Uber, Lyft, and the Subscription Opportunity
Go big or go home
Of course the age of abundant information is prone to beg, borrow, and steal your attention. Reading blogs, news, and research has always been an inefficient user experience — finding needles in haystacks. But now, Annotote is the antidote, check it out: Don’t waste time or attention; get straight to the point.
The following excerpt got me thinking about revisiting a strategy I’ve discussed before…
From The Verge: “Lyft is testing monthly subscription plans for high-frequency users, a sign that the company is shifting toward a Netflix or Spotify model for transportation. The terms of the subscription models seem to vary, but appear targeted at users who spend up to $450 on ride-hailing a month…”
Uber flirted with a similar idea last summer, and it makes a lot of sense: locking up demand is the best way to lock up supply. I still believe Uber should have spent more extravagantly here — that they didn’t is one of the biggest red flags about just how valuable this market might be.
— Ben Thompson, Stratechery (highlights courtesy of Annotote)
If I’m Uber, I’d go all-in on subscriptions before Lyft accrues the benefits therein. Subscriptions are a way to consolidate modal riders — the most active users — or as Ben says above, “locking up demand is the best way to lock up supply”.¹ But, there’s much to consider when evaluating such a magnanimous strategic shift…
Consumer preference
For me and many other mass market commuters, our households’ secular transportation planning is not determined as much by “experience” as financial breakevens — opportunity costs. But, before we dive into the dollars and cents, let’s take step-back to look at the whole equation, both financial and non-financial.
It’s really difficult for a consumer to predict the more qualitative factors in his/her transportation outlook. There are just too many soft variables, like capacity utilization and unscheduled downtime. So, most consumers opt for buying or leasing a car due to the certainty, taking solace in not only the ready availability of their own car, but also the relatively defined budget: ‘It may cost $250 extra every month, but at least it’s a known quantity.’
That all said, ridesharing has proven itself as a viable, reliable mode of transportation — eliminating a lot of those more qualitative concerns. Accordingly, ridesharing is now free to compete with other mediums on a more quantitative basis. As such, its remaining hurdle is proving that it’s consistently cheaper than car ownership. (Although, you could make an argument that “cheaper” isn’t necessarily the bar, but merely “cost competitive” would suffice. For example, in terms of reliability, you could argue that ridesharing networks are theoretically more resilient than a lone family car, but we mis-attribute these risk factors due to availability heuristics and probability neglect. Nevertheless, I’ll stick with “cheaper” as the hurdle, because the low costs of switching to next best alternatives, like Lyft, might warrant a competitive discount for those commuters who are open to the ridesharing-only option.)
To wit, consumer preference is biased against open-ended liabilities — a bird in hand is worth two in the bush. Buying or leasing bear a lot of tangible costs, but those are very predictable costs (e.g. amortized payments + insurance + gas + maintenance + … + parking). To quantify that:
“In 2017, cars cost their owners an average of $8,469 a year ($706 a month, or $23 a day) … ranging from $6,354 for small sedans to $10,054 for a pickup truck.”
While there’s plenty of margin in there that ridesharing could easily undercut, consumers’ monthly spend is still too unpredictable in the current, pay-per-ride regime: Are you going to take a mere 40 trips this month — just back and forth from the office every business day — or are you going to face considerably more impromptu errands and events? How much does surge pricing increase the variability of your aggregate spend?
Enter subscriptions…
Subscription economics
Now, Uber cannot undertake subscriptions half-heartedly by offering competitive pricing for both pay-per-ride and subscription services. Such half-measures could threaten it with illiquidity (i.e. negative demand shock among riders precipitating a proportional supply shift among drivers) or adverse selection (i.e. the most active users subscribing to save money vs pay-per-use). In other words, Lyft’s aforementioned experiment in “testing monthly subscription plans for high-frequency users” is destined to fail.
In contrast, going all-in on subs promises the following:
- Increased ARPU
By allowing consumers to comp ridesharing costs vs car ownership on an apples-to-apples, Uber would become a more perfect substitution for core transportation solutions, unlocking higher monthly spend per user. Compared to the $706 average monthly spend for car ownership, mentioned above, Uber’s most frequent riders spend only $95 per month and its average rider spends only $50. There’s considerable headroom there for ridesharing — and don’t forget that a lot of Uber Pool riders benefit from pre-tax dollars too! - Increased aggregate demand
DAUs and volumes would spike thanks to the flat fee encouraging “all-you-can-eat” consumption effects. Currently, rough math says Uber’s 75 million MAUs average 6 rides per month. Ramping those 6 rides up to the aforementioned 40 required for the average commuter to get back-and-forth from work every month translates into a $333 monthly subscription fee at current rates.² Again, there’s considerable headroom there for Uber to increase pricing; drivers to increase takehome pay; and riders to breakeven relative to owning/leasing — all while ramping usage! (More on this in a moment.) - Increased demand predictability increases supply efficiency
- Increased positive working capital cycles
Leverage negative cash conversion cycle a la Amazon, wherein Bezos was able to self-fund growth, organically, out of cash flow, without raising outside capital, since Amazon’s customers paid it before Amazon had to pay its suppliers. - Sustained profitability
Of course, Uber could still maintain its on-demand option for a-la-carte/pay-per-use rides, but they’d have to increase the premium and decrease the user experience such that the monthly subscription is a no-brainer. That’s what I mean by going “all-in” on subscriptions: Fully optimizing for that one specific product-market fit — often at the expense of non-core users. (This makes me think of the Vail Resorts pricing model for skiing lift tickets vs seasons passes, which over-serves the latter at the expense of the former.) For example, Uber could structure its algorithm so all a-la-carte rides pay undiscounted full fares and are accompanied by either a Pool co-rider, food deliveries, or both.
For a number of obvious reasons, Uber would unfortunately have to cap monthly usage. Above a certain quantum of milage, users could perhaps pay a 30% discount to the prevailing a-la-carte rate. So, were Uber to cap subscribers’ usage at 400 miles per month, then $400 is actually a pretty reasonable monthly price point — not even accounting for the prevalence of pre-tax dollar benefits. At worst, that works out to a $1 per mile fare equivalent, which exceeds Uber’s current rates at $0.60–0.80 per mile (depending on the geography). That’s also a material discount to car ownership’s $706 comp, and power-users can withstand plenty of overage charges, since the $400 starting-point is significantly lower than the average American’s total transportation run-rate of $756 per month as well. Furthermore, it gives the average subscriber 10 miles per leg of his/her commute, which is substantially more than the current average Uber trip at 6.4 miles.
In sum, Uber’s subscription model might not necessarily be a fixed budget for every rider, but it could be sufficiently predictable relative to the next best options.
The Innovator’s Dilemma
Now, this isn’t for everyone. Specifically, Uber could be alienating a lot of its major metropolitan riders — for whom it’s probably not worth $400 per month for a short-haul car service. That’s the Innovator’s Dilemma for the subscription opportunity here: An incumbent doesn’t want to cannibalize a massive revenue base or hard-earned customers.
But, those losses are likely worth the gains, based-upon all of the “promises” listed above — and they might not cut as deep as you’d expect. Not only is surge pricing a testament to the surprising demand inelasticity across a big segment of this market, but many of Uber’s short-haul trips are better served by other solutions — better for Uber, riders, and drivers:
“Why Is Uber Buying an Electric-Bike Company? Because short car trips are costing it a fortune — and riders might be seduced by a bicycle with some extra oomph.”
Win-win-win! Uber has the best possible ecosystem for this model — given UberEATS, JUMP electric bikes, etc. Scooters also seem like an up-and-coming option to add to the mix, perhaps better serving urban short-haul riders than automobiles. In addition to retaining marketshare, those supplemental services really layer-on subscribers’ perks. I don’t see how Lyft could compete on this plane, which is why the subscription opportunity is particularly ripe for Uber.
Thing is, Uber doesn’t have to pivot down this path. They already have a scale advantage and seem content to try and bleed Lyft dry — or at least subsist as the alpha in a two dog litter. Overhauling their business model is a risky strategic pursuit too, as discussed. However, the reward is substantial in both absolute and relative terms. Furthermore, Uber’s chance to suffocate Lyft seems to have passed, so all of the ridesharing market’s projections for future TAM, scale, and profitability need downward revisions, changing the calculus on the prospects for gross margins and net income for the legacy business model.
A series of self-inflicted wounds, like PR gaffes and a generally toxic culture, have taken Uber from the brink of consolidating the entire ridesharing market upon its failed Lyft acquisition in 2016 to Lyft getting a series of capital infusions that fueled its comeback, which saw its marketshare rise from 20% to 35% in the intervening period. (The fact that Uber was on the brink of total domination but lost on account of own-goals should not be lost in future case studies thereof.) As a consequence of this increasingly entrenched duopoly, it will be much harder to gain sufficient marketshare such that Uber (or Lyft) were to reach profitability. Lobbying and litigation expenses aside, there are now increasing costs to acquire and retain drivers on the supply side. Once upon a time, young Uber’s projections would have expected these costs to wane as it consolidated marketshare, as suppliers would have had to come onto Uber’s service on Uber’s terms — per Aggregation Theory’s edict. In contrast, there are now substantial contra-revenue and cost of revenue line items that have to be discounted forward.
So, the ridesharing market’s resurgent competition means it no longer presents a ‘small margin multiplied by high volume’ path-to-profitability. Hence, Uber would be wise to take its small cost advantage and win the first-mover-advantage in subscriptions, which would grow its TAM by expanding into the market for car ownership — as a viable substitute thereof.
An example of consumer preference…
¹ Re: “locking up demand is the best way to lock up supply”
I personally don’t prefer this maxim, because a catalog of past multisided networks show that businesses can approach the chicken-or-the-egg conundrum from either the supply side, the demand side, or both!
² 15 million rides per day * 30 days per month = 450 million monthly rides / 75 million MAUs = 6 monthly average rides per MAU;
40 average monthly rides per commuter / 6 monthly average rides per MAU = 6.66 * $50 monthly average spend per MAU = $333
