 billion.

The man responsible for that fundraising and the design of the company’s upcoming IPO was DoorDash’s CFO, Prabir Adarkar. Adarkar had an engineering degree from the prestigious Indian Institute of Technology system, and an impressive résumé. He was a former McKinsey & Co. project manager, Goldman Sachs TMT banker, and senior finance executive at Uber, one of the world’s largest consumers of corporate capital. Adarkar left Uber after three years to join DoorDash in 2018. In a July announcement that year, Xu offered a sharp assessment of his latest hire. “Having spent over 40 hours in the past two months with Prabir, I can tell you that he’s got a sharp mind, possesses an owner’s mentality, and leads from the front.”

Adarkar had a keen interest in the design of DoorDash’s public listing, and his engineer’s mind liked to deconstruct the process. Like Airbnb, DoorDash considered a direct listing. It quickly ran into the key shortcoming of the direct listing: that it couldn’t help a company raise capital. Food delivery was capital-intensive, and executives knew that if growth continued, they would need more money.

That could be done within the confines of the direct listing process if DoorDash wanted to wait. The SEC needed to approve the NYSE’s rule change allowing a capital raise concurrent with a direct listing. Then the company would have to haggle over any remaining details. DoorDash executives held talks with their bankers at Goldman Sachs and JPMorgan, coming to realize that it would likely take another eighteen to twenty-four months if they wanted to go that route.

That was too long. Adarkar didn’t want to waste the opportunity to go public in what looked at the end of 2019 to be a robust market for IPOs. Who knew what the markets would look like in two years’ time? Put simply, the risk/reward calculus ruled out a direct listing. One person close to the process summed it up as a “science experiment” that departed too far from the company’s primary goal, which was to raise money. Nonetheless, Adarkar and Xu were interested in exploring alternatives to the traditional offering and had the support of a board that included Alfred Lin and John Doerr. On February 13, 2020, DoorDash filed documents confidentially with the SEC. Disclosed later, the prospectus showed Goldman Sachs in the lead-left position, followed by JPMorgan. A week later, the company raised $340 million in convertible debt. And on February 27, DoorDash issued a press release stating that it had filed its draft prospectus.

The WHO’s pandemic declaration would come less than two weeks later. DoorDash set aside its IPO plans to focus on doing what it could to keep its restaurant partners open for business and its drivers safe.

In the first few months of the pandemic, DoorDash’s IPO plans remained on the back burner as executives concentrated on meeting the orders surging into its app and doing whatever it could to help its restaurant clients meet the uptick in delivery orders. In April, as the full impact of the pandemic had begun to sink in, DoorDash accounted for 45 percent of all third-party delivery orders in the United States, according to Edison Trends, a market research firm that analyzes anonymous e-receipts from consumers. That was up from 28 percent in March 2019, when the company took the lead from Grubhub.

In June, the company used that momentum to raise $400 million, turning to investors including Durable Capital Partners, run by former T. Rowe Price portfolio manager Henry Ellenbogen; Fidelity; and T. Rowe Price. The investors purchased shares in DoorDash that valued it at $16 billion. The money helped DoorDash start a Storefront product to help restaurants create their own websites for taking delivery orders.

By July, DoorDash executives felt confident enough in the state of the business and its partners to dust off its IPO plans. It submitted a new draft registration statement to the SEC, responding to the questions the agency had outlined in a letter it sent to the company on March 12, one day after the WHO’s declaration of a pandemic.

DoorDash’s financial condition served as a strong foundation for a management team considered by at least one banker to be one of the smartest in Silicon Valley when it came to raising money and accessing the capital markets. The need for cash had driven the company through many rounds of fundraising, but Adarkar’s analytical background and curiosity about market mechanics kept the company focused on designing a bespoke IPO process.

On August 3, Adarkar moved to shore up his team, bringing in Andy Hargreaves as head of investor relations. Hargreaves had known Adarkar and Keith Yandell, DoorDash’s chief business and legal officer, for several years. He was already in frequent contact with the company. Hargreaves was a long-time securities analyst and had recently been doing contract work for Sands Capital, a small investor in DoorDash.

Hargreaves immediately got to work fine-tuning the draft registration statement that would become the public prospectus shared with investors. He started working with Goldman Sachs and JPMorgan bankers to draft materials DoorDash would share with investors, making a list of which ones the company wanted to meet and how often.

This would be the first IPO process for Hargreaves, much as it was for Adarkar. Like Adarkar, Hargreaves favored an analytical approach to thinking through DoorDash’s particular objectives. While the direct listing was off the table, the executives could still thank Spotify for showing that the time had come to question the status quo.

DoorDash executives weren’t sure what to make of the system. They wanted to do whatever they could to maximize the proceeds of their IPO by selling shares for as high a price as they could persuade investors to pay. The company also wanted to be able to pick its shareholders—due to its fundraising activities, it already had relationships with some of the world’s largest IPO investors. Management wanted to ensure it could allocate shares to those investors who had been along for the ride and continued to act as strong supporters.

But they also realized that there were parts of the process that didn’t make sense in 2020, particularly around the greenshoe option, the lockup provisions that Unity had already challenged, and the common wisdom that shares should be priced at between $10 and $25. Airbnb had also briefly considered doing away with the greenshoe, but ultimately kept it.

As DoorDash’s management team met with its bankers and its board, they spent hours evaluating the greenshoe. The option was a common part of the traditional underwriting agreement and something that other companies had simply taken for granted. Pioneered in 1963, when the Green Shoe Manufacturing Company held a secondary offering underwritten by Paine, Webber, Jackson & Curtis, the feature allowed underwriters to sell more shares, typically 15 percent more, than those the company planned to sell in the IPO. This could be done if demand from investors dramatically exceeded supply. The additional supply from the greenshoe could be used to dampen volatility.

It also acted as an insurance contract of sorts. By selling extra shares, the banks were effectively short the stock and needed to buy shares to balance the position. If the shares rose after the IPO, the banks typically had thirty days after the issue to purchase the additional shares from the company. But if the shares fell below the IPO price, the banks could buy the shares they needed in the secondary market, thereby providing demand that would serve to prop up the share price.

Critics liked to suggest that on popular IPOs the greenshoe was easy money for the banks. If the stock price surged in trading, the banks could buy the stock from the company at the IPO price and sell it at the much higher trading price, pocketing the difference in profit.

Jamie McGurk, who had left Andreessen for a senior position at Coatue Management and advised DoorDash when it was considering a direct listing, summed it up. “Investment banks make a lot of money on the greenshoe,” McGurk said. “It’s a very lucrative part of the IPO for them, and the reality is if you have a robust market and a well-priced and well-opened IPO, you don’t need a greenshoe. Maybe on a particularly volatile day, or a set of circumstances where something’s mispriced, it’s a useful tool to have, but it’s almost like when things didn’t go well that’s when the greenshoe is actually useful. Otherwise, it’s kind of an extra incentive for the underwriter.”

The board, including director Alfred Lin, had a dim view of the greenshoe. DoorDash executives examined the data and found that it seldom acted as the shock absorber its supporters suggested. In cases where the stock declined in price, their research found that the extra buying power it created often wasn’t enough to prevent the shares from continuing to decline. On an average IPO, only 40 percent to 50 percent of the shares are available to trade on the first day. Buying an extra 15 percent was often not enough. And if things went well and the stock price increased, the banks would benefit at the expense of the company, which was contractually obligated to sell the shares cheap even though they were trading at higher prices in the market.

As Adarkar asked around, he found that many large investors didn’t care about the greenshoe because they were reasonably confident that they would still get the allocations they wanted without it. In other words, the extra shares are typically shared with less important investors lower down in the order book.

While DoorDash executives spent a considerable amount of time understanding the greenshoe, they also considered eliminating or adapting the customary 180-day lockup. Many practitioners blamed it for artificially limiting the supply of shares in the market. The DoorDash executives wanted to explore ways to reduce those limitations.

In considering the lockup, the executives pulled data on average daily trading volume for shares of other recent IPOs. They considered the potential effects on stock price volatility and liquidity of unlocking some percentage of the shares at various points in time. They weighed those quantitative factors with more qualitative factors, such as considering the rights of old and new investors, the asymmetry of information possessed by both types of investors, and the traditional thinking that new investors should get two quarters of earnings results before old investors could cash out. The results of the analysis were far from conclusive.

When Adarkar sounded out investors about the lockup, he received a very different answer than the one he had received about the greenshoe. Investors enjoyed the lockup provision because it protected them from a surge in supply hitting the market and gave their investment time to find a stable trading price. They didn’t want to see it eliminated.

DoorDash executives were more focused on the influence of a lockup on the first-day trading performance than they were on the aspect that Unity executives worried most about, the fact that it disadvantaged insiders. DoorDash executives knew that the relative lack of supply on the first day of trading was to blame for some of the first-day pop, as Andreessen partners Alex Rampell and Scott Kupor explained in their August blog post. The selling that accompanied the end of the lockup often led to volatile price swings and a depressed stock price. Some institutional investors had even discovered that on a popular transaction, they could ride the stock price higher, sell for a quick gain, and then buy back the shares after the lockup expiration pushed the price down.

In the end, DoorDash decided on a two-stage lockup provision, giving employees and early investors two chances to sell stock and relieving some of the pent-up supply pressures. They’d get the first chance at least 90 days after the IPO, as long as a list of other conditions had been met. The management team and directors could sell 20 percent, while all other equity holders could sell as much as 40 percent. And then everyone would get a chance to sell their remaining shares after the customary 180 days.

The executives also decided against splitting DoorDash’s shares in a way that would have brought the IPO price into a range of $10 to $25. The bankers told the executives that lower prices tended to attract more retail interest than higher prices, but DoorDash also knew that retail investors brought volatility. The executives ultimately did split the shares. But, like Google sixteen years before, DoorDash decided against splitting the stock in a way that artificially courted retail investors with a lower price.

With those final details ironed out, DoorDash publicly filed its IPO prospectus on November 13. The document listed Goldman Sachs, JPMorgan, and ten other investment banks as co-managers or other members of the selling group. The bankers could now begin talking to investors and collecting orders for the upcoming IPO.

By this point, Adarkar had witnessed Unity’s IPO and the stock’s modest 30 percent gain on the first day of trading. Armed with that success, Goldman was touting its role in developing the new blind order-entry system and offering it to any client that wanted to use it.

Adarkar reached out to Unity CFO Kim Jabal to get a better understanding of the process. He learned that one of Unity’s goals was to run a meritocratic and democratic process, an objective DoorDash executives weren’t opposed to, but one that wasn’t high on their list. They wanted to make sure they could sell the stock at a high price and avoid dilution.

As the company mulled Goldman’s order-entry system, it began its roadshow. Just as Unity had done, DoorDash met with investors, examined their turnover data and other investments, and attempted to come up with a nuanced understanding of which of them could reasonably be expected to hold the shares over several years.

Members of DoorDash’s finance team looked at the last hundred tech IPOs going back to 2015 or so and found that many investors sold sooner than expected. While the market shorthand was that mutual funds held while hedge funds flipped, DoorDash’s analysis, much like Unity’s, showed that the diversity of investors in the current market made that axiom obsolete.

That was in part because there were many buy-and-hold hedge funds focused on specific industries like tech. Large mutual fund complexes were so big that any allocation they got—even if it was among the biggest—got split among various funds. It was hardly material to their bottom line unless the portfolio manager could build it out in the aftermarket. Some managers didn’t like to keep stocks unless they represented a meaningful holding.

As the company sought to understand what mutual funds might be committed long-term holders, they found that hedge funds had done some of the best analysis on the company. The fact that hedge funds could be more nimble or global in their investment choices than some mutual funds meant that they had a good understanding of the worldwide implications of the food delivery industry.

The order-entry system that Unity had developed, by requiring both order size and price, gave DoorDash executives better insight into which investors were likely to hold. If an investor showed interest at share prices well in excess of where the company eventually set them, DoorDash could reasonably assume a bullish outlook. Every investor will tell a management team that it will act as a long-term holder, but DoorDash hoped that by looking at the bids, they could come up with a more data-based metric.

On November 30, the company amended its prospectus to show that it planned to sell 33 million shares in an initial range of $75 to $85. The price would value the company at $32 billion, roughly double what it received in the June financing. The range was the subject of some significant debate between DoorDash and its bankers. It was higher than traditional IPOs and ran up against a preference among bankers to set the range low in order to attract as much interest as possible. The problem, Carl Chiou and others believed, is that the lower price gets anchored—the company has to amend the prospectus every time the price increases, or decreases, by 20 percent or more. But bankers and company executives didn’t like to amend the prospectus more than a few times, because it’s cumbersome and time consuming.

Adarkar and other members of DoorDash’s management team pushed the bankers to increase the initial range. As a consumer-facing brand, DoorDash didn’t feel like it needed help drumming up interest by keeping the range too low. In the end, Goldman’s bankers talked the executives down and the two sides settled on $75 to $85.

Investors soon showed that they knew how the game was played and provided support for the behavioral finance theory that investors anchored to the initial range. When the bids came in, most of them were exactly 20 percent above the range. DoorDash had effectively wasted its first round of bids.

On December 5, DoorDash updated its prospectus to show the new range. Now the company said it planned to sell shares between $90 and $95, or 15.6 percent higher than the initial range. The IPO was just days away.