Data-Driven Utilization Forecasting: The Overlooked Pillar of EV Charging Success

(Part 1)

Electric vehicle (EV) charging infrastructure is growing at a record pace. Cities, utilities, EPCs, and charge point operators (CPOs) are racing to meet ambitious electrification targets. Yet despite the momentum, a surprising number of EV charging projects continue to fall short, whether due to over-building, under-utilization, or misalignment with actual vehicle behavior.

One of the most frequently overlooked culprits? Poor charger utilization forecasting. In this article, we will discuss this second of five pillars essential for consideration when planning an EV project. (You can read our opening article in this series about the five pillars here, and read about pillar #1 - Financial Modeling - here.)

The Invisible Risk: Misjudging Utilization

In our discussions with EPCs and CPOs, we’ve found that most understand the importance of site selection and grid interconnection, but we’re often surprised to hear about how few projects devote equal rigor to utilization forecasting. And yet, this is a key pillar that directly influences the success or failure of an EV charging site.

What happens when this is neglected?

• Chargers sit idle for weeks or months with insufficient demand;

• Or on the other side of the coin, queues back up due to under-provisioned infrastructure, damaging user experience and operational efficiency;

• In commercial fleets, even modest underestimations of utilization can ripple into delayed routes, unplanned charging windows, or downtime that erodes ROI.

These are not rare cases happening on the margins - they are increasingly common, leading to very preventable failures stemming from outdated modeling techniques or blanket assumptions applied across different geographies, fleet types, and usage profiles.

Why Many Projects Get Utilization Forecasting Wrong

The typical approach to forecasting charging behavior is static and overly simplified. It often relies on national EV adoption rates, generic assumptions about vehicle turnover, or rough averages for charging duration. Unfortunately, these high-level estimates don't translate into actionable forecasts at the site level.

Some of the factors we’ve discovered that are commonly overlooked include:

• Local traffic patterns that impact visit frequency and dwell times;

• Fleet shift schedules that concentrate charging into narrow time windows;

• Seasonal or event-based demand surges in commercial hubs or tourist zones;

• Charging behavior differences between light-duty, medium-duty, and heavy-duty fleets, and;

• Behavioral quirks - such as driver “top-off” tendencies or queuing aversion - that change how stations are used in practice.

• Utility Demand Fee set points may reflect a 1-time “snap shot” charging event, making considering the need for / viability of battery storage to mitigate / avoid Demand Fee triggering events invaluable.

When these nuances are missed, the risk isn’t just academic. Misaligning capacity with actual demand impacts capital efficiency, user satisfaction, and long-term site viability.

The Cost of Over-building vs. Under-building

There is a critical financial trade-off at play. Overbuilding leads to stranded assets and sunk capital, while under-building creates bottlenecks that limit throughput and utilization which in turn negatively impacts revenue and ROI. Neither option is acceptable for infrastructure expected to scale over 5 to 15 years.

• Over-building = Unused chargers, increased maintenance costs, and negative optics with investors or public stakeholders.

• Under-building = Frustrated drivers or fleet managers, congested queues, and lost revenue due to unmet demand.

Neither option is acceptable for infrastructure expected to scale over 5 to 15 years. For commercial fleet operators, poor utilization modeling hits even harder:

• It can disrupt route reliability and charging windows for delivery schedules.

• It reduces asset utilization (vehicles waiting to charge, rather than delivering value).

• It causes higher-than-expected operational costs, as vehicles are rerouted or idled.

In all of these scenarios, incorrect or sub-optimal utilization modeling becomes a bottleneck not only to performance, but to future scale.

Existing Tools Are A Valuable First Step in the Right Direction - But That’s Not Enough

It would obviously be unfair to say the industry has ignored this problem entirely. Several tools now offer charger siting optimization and demand modeling. These include GIS-based platforms, utility incentive calculators, and charger OEM planning tools. And indeed, they represent a valuable first step. Many help planners:

• Identify general charging demand zones.

• Estimate how many chargers might be needed.

• Evaluate spacing and proximity to existing infrastructure.

But these tools are rarely tailored to fleet behavior, time-of-day usage, or local vehicle activity. Their forecasts are often based on aggregate demand, without incorporating site-specific behavior patterns that are vital to true capacity planning. That’s a limitation with real consequences. For example:

• A station near a logistics hub might see intense usage from 5–9 a.m. and again at 5–9 p.m., but near-zero activity in between.

• A fleet depot operating 24/7 in rotating shifts may require round-the-clock charging access — with downtime scheduled by design.

• A retail site may see weekend peaks driven by shoppers, not weekday commuters.

Without the ability to simulate these patterns, decision-makers miss the mark on the right number, type, and timing of chargers.

Key Developments Reinforcing the Need for Granular Forecasting

Several macro-level trends further emphasize why utilization forecasting must evolve from a back-of-the-napkin exercise to a data-rich discipline:

1. Shift from Pilot to Scale

As charging infrastructure scales, the stakes grow. Scaling means new locations, customer types, and utility interconnection challenges. What worked at a handful of pilot sites may not work across dozens of diverse locations.

2. Tightening Capital Constraints

Increased interest rates and a shift toward profitable, ROI-positive projects make it harder to justify speculative charging infrastructure. Every charger must earn its keep.

3. Growing Fleet Electrification

Fleet managers require certainty. They need to know a site won’t bottleneck vehicles and disrupt operations. Predictability matters — and only data-backed forecasts can offer it.

4. Planning for Multi-Use Sites

Public and semi-public sites are increasingly multi-use — with passenger EVs, delivery vans, and even ride-share drivers sharing space. Modeling for such mixed utilization is complex but vital.

5. Incentive Alignment

Many public funding programs require utilization metrics as part of reporting or ROI modeling. Projects that cannot show likely throughput may not qualify.

Coming Up in Part 2

In the second part of this 2 part article, we’ll explore how a more systemic approach to project analysis can address these issues. Using the Brighterge platform as an example of a platform that has adopted such a systemic approach, we’ll explore:

• How Brightmerge’s utilization forecasting works.

• The kinds of parameters it includes (e.g., shift schedules, fleet rotation windows, dwell times).

• Examples of how these forecasts influence infrastructure planning, site optimization, and capital allocation.

• How utilization insights connect to other critical pillars like smart charging and ongoing asset optimization.

Utilization forecasting isn’t just a box to check. It’s a critical data layer in a systems-level approach to charging infrastructure that works — financially, operationally, and strategically.

Stay tuned.