Installer SupportResource Center

2026-11-17 · Camille Lefevre

How to Install EV Charging at a Commercial Site: My 7-Step Checklist After 14 Months and $186,000

A procurement manager's step-by-step checklist for installing commercial EV charging alongside on-site solar and battery storage — covering the transformer capacity check most teams skip, how we scoped Tesla Powerwall 3 against hybrid inverter and portable options, what we learned evaluating Tesla Solar Roof, and why we paid extra to lock a delivery date instead of accepting a target.

Two years ago I assumed commercial EV charging projects fail on the electrical side—wrong breaker, undersized panel, a load calculation someone did in a spreadsheet at 11pm. Fourteen months, $186,000, and one blown deadline later, I'll tell you that's not where they fail. Ours nearly died twice, and both times it was procurement, not engineering.

I'm the procurement manager at a 240-person commercial property management company. I've owned our facilities capex budget ($2.1M annually, roughly $400,000 of it energy-related) for eight years, negotiated with 30+ vendors in that stretch, and logged every order in our cost tracking system. What follows is the checklist I now run whenever we scope on-site solar, storage, or charging. Seven steps, in this order, for a single-site retrofit or a multi-building portfolio.

It assumes you're the buyer, not the installer. If you're the electrical contractor, some of this will be obvious. Step 2 won't be, and step 6 is where the money is.

Step 1: Pull 12 months of utility bills before you look at a single product

When I first started scoping these projects, I assumed the starting point was equipment selection—which charger, which battery, which modules. That's backwards. The bills tell you what the system is for. We pull 12 months of interval data and reduce it to four numbers: peak demand in kW, demand charge as a percentage of the total bill, the time-of-use windows, and any power factor penalties.

One of our sites showed demand charges of $1,140/month (about 31% of the bill) for a load that peaked for maybe 90 minutes a day. That single number changed the entire scope—it meant peak shaving had real value, and it meant we should size storage against the demand curve rather than against outage hours.

Checkpoint: you can state your peak demand, demand charge, and the hours it occurs, in writing, before you request a single quote.

Step 2: Get transformer capacity confirmed in writing—before anything else

This is the step most teams skip, and it's the one that decides whether your project finishes this year or next. What most people don't realize is that the free site survey you get from a vendor usually covers the electrical service on your side of the meter. It does not include a utility capacity study. Those are different documents, produced by different people, on different timelines.

If your site needs a service upgrade, you are now in a queue. Public supply-chain reporting through 2024 put large power transformer lead times at roughly 100–120 weeks in North America, with wide regional variation. Verify current numbers with your utility rather than trusting a blog post (including this one).

Practical rule: submit the service request to the utility in week one, before you've chosen equipment. You can change a battery model in week twelve. You cannot compress a transformer queue.

Checkpoint: a written capacity confirmation from the utility, with an estimated timeline in weeks.

Step 3: Decide what the battery is actually for, then pick the hardware

"Battery" covers three different purchases that get conflated in every sales conversation I've sat through: peak shaving, backup power during outages, and deferring a grid upgrade. Write down which one you're buying. The answer changes the sizing math completely.

We ended up with Tesla Powerwall 3 units on two buildings. The spec that mattered to us was the integrated inverter (13.5 kWh usable, 11.5 kW continuous output per unit)—fewer discrete components between the meter and the load, which at the time reduced our balance-of-system cost and gave us a cleaner single warranty path. Two of the three qualified installers on our shortlist were Tesla certified, which shortened the commissioning schedule. I'm not going to tell you the cells are better than everyone else's; I don't have the test data for that, and neither does your sales rep.

We priced two other architectures on the way. A hybrid inverter setup—the EG4 6000XP hybrid inverter came up repeatedly in forums and in two of our quotes—is a 48V, 6 kW-class unit aimed at off-grid and small hybrid systems. Our main building is 480V three-phase with a 410 kW peak, so it was never in scope there. It could make sense for a detached site office or a yard building. Different tool.

And at the small end, a Jackery 1500 watt solar generator (the model everyone forwards me) is genuinely useful for a job trailer or temporary site power. It's not a building backup system and I'd be doing you a disservice to pretend otherwise. Match the tool to the load.

Checkpoint: one paragraph stating your use case, required runtime, and required continuous kW. If you can't write it, you're not ready to get quotes.

Step 4: Size solar against the load curve, not the roof area

We evaluated Tesla Solar Roof on one low-rise building because two tenants asked about it by name. It's a legitimate product, and on a pitched, visible roof with clean geometry, the economics and the aesthetics both work in its favor.

Ours was a low-slope commercial roof with HVAC penetrations, drains, and a maintenance access path. The tile format added install complexity and cost that didn't clear our payback threshold in the model. So we went with conventional modules on that building. That's not a knock on the product—it's a statement about our roof and our load profile.

The part that actually moved our numbers was matching generation to the daytime load curve rather than to available square footage. Filling the roof looks good on a proposal and exports power at a rate that's usually worse than what you'd save by consuming it on site.

Checkpoint: a 12-month production model paired with your interval load data, plus a structural load report from someone who has physically been on the roof.

Step 5: Plan the charging install around parking behavior, not headline kW

On how to install EV charging station hardware at a commercial site, the biggest decision isn't the equipment brand—it's whether you put capacity behind the meter with load management, or upgrade the service. We did the latter on one site and regretted it once we saw the transformer timeline. On the second site we used metered, load-managed Level 2 charging and skipped the upgrade entirely.

Some numbers from our own site, for context: 6–7 kW per port covered the overwhelming majority of dwell times in our lots. We spent our budget on more ports at lower per-port power rather than fewer ports at high power, and utilization went up. Publicly quoted commercial Level 2 install costs generally land in the $2,000–8,000 per port range depending heavily on trenching, panel distance, and utility work (based on publicly listed figures, 2024–2025; verify current pricing for your site—it varies enormously).

Checkpoint: assumptions on simultaneous usage, per-port power, payment handling, and ADA-accessible stall counts, written down before you sign. The payment system is the most commonly underestimated line item in the budget.

Step 6: Pay for a locked delivery date, not a target

Here's something contractors won't volunteer: standard turnaround times often include buffer the contractor uses to sequence their crews across multiple jobs. It isn't necessarily how long your work takes, and a "target date" is a scheduling intention, not a commitment.

We got this wrong on the first building. The contractor said "sometime in Q3" and we planned tenant communications around it. The site went live nine weeks late, which cost us a $28,000 concession on a lease tied to charging availability. On the second building, we paid a $9,400 scheduling hold fee to lock a firm date and crew allocation. To be fair, that's real money and it isn't always justified. But when I compared the two projects side by side—same scope roughly, one floating and one locked—the difference wasn't speed. It was that we could actually plan around one of them.

In my experience, "probably on time" is the most expensive phrase in a construction contract. Paying to remove that uncertainty is a line item, not a gratuity.

Checkpoint: a signed date with a defined penalty or credit, a named crew allocation, and the milestone dates for utility approval and inspection written into the schedule.

Step 7: Meter and commission before you release final payment

"It turned on" is not commissioning. We hold 10% until we have four things: an independent production/consumption meter, as-built drawings that match the installed system, a commissioning report with the actual measured output, and the warranty start date documented in writing.

The warranty date is the quiet one. Ours started at substantial completion, not at the date we first used the system—and nobody mentioned the gap until I asked. Read that clause.

Checkpoint: all four documents in hand, and the meter data reviewed for 30 days before final payment releases.

What I'd tighten up next time

  • Put the transformer lead time on the project schedule in week one, not week nine. (Note to self: I've now written this in three post-mortems and done it late twice.)
  • Get the utility capacity confirmation before the equipment deposit, not after. A deposit on hardware you can't energize is a very expensive paperweight.
  • Don't let a vendor's free survey substitute for an engineering study. They're different deliverables from different parties.
  • Ask each shortlisted contractor for their last three projects' actual completion date versus contracted date. The gap tells you more than any reference call.
  • Budget for the payment system and network connectivity. It's the line item everyone forgets and it's never free.

None of this is glamorous. But the difference between the project that missed its date and the one that hit it wasn't the battery, the modules, or the chargers. It was the order in which we did things.

Pricing and lead-time figures cited above reflect publicly available information as of early 2025 and our own project records. Verify current rates and timelines with your utility and local contractors before committing budget—regional and site-specific variation is significant.