• Flywheel
  • Posts
  • Flywheel | February 23, 2023

Flywheel | February 23, 2023

An example of using geospatial analysis for TAM assessment with Chris Hannesson of Magna and featuring the top 5 vehicles of the week

Hey!

Welcome to Flywheel, a weekly exploration of the used side of owned micromobility. Each newsletter will highlight an observation of trends emerging in the industry and feature five of the most interesting used vehicles being sold in the secondary market.

If you were forwarded this post and found it interesting, I would appreciate it if you subscribed to Flywheel!

The observation of the week is Part 2 of a guest column by Chris Hannesson of Magna exploring geospatial analysis for TAM assessment. This week’s featured vehicles are a recently launched long-tail cargo bike, two commuters, an ultra lightweight road ebike, and an ultra portable scooter.

If you’re viewing this newsletter in your email inbox, it may appear truncated due to the length of this report. I recommend reading this report in your browser by visiting the post here, clicking on “Open in browser” at the top right of the email, or clicking on "View entire message" at the end of the email.

Observation of the Week

Please welcome back Chris Hannesson (Director of Product and Tech Investments at Magna), who is returning as a guest columnist on Flywheel for a two part series on using geospatial data to unpack market sizes and opportunities for micromobility. Last week’s Part 1 can be found here, and Part 2 can be found below:

Exploring Geospatial Analysis for TAM Assessment, Part 2: A guest column by Chris Hannesson of Magna

Hey, my name is Chris, and in Part 1 we talked at a high level about how geospatial software is an unloved way to calculate TAM and SAM. In Part 2, let’s actually do it. 

Reminder: These are my opinions, and I do not speak for Magna in this article, nor is any product or business or website I refer to an endorsement.

Part 2: Geospatial TAM estimation and opportunity discovery

  • Orientation: architecture, existing tools, and data sources 

A thorough overview of the architecture of geospatial software is out of scope. To keep it simple, distinguish (1) tool and (2) data. You need both, and they are not the same. 

Where should you start? If you have no experience, a paid service like ESRI or Mapbox is the easiest.

(1) Tools (ie things that you load data into and perform analysis with to get interesting answers)

(2) Data (ie the information we talked about in those boundary conditions we want to use tools to tell us interesting things about)

  • Scoping our example

Let’s look at how you could come at North American addressable market sizing using ebikes for delivery and most suitable launch markets.

So first, you do the typical thing - try to estimate top down and bottom up various things from your traditional public and private sources. Fill in the gaps with expert research calls where you just can't find what you are looking for. Be curious and skeptical.

Next, repeat your affirmations to the gods of false precision – if we go fast, this is directional analysis only and fraught with many potential sources of error. Retiring these risks efficiently is why you get paid the big bucks!

  • Step 1: All cities are not created equally. Quickly filter for most suitable ones.

Let’s assume a homogenous population with respect to spend/month and price elasticity so we just care about deliveries per hour, the number of people and relative speed versus a car (the alternative mode). We can use geospatial tools to quickly sort by (1) high population density and (2) speed limits (ie you can’t go fast even if you could).

Then you’d want to visualize speed limits to see if there are areas where the government makes cars go as slowly as ebikes. And cities like Chicago (left) are more suitable in some areas than Houston (right).

  • Step 2: Get more real.

People actually don’t spend on restaurant food at the same level (income, willingness to pay, proximity, preferences, etc). This shows a slice of Chicago east of O’Hare with two variables at the “census block group” level (smaller than a ZIP code, it’s ~600-3,000 people): population density (colour of circle, darker is denser) and annual spend (small circle is national average indexed to 100, largest circle is >3x that).

  • Step 3: Find the sweet spots.

Let’s zoom out to Chicagoland and isolate “Target” retailers. The sweet spot emerges where ebikes can be just as fast as cars, and where there is a far higher of order and spending density than the rest of the city, especially the suburbs. 

Sweet spots exist in other places as well, so let’s switch to Toronto to get specific (highlighting a different retailer with a large footprint, and the >30mph roads).

  • Step 4: Measure SAM accurately.

If geospatial software has one party trick, it’s that it can look at a bunch of layers of data within a shape (a polygon) you define. So you can look at an entire city or at a census block group or a sweet spot that you define, and then just ask it to spit out how many people are there, how much income they earn, how much they spend on restaurants or whatever you want – as long as there is census data or credit card data or other data available to you, you have your TAM. Super flexible.

Red is showing the TAM, which I’ve defined here as a car can get there in 12 minutes (a polygon called an isochrone that we can create with another geospatial tool, see eg link): ~2.8 million households with a median household income of ~CA$90,000, so ~CA$250 billion. The TAM for restaurant spend would be some fraction of that. In the example below, downtown Toronto, the amount of annual restaurant spending is ~CA$1.2 billion.

How much of this is serviceable with an ebike fleet? We can measure this too, by keeping ebikes on <=30mph streets only and looking at travel speeds in an ebike (green) and a car (blue) to identify where there is parity in speed (or what % of destinations can an ebike deliver to at the same speed). In this case it is ~96%, so SAM is ~CA$1.15 billion. Sweet spots are important because you are pretty much guaranteed to have a really high level of parity (most food is delivered during peak congestion, meaning cars have no speed advantage).

  • Step 5: Extend this analysis nationally to calculate TAM and sort to figure out where to launch.

To finish off our example, we can calculate summary statistics for our reference city (or cities small, medium and large), call them “archetypes” to sound very authoritative, look at a list of comparably populated and dense cities and start multiplying then adjusting for population density or median household spending or income, and the product of this whole effort would be a credible TAM estimate. You can do the same with SAM, and by sorting in Excel, you can prioritize where to launch and operate down to specific blocks.

  • Bonus: Cool stuff you can bolt onto this

Bolting on things like vehicle route planning and more generalized optimization analyses can get really powerful. I find this link helpful for what to do on route planning but I haven’t found an easy to interface with optimization tool yet to solve complex business problems that beats Excel...

Have a vehicle you'd like to feature and sell on Flywheel? 

Submit vehicle information at submit.rideflywheel.com or click the button below.

Looking for a used vehicle of a specific brand or model?

Fill out the form below to receive weekly Flywheel alerts of all available used vehicles that fit your criteria, directly in your inbox.

Top 5 Vehicles of the Week

The Lectric is a class-2/class-3 long-tail cargo bike. Announced just this past Tuesday, the XPedition seems set to become the most capable minivan replacement at the ultimate price. Capable of hauling a startling 450lbs, the XPedition’s powertrain features an 85Nm rear hub motor and a 672Wh battery pack, and there’s even an option for a dual-battery set up to double your range. Lectric has implemented a few smart motor controls to ensure that the XPedition is both responsive and easy to use. In addition to a throttle that makes the XPedition so much more practical for cargo hauling, the XPedition also uses a new “Pedal Assist Wattage Regulation” controls approach to regulate pedal assistance levels by power as opposed to speed. Both of these features should result in a more intuitive understanding of battery consumption as well as a smoother response when pedaling. Lectric gets a lot of things right with this new vehicle. A standard XPedition includes rear rack cushions and boards in addition to other must-have accessories for urban riders (fenders, kickstand, lighting), and the vehicle comes fully assembled in the box when shipped. Not to mention the fact that it retails at an unbeatable price of $1,399, the XPedition is a no-brainer choice that makes it as easy as possible for new riders to go from hitting order on the website to using their new vehicle at its max utility. Lectric has quickly become one of the most important OEMs in the ebike sector. They’ve been on a tear this past year, releasing the XP 3.0, the XTrike, and now the XPedition. The company produces unbelievably accessible and highly reliable vehicles, all at unbeatable prices. Listing can be found here.

The Haibike Urban Plus is a sporty class-3 commuter. Featuring a custom TranzX powertrain that consists of a 70Nm mid-drive motor and a 490Wh battery pack, the Urban Plus is one of the few Haibikes that doesn’t use a Bosch or Yamaha system. However, TranzX powertrains are commonly used by Haibike’s parent company across their other brands (Raleigh and IZIP) and provide ample power for day-to-day riding in a smaller and cheaper package. The stiff frame and hydraulic brakes give the Urban Plus agile handling, although many newer riders do complain that the Urban Plus borders on feeling too stiff. The Urban Plus comes standard with a built-in rear rack and fenders, which are surprisingly uncommon on dealer ebike brands. Its software system is also unique and markedly better than that of many other dealer ebike brands. Urban Pluses come with a COBI smart hub, which allows you to dock your smartphone and use COBI’s highly-integrated mobile app as the UI for your vehicle. This listing has a mileage of ~580mi, and is a great option to get a high-quality class-3 commuter for the price of a class-1 or class-2 budget bike. Listing can be found here.

The Urtopia Carbon One is a class-3 road ebike designed by award winning automotive designer Mathis Heller. In addition to its unique Möbius band design, what’s most notable about the Carbon One is that it only weighs 33lbs. Due to its luxury carbon fiber frame, the Carbon One’s frame is not only ultra lightweight but also has better vibration dampening than aluminum alloy frames. The powertrain features a 35Nm rear hub motor controlled by a torque sensor, a 360Wh battery pack, and a Gates carbon belt drive. While the components and materials choices make the Carbon One extremely smooth to ride, its modest powertrain feels a bit weak for hillier cities like SF. The Carbon One is also packed with industry-leading electronics and software, which enable lots of smart features like a bluetooth speaker, turn-by-turn navigation, voice control, anti-theft GPS tracking, and built-in turn signals. This listing is in mint condition (Flywheel estimated 194mi of usage) and has been “gently ridden only [a] few times.” Listing can be found here.

The Stromer ST1 is a premium class-3 commuter by the brand that has long set the standard for high-quality urban ebikes. Launched in 2013, the ST1 Platinum was once upon a time Stromer’s most premium vehicle with better top-speed, transmission, and range. Its powertrain features a 40Nm gearless rear hub motor (capable of regenerative braking and controlled by a torque sensor) and a 522Wh battery pack. The ST1 Platinum is buttery smooth to ride, but it’s quite hefty (weighs 62lbs) and can be difficult for smaller riders to maneuver. This listing has a mileage of 386mi and is sold by LA bike shop Orange County Cyclery, who recently inspected and tuned the vehicle before posting it for resale. Orange County Cyclery is a particularly interesting shop as they not only retail new bikes, but also very actively buy, sell, and trade used ebikes. In fact, they have listed 8.5% of all used ebike listings in LA on Craigslist over the past 14 months. Listing can be found here.

The E-TWOW GT SE is an ultraportable, connected e-scooter and the Swiss army knife of commuter micromobility vehicles. E-TWOW’s flagship vehicle packs an incredible performance and stability despite its low weight of 29lbs. The powertrain features an 18Nm, 700W motor and a 504Wh battery pack. There’s three braking systems, but most notable is the regenerative braking that is inspired by the KERSs (kinetic energy recovery systems) of F1 cars and is claimed to be capable of recovering up to 60% of the braking energy. The GT SE’s connected electronics features primarily focus on security and safety. There’s an anti-theft lock and alarm that make a loud noise and engage the brakes when unexpected movement is detected, and there’s an extremely loud horn that is a welcome feature to ensure that car drivers can hear you. This listing has a significant mileage (1K+ miles) and is sold by Sella, an interesting 3rd party service that makes buying used products more trustworthy. Sella “helps people sell their used goods” by working “directly with the owners to review and list each item” and coordinating delivery or pickup with Sella reps. They even offer a money-back guarantee if items sold are “verifiably different from the details” on a listing. Given the massive networks that already exist on platforms like Craigslist, it’s exciting to see a service that meets customers where they’re at and sits directly on top of these marketplaces to help bring better trust, transparency, and UX to the secondary market. Listing can be found here.

That’s it for this edition. Thanks again for joining, see you next week!

- Puneeth Meruva

🚲 If you have any suggestions, interesting listings, comments, questions, etc., please email me at [email protected].

🚲 I would greatly appreciate it if you could share Flywheel with your friends!

🚲 Follow Flywheel on Twitter!