The lede: If we are to have a fleet of predominantly full-electric vehicles on the roads in the next decade, there’s a key dependency that needs as much focus as the auto battery and initial electricity generation: EV charging—install & deployment, UX, predictability and more.
Lo-fi Background
The EV charging experience is fairly wide ranging now, from “drip charging” on the street with retractable cables mounted to phone poles, to “super charging” a Tesla at their purpose-built charging stagings, to DC charging with home installs, to a range of voltage and expectations around amperage, etc. (Note: I’m not an electrician, a physicist or anything in between, so see an expert for all things voltage/amperage.) Interestingly enough, there hasn’t been much variation in this space for the last decade, despite all of the hype. See this blog post from 2012.
This post takes a look at a few areas for improvement that in a kaizen-like fashion is designed to bring minor improvements to the experience and adoption of EV charging.
The real crux here is we NEED this entire system to be improved to get full EV adoption in the market. The fleet needs the supporting infrastructure. The humans operating the fleet need to understand the infrastructure and tools in order to adopt them. And the infrastructure needs to be there to begin with in order to be used.
Suggestion #1: Improve the UX
AKA WTF am I supposed to be doing here? The image below is represents a user experience that’s pretty similar across most providers.
I walk up, I wave my hand in front of it, the screen lights up with a logo. … Okay, now what? I’ve seen this again and again. This is a simple “cold start” and product education challenge. Simple step-by-step workflow please. Treat me like I’m a dummy (as I, personally, am) and tell me exactly what I’m supposed to be doing. It’s not habit yet, and it probably won’t be for a few years.
Simple tactics:
Branding is secondary here. Make sure the experience works. Give your UX staff purview over your branding design staff. #kidding #notkidding
Simple workflow please. On recognition of a human observer, illustrate instructions in an enumerated, sequential fashion. Just like simple self-service pay stations at many pay lots. Don’t get fancy. KISS. And bear in mind the screen might not light up due to some motion detector failure or screen outage, so consider physical signage.
Enable a way for the user to re-trigger the screen in the event that motion detection fails. Always have that fall back for critical functions like this on physical devices (eg. with smart phones, you can shut down via the software interface OR you can simply long-hold the power button…and this is almost universally the case across all devices manufactured regardless of make, model, yada yada).
Let me pay you. Please. Most services require a proprietary app to use their charger. I hate to tell y’all, but you have to earn that privilege. Shoot, no one expects me to install a custom app to pay for gas at an OG petroleum-product fill-up. // I can’t tell you how many times I’ve struggled with apps and simply gave up on non-Tesla chargers and decided just to wait for my next supercharge visit.
I could go on…but I know there are industrial designers and six sigma types out there that could 10x my suggestions. This short list should get the ball rolling…
Also, what do those little red dots mean? Are they telling me about charging progress? Network access indicators? Are they red herrings and reading anything into them is a quixotic task?
And how about removing the nozzle? Watch the video above. Am I incompetent? (Yes, you can say “yes” here.) Or am I representative of the norm here?
Simple tactics:
Again, point me in the right direction. If the nozzle cannot be removed prior to some other action being performed, say so explicitly…at least 3x…digitally and physically.
Insert responsive UX here. (AKA usability over branding Pt. II) Those inexpensive sensors that know when a nozzle has been removed? Use them to also indicate when someone is attempting to remove the nozzle. If they’re trying and struggling, show some messaging on the screen. Don’t show your brand name ad infinitum…as at this point I’ve just developed an association with your brand as a creator of non-functioning products.
Let me just unholster the nozzle, damnit. Follow Tesla’s lead here. There supercharger stations will let you pull out that nozzle and whip around that cable to your heart’s content. No sweat off their back. Why not? This locking is serving no real JTBD purpose that I can identify.
See #1 and #2 above. Rinse and repeat.
ADA compliance. Side note: I’m pretty sure none of the screen’s I’ve seen are optimized for various color blindness and other visual impairments that don’t prevent one from being a licensed but DO prevent you from using most of these damn machines when there’s (a) low light, (b) too much light, or (c) any variant of light. Perhaps don’t optimize for sexy LCD screens but just use some highly functioning, low energy consumptive e-ink. B&W, baby. With no shiny surface. Kindle since Day One, FTW.
Suggestion #2: Improve Network Access at Charge Ports
I’m in the garage pictured above. The one with the two empty parking spaces. The always sad empty spaces. I digress.
Those two spots, rest of the UX aside, are probably sitting empty the majority of the time because there is no reception (CDMA or GSM, or Wi-Fi) in the garage. Because I’m required to authenticate in an app (after verifying an email on first use…which is the majority of the users on these charge ports) to enter a payment mechanism so that I can just just pay for the electricity already…damnit. Because I need all of that data transaction, because I can’t pay via the charger device itself (see comment above re: paying directly) I must have network access on my device. No network means I can’t pay, which means I can’t charge. You see where we’re at here?
Sad fact: many large garages like this one (at a mixed use commercial / retail area), or those in large hotels, or shopping malls, or…you get the point…many of these facilities, if underground, lack consistently available network devices. WTF? How was this critical dependency forgotten?
This example isn’t just my single person experience either. I talked to an installer recently and he said this is frequently the case. He finds that others did an install and failed to call out network access as a requirement (see section below about install/deployment staff).
Suggestion #3: Bear in Mind Basic Physics - Gimme Some Slack
The shortest distance between two points is a straight line. Cool. Basic physics established. We (the collective we of service providers in the space) can save money by having just enough cable deployed to provide a direct line between the holding box on the utility pole (that contains the spooled charging cable when it’s not dispensed by the machine) and the charge port on the car. Again, cool.
Quick reminder: auto manufactures have distributed ports on all sides of vehicles in the market. Driver side. Passenger side. Front (below image). Back. And some parallel street parking is driver side to curb and some (most) is passenger side aligned to curb. Pretty please, don’t forget to account for all those possible variants. 🙏
So, lowest common denominator here, and sufficient slack to cover variants. Else, empty street parking spots just like those in the parking garages above, but for a much different but equally annoying user experience challenge.
Suggestion #4: Improve Availability / Wait Time Predictions
This one is specific to Tesla…today, but will likely be relevant to many other manufactures with their own custom high voltage charger stations or even manufacturer agnostic charging providers.
When I roll up to a gas station (unless it’s the US circa ‘79 or a few other rare instances) I typically don’t have to wait more than 5 minutes to fuel up, and then that filling experience is 2-3 minutes.
When I roll up to a supercharger station, I might wait 15-20 minutes to plug in, and then there’s the 45 minutes for the charge itself. This means lots of manual changes on my part as a customer’s been so used to ease and convenience in re-fueling my entire life.
As you can see below, Tesla tries to manage utilization distribution by showing kWh pricing variance to guide some drivers to the low utilization late night/early morning windows. That’s a simple start. I appreciate it.
Now notice it says “short wait”. What is “short” when it comes to a wait? Two minutes? 10 minutes? 30 minutes? At this point it’s a placebo copy block, like a “close door” button that does nothing.
You’re Tesla. You have brilliant statisticians, mathematicians, data scientists galore. Your “small data” problem is no longer an impediment, or at least not enough to preclude you from doing some decent estimations of actual availability. You know typical egress timings after a car is fully charged, and you probably know the variants for those just doing a top up vs. those doing a full charge. Random aside: my hypothesis is the egress timing here—i.e. idle time where my car is blocking a perfectly good charger as I’m sitting in the necessary spot that provides access—is higher for longer charge times. You also have an indication of typically how long someone will wait idly for a spot versus those who will leave the queue (in frustration)—and you can probably slice and dice that cohort endlessly as well.
Rambling above aside, you can probably give me more of an indication than a “short wait”.
(Side note: I wonder if someone would pay for access to “express charging access” much like you can pay for faster delivery on DoorDash or Uber Eats, et al? And not would they, but rather how much more would they be willing to pay for access to a "fast pass slot”?)
Suggestion #5: Workforce Development is Needed ASAFP
Remember the comment above regarding a service provider who noted how so many parking garages lack network access for users near charging stations? Well, that guy’s company—who’s contracted by cities, companies, schools, individuals—has a number of teams of installers. But he can’t keep up with the charger install/deployment demand because of labor.
He noted there just aren’t enough electricians. I’ve done next to no research to confirm this, but seeing that similarly specialized “blue collar” jobs seem to have fewer people entering those fields and the existing labor pool is aging, this is consistent with US manufacturing and other historically unionized classes of labor.
This is an opportunity for service providers to spend into the creation of trade school programs for electricians and similarly skilled folks who will lay the cable and perform the install on network repeaters, set up the chargers vectorized in the image above.
To scale the infrastructure to meet the needs for an EV vehicle fleet the US government is striving for, we will need to more people trained to do the job. And given this is high-voltage electricity we’re talking, it’s critical people are being trained appropriately and sufficiently.
Throwing Stones in Glass Houses
This piece isn’t intended to target any particular service provider, despite me calling out ChargePoint, EVgo, and Tesla by name. Those selections were just a function of laziness and recent personal anecdote (thanks, recency bias). All of the suggestions above are applicable to pretty much the entire industry.