[I own DoorDash stocks and I may change my opinion anytime. This is not investment advice and please do your own due diligence.]

DoorDash recently released its 2024 Q2 financial results. Its top line is growing at a clip of approximately 20% YoY with the bottom line rapidly approaching the point of GAAP profitability breakeven. User order frequency is at an all-time high and still growing; newer cohorts of users are ordering as frequently as the “pandemic” cohorts, if not more frequently. DoorDash’s strong operational results fly in the face of the general environment where consumer spending seems to have somewhat weakened.

During the past couple of months, DoorDash’s stock price has experienced one of its worst drawdowns in history, with a peak-to-trough decline of over 30%. In my immediate two prior blog pieces on DoorDash, I highlighted several sources of risks, such as heightened stock valuation. After the meaningful drawdown, I feel that DoorDash’s stock is now in a more balanced space. “Tourist” investors have probably sold out to a large degree, as a multi-month 30% drawdown is not what “tourists” can easily stomach. For investors who follow fundamentals, what they now get is a moderately cheaper company that has just reaffirmed its strong business fundamentals. That is why I feel the stock is in a more balanced space now.

And this time, I want to share two thoughts about DoorDash.

Artificial Intelligence (AI)

I believe that DoorDash will likely be a beneficiary of AI. And, AI can be seen as an undervalued “call option” that is embedded in DoorDash’s stock price. This is how I see it.

DoorDash is a B2B merchant services business. Its primary focus is to serve merchants, mostly in the following two areas: last-mile delivery and customer service. The former incurs “delivery cost”; the latter incurs “customer service cost.” Both happen to be extraordinarily large cost items.

Delivery Cost: DoorDash collects commissions and fees from both merchants and consumers and uses that sum to pay delivery drivers. Based on my research, in the U.S., for an average food delivery order, DoorDash pays out around 50% (or a bit more) of that sum to delivery drivers. Look at it from a different angle: If not for the delivery cost, DoorDash’s revenue would instantaneously double. Wink Wink.

If AI is going to bring us true autonomous driving, “delivery cost” would be brought down to a drastically lowered level. Use ridesharing as a reference point here. It is somewhat a consensus that autonomous driving can slash “cost per mile” of a ridesharing ride from $2.3 today to under $1 (a cost reduction of over 50%). For food delivery, there are reasons to think the degree of cost reduction could be much higher than 50%.

Today, an average food delivery driver in the U.S. can complete about one to two orders per hour. In a world with autonomous driving, one can imagine that the “middle mile” between restaurants and consumers will be driven by driverless cars (at a much lower cost level than human workers). Those driverless cars could shuffle between a cluster of restaurants and a cluster of end consumers, carrying multiple bags of food for a given trip. Human workers, then, could simply meet those driverless cars at where end customers are, pick up multiple bags of food from these driverless cars and drop these food to end customers who are all located nearby. It would be a much more efficient logistics model. So, instead of completing one to two orders per hour, human workers would be able to complete four, five or more orders per hour. For the same hourly earnings, more orders can be completed. So, delivery cost on a “per order” basis would be drastically reduced. That is why I argue, with autonomous driving, the delivery cost in the system can be reduced by more than 50%, if not significantly more. Reduced delivery cost would first go to expand DoorDash’s revenue line and then flow directly to DoorDash’s bottom line. And remember, at the moment, DoorDash pays out 50% or more of what it collects as delivery cost.

Therefore, if we indeed get autonomous driving that is as good as we want it to be, because of the reduction in delivery cost, this new technology could potentially have an explosively positive impact on DoorDash’s top line and bottom line.

Customer Service Cost: Within the DoorDash organization, the customer service department is likely to be the single biggest department by headcount. Among the 30 some countries that DoorDash operates in, most are not English–speaking countries. Combined with DoorDash’s preference for providing users with superb customer service, my sense is that DoorDash cannot and will not entirely outsource its customer service function to lower-cost countries. Anecdotally, I have heard in some European countries, over half of the local DoorDash (more specifically, Wolt) team members are customer service agents who are local people and who speak local languages. According to DoorDash’s 2023 annual report, the company has “over 19,300 employees worldwide” (as a comparison, Lyft has fewer than 3,000 employees for the same time period) — so, I would not be surprised if some 30% to 50% of DoorDash’s employee base consists of customer service agents.

If AI is going to bring us AI–powered customer service capability that is indistinguishable from what human agents can do and can speak all the languages that we want it to, one can reasonably expect that DoorDash would no longer need to employ as many human customer service agents as it does today. DoorDash’s employee headcount could be drastically reduced and incremental cost savings would drop directly to DoorDash’s bottom line.

So, while AI poses extreme downside risks for many types of businesses, for DoorDash, AI presents extreme upside potential. And, unlike some companies that have to rely on the “AI story” to “keep up” their stock price, even without AI, DoorDash is already a good business.

DoorDash is in such a great position that its management team doesn’t even feel the need to mention AI to investors. Consider this: Never even once had DoorDash’s management team ever proactively mentioned AI at any of DoorDash’s earnings calls! It also means, the potential AI upsides in DoorDash are currently under-appreciated by investors. That is why I say, AI can be seen as an undervalued “call option” that is embedded in DoorDash’s stock price.

GAAP Profitability

Back in 2022, I wrote (link) that DoorDash can turn GAAP profitable “meaningfully earlier” than the contemporary consensus of 2026. As of today, DoorDash reaching GAAP profitability feels “imminent” to me — highly likely, I feel, it could realize within the next few quarters, if not in this upcoming quarter.

DoorDash has been FCF positive (on a T12M basis) since 2020. Whether to show GAAP profitability or not, as I wrote in another piece in 2022 (link), is a “choice.” Take this past quarter of 2024 Q2 as an example. On the one hand, DoorDash reported a GAAP loss of $158M. On the other, DoorDash chose to take two charges that are both more or less one-off in nature: A $83M charge for “office lease impairment expenses” and a $85M charge for “litigation reserves.” If DoorDash’s management were desperate to show GAAP profitability, they could have chosen to not take these two charges for this quarter and DoorDash could have been GAAP profitable already (-$158M + $83M + $85M = $10M). If nothing else, these seemingly arbitrary adjustment items appear to serve the purpose of making the eventual GAAP profitability more durable — i.e., by clearing out these adjustment items now, DoorDash reduces the risk of having to “reverse” back to GAAP non-profitability in the future.

Moreover, observant readers would notice an inconspicuous change of wording on DoorDash’s 2024 Q2 press release. In the notes area under the main body of text, the phrase “GAAP net loss” that historically was there has been replaced by “GAAP net income (loss).” The change of words, I speculate, reflects a collective sense shared among the company’s management team about an imminent GAAP breakeven.

After it reaches GAAP profitability, what’s next for DoorDash? If I have to guess, it would probably be the inclusion of DoorDash into the S&P 500 Index — an event that would broaden DoorDash’s investor base. However, this event is inherently difficult to measure, as the decision to include a stock or not is a discretionary decision on the side of the index company.

Overall, I continue to have a favorable view of DoorDash. I see the stock being in a more balanced space now versus the immediate past few months.

Disclaimer: Jackson Zhu (the “Author”), individually or through one or more entities controlled by him, holds long positions in the securities of and derivatives associated with DOORDASH, INC. (the “Company”) described herein and stands to benefit from an increase in the price of the common stock of the Company. Following the publication of this post, the Author intends to continue transacting in the Company’s securities, and may become long, short, or neutral on the Company’s securities. As such, the Author may change his view on the analysis presented as of any date following the date of initial publication. Likewise, the discussion contained here is not designed to be applicable to the specific circumstances of any particular reader, and you should consult with your own advisers to determine if any investment ideas discussed here are appropriate for your circumstances. The Author has obtained all information herein from sources believed to be accurate and reliable. However, such information is presented “as is,” without warranty of any kind, whether express or implied. The Author makes no representation, express or implied, as to the accuracy, timeliness, or completeness of any such information or with regard to the results to be obtained from its use. All expressions of opinion are subject to change without notice, and the Author will not undertake to update this publication or any information contained herein. Please read our full disclaimer at “jacksonzhu.com/about/.”