A peace plan to end the wireless wars

No one would have predicted that the three of us would ever find ourselves on the same side of the corporate patent wars, let alone speak with one voice about how to end them.

That’s because one of us is the patent chief at a global smartphone maker (and an influential critic of patent licensing abuses); another is the former licensing chief at Apple and current chief executive of a non-practicing entity (NPE) patent licensing company that has been a target of criticism from product manufacturers; while the third is president of a patent pool operator, who has criticized companies on both sides of the patent wars for their gamesmanship, lack of transparency, and litigiousness.

We have come together because we see that patent owners and product makers have become trapped in an endless cycle of demands, counter-demands, and unproductive litigation. Unless we find a way out of this conflict, we will almost certainly see a repeat of yesterday’s costly and wasteful smartphone wars in tomorrow’s wireless connected car sector.

Product makers accuse patent owners of threatening lawsuits and using the expense of the legal process in order to demand extortionate royalties for their patent rights. For their part, patent owners say product makers refuse to pay fair compensation for the patented wireless, audio, and video features that give their products value as communication and entertainment devices.

The truth is, both sides have a point. That’s because patent owners and product makers are caught in a classic “prisoner’s dilemma,” in which the lack of transparency and fair ground rules in patent licensing lead companies on each side of a patent dispute to try to game the other. This only ensures that both sides suffer a negative outcome in outrageously-expensive litigation.

Unlike in the real property business, in intellectual property (IP) licensing there is little or no independent appraisal of the assets (i.e., patents) or transparency as to how prices are determined. And because most patent license agreements are confidential, there is little or no information or “comps” on what others have paid for similar patent rights. Nor are there any widely-accepted ground rules for what constitutes fair negotiating practices between buyers and sellers.

This is especially true in regards to standards-essential wireless patents, which are supposed to be licensed on fair, reasonable, and non-discriminatory (FRAND) terms. But what’s fair or reasonable about the fact that an impossibly-large number of LTE (4G) cellular patents — more than 60,000, in fact — have been declared “standards essential” without any independent evaluation of those patents whatsoever?

That’s right, those 60,000-plus patents have all been self-declared “standards-essential” by companies each seeking their own commercial advantage. What you’ve got is a wireless gold rush — with plenty of fool’s gold posing as real gold.

So the three of us, working with industry leaders on both sides of the patent owner vs. product maker divide, have developed a three-pronged plan for ending the wireless patent wars and creating a more productive and less litigious patent licensing sector.

First, whittle down this ridiculous mountain of self-interested wireless patent claims to the fewer than 2,000 patent families that most experts believe are truly essential to smartphone handset makers. We can do this by excluding duplicative patents, expired patents, patents not in force in major economic markets, and patents for base station, infrastructure, and other innovations not relevant to handset makers. Independent, neutral evaluators will then confirm each patent’s relevance to the LTE standard for handsets.

Second, base royalty prices not on the subjectively-argued value of each individual patent examined in a vacuum, but on the objective value of the entire stack of LTE patents in a phone. A recent court judgment valued that LTE stack at roughly $20 for a smartphone with an average selling price of $324, but with greater price transparency from both sides, the market itself will likely set a rational price for the LTE stack. Royalties can then be paid to patent owners roughly proportionate to each patent owner’s percentage share of the total LTE patent stack.

And third, ensure greater transparency by promoting collective licensing solutions such as patent pools that openly publish their pricing frameworks and offer consistent terms to all licensees. Given the “prisoner’s dilemma” dynamics in patent licensing today, it is unrealistic to expect any one patent owner to unilaterally forego potential business advantage by revealing its pricing strategies. But collective licensing approaches such as patent pools reduce the risks of transparency for everyone.

As the IP journal Intellectual Asset Management recently noted, “There’s a growing sense that a collective approach to licensing could help solve some of the problems of the industry which, in sectors like mobile, has been scarred by long-running and costly disputes between patent owners and potential licensees.”

Our “peace plan” would eliminate many of the incentives and opportunities for gamesmanship in wireless patent licensing. And most importantly, it would help patent owners and product makers avoid a repeat of yesterday’s costly smartphone wars in tomorrow’s connected car, autonomous vehicle, and Internet of Things (IoT) industries.

It’s time for a new realignment in the industry — one in which the conflict is no longer between product maker and patent owner, but between those who license patents on a fair and transparent basis, and those who do not.

How to understand the financial levers in your business

How can an electric scooter ride-sharing company like Bird possibly make money?

If you live in a select number of cities in the U.S., it’s hard not to see electric scooters appearing on sidewalks all over the place. Electric scooter ride-sharing services are also remarkably cheap: $1 to start a ride and another $0.15 per minute after that. But electric scooters aren’t cheap, and the logistics of a shared network are off-the-charts complicated.

As someone working in venture capital for hardware startups, the above question is obviously rather prescient.

How does a company like this possibly make money?

Here’s a closer look at the basic unit economics of Bird, the electric scooter ride-sharing company based in Santa Monica, Calif. There’ll be a super simple model and test scenarios that show how critical it is to understand and manipulate the key levers of any startup  —  in fact, it can determine whether a business sinks or swims.

Building a Scooter

Two young people in love, gazing adoringly at a pair of electric kick-scooters. Because that’s totally a normal thing to be doing. 

It looks like Bird is using the Mi Electric Scooter as the base of their platform. The Mi’s recommended retail price is $499, but it’s probably fair to assume that Bird gets a bulk discount and can buy the scooters at around $300 apiece.

On top of the base cost of the scooters, Bird needs modules to turn the scooters into sharing economy units. That doesn’t have to cost a lot of money. A Particle 3G asset tracker in a box, plus some custom code to deal with the scooter’s power management, is all that’s needed, so let’s call that $80 per unit. That takes the total cost per finished scooter to $380; plus, we’ll toss in $20 for final assembly.

My back-of-the-envelope calculation puts Bird’s road-ready scooter at $400 per unit.

Deploying a scooter

Scooter startup Scoot operates electric scooters that are more like motorcycles than electrified Razer kick-scooters. 

One of the biggest problems electric scooter companies must solve is distributed charging. Scoot solved this problem by building a massive network of charging stations, distributed around San Francisco — a big infrastructure push, but necessary, given the robust profile of the scooters. Unlike Scoot’s wheels, which need to be returned to a charging station for charging, Bird scooters can be easily picked up and taken inside a user’s apartment or office, creating an instant and nearly infinite distributed charging network called “available wall sockets.”

This completely changes the charging game in Bird’s favor, so much so that Bird offers people $5 per charged scooter. This creates an elegant user experience and is a sign that it’s a key lever in their financial model.

A tale of two financial models

A lot of assumptions go into building a financial model. This one came together over a few beers on the weekend and is an example of the kind of “quick and dirty” math all founders should do as they pressure test ideas. You can follow along in this spreadsheet, and if you want to experiment with the numbers, you can duplicate the sheet and plug in your own numbers.

For both models, we’ll assume the following:

We’ll be playing with the average lifetime of a scooter, the average number of rides customers take per day, and some metrics around charging, to see how that effects gross margin.

Model 1: Uh-oh, this looks like trouble.

For the first model, let’s look at these dynamics:

  • Average lifetime rides per scooter — 300
  • Average rides per scooter per day — 5
  • Average ride length — 20 minutes
  • Percent of consumer charges: 50%

If those assumptions are right, it takes 220 rides (or 44 days) to reach break-even on the scooter itself.

After 400 rides (when a scooter is written off), the company has generated $147 of profit, at a relatively meagre 10.3% profit margin.

Suffice to say: That doesn’t look like a particularly sustainable business.

Model 2: A more optimistic outlook

However, you don’t have to change the assumptions much for it to be a much more attractive business.

What if Bird was able to extend the life of a given scooter, decrease the average ride length but increase the average rides per day, and push more of the charging burden to consumers?

Let’s look at these dynamics:

  • Average lifetime rides per scooter — 500
  • Average rides per scooter per day — 7
  • Average ride length — 20 minutes
  • Percent of customer charges — 75%
A very different story. Note here that Bird did not reduce the hard costs of charging a scooter (it’s still $5 for a consumer to charge and $20 for Bird to charge), but they did find a way to encourage customers to charge for them, reducing the overall charging cost.

Changing these four variables means that it only takes 165 rides (24 days) to break even, and the lifetime profit of a scooter is $813 — or a gross margin of 41%.

Financial models 1 and 2 side by side, for ease of comparing. 

So, do these unit economics make sense?

Investors certainly seem to think so. In February, Bird raised a $15m Series A, and only a month later, the company raised a $100m Series B. A company like Bird would be struggling to raise money on a 10.3% profit margin (as in Model 1), but if the numbers under the hood are closer to Model 2, it’s easy to understand how Bird starts to look like a rather attractive business.

Isolate variables to find your levers

In the case of Bird, you might be surprised to learn that three levers dramatically affect the finance model: average ride length, cost of charging, and usable life per scooter. Isolate variables and play with the numbers to figure out which ones are key levers.

Taking Model 2 above as a starting point, let’s explore by manipulating one set of variables at the time:

Ride length

Ride length has some impact on gross margin, but not as much as you’d think.

Usable life / Longevity

The durability of the scooters has quite a bit of impact, especially if the scooters fail early.

Cost of charging

The cost to charge a scooter has a huge impact on margins, so perhaps that’s a good place to focus.

Pulling the levers: SuperScooters

Once you know your levers, it’s fun to pull them a bit. If we were to optimize Bird to be as profitable as possible, it might be tempting to try to influence people to ride longer. But how? People’s commutes are probably relatively fixed, and you’re unlikely to be able to get them to change their commuting route. As we saw in an earlier example, though, the cost of charging has a huge impact on the overall business. What if we could find a better way to solve that?

Model 3 — SuperScooters

Say there was a different scooter available on the market— a SuperScooter —  with a swappable battery pack that clocks in at a hefty $1,000 MSRP.

The scooter has a higher up-front cost, but it’s more robust. Instead of a 500-ride lifespan, it has a 1,000-ride lifespan. The replaceable battery pack enables the Bird Service Crew to quickly replace scooter batteries out of charging racks in the back of the vans they’re already driving around town to redistribute scooters.

Let’s say that reduces Bird’s “recharging” cost to $3 per scooter per day, even cheaper than the consumer charge in Models 1 and 2, totally eliminating the need for consumers to charge the scooters at all.

In addition, let’s say Bird invents their own asset tracker that they can build for $30 per unit, rather than $80 off the shelf, and their manufacturer agrees to install it at the factory, taking the $20 in-house final assembly cost to zero.

By implementing the changes above, you end up with $2,467 profit per scooter, break-even at 34 days, and a gross margin of 62%. In other words: If such a scooter were available, it’d be a no-brainer: You’d want to replace the entire fleet as quickly as you could.

Comparing all three models shows that optimizing for longer scooter usable life and cheaper charging costs (Model 3) would have an extraordinary impact on the gross margin per scooter deployed. 

Build your model. Know your levers.

Obviously, I have grossly simplified the financial model here — if you were to model out the entire business of Bird, you would need to look at customer acquisition costs, customer lifetime value, churn, R&D costs of the elusive SuperScooter, and so on and so forth. Models get complicated quickly, but they also allow you to explore the impacts of changes you might make before you make them, which is invaluable.

Whatever your business, build a business model that includes all of your assumptions — and build the model so you can pressure-test variables and find your levers. Once you’ve identified them, build MVPs to test those assumptions in more detail. It’s really important to experiment early and get some good data on what works (and what doesn’t), before you start ramping up and pouring lots of money into marketing and execution. Some changes can have exponential effects — for better or for worse.