Parking utilization studies are among the most common analytical exercises in parking management — and among the most frequently done poorly. A utilization study that uses the wrong observation windows, inadequate sampling frequency, or inconsistent counting methodology can produce data that actively misleads planning decisions.
This guide covers how to design a study that produces reliable, actionable data, what the numbers actually mean once you have them, and the most common interpretation errors that lead operators and planners astray.
Why Utilization Studies Matter
The fundamental question a utilization study answers is: how much of our parking supply is actually being used, when, and where?
That sounds simple. The implications are not. Underutilized parking represents capital tied up in asphalt and concrete generating less revenue than it could. Overutilized parking — facilities routinely exceeding 85 percent occupancy during peak periods — causes measurable congestion, customer frustration, and in some contexts, a diversion of demand to less appropriate locations.
In urban planning contexts, utilization data is foundational to minimum parking requirement reform, shared parking agreements, and transportation demand management programs. Without it, decisions are made on assumption and anecdote.
Defining the Study Purpose First
Before designing any study methodology, define what decision the data will inform. This sounds obvious but is frequently skipped, resulting in studies that collect the wrong data for the actual question.
Common study purposes and their methodological implications:
Evaluating whether supply is adequate — requires peak period observations across multiple days, with enough temporal resolution to identify true peaks vs. transient spikes.
Informing a shared parking agreement — requires time-of-day profiles for each party’s demand, including how demand patterns overlap or complement each other.
Supporting a pricing or rate structure decision — requires occupancy data segmented by price zone, user type if possible, and duration. Turnover data is as important as occupancy.
Justifying a development project parking reduction — requires data sufficient to satisfy local regulatory standards, which often specify minimum observation periods and methodologies explicitly.
Identifying operational problems — may require finer spatial granularity, looking at section-level or aisle-level occupancy to identify localized congestion or avoidance patterns.
The study design that serves one purpose may be entirely inadequate for another.
Core Methodological Decisions
Observation Windows
The observation window — the days and hours during which data is collected — is the most consequential design decision. A study that only observes Tuesday through Thursday misses weekend patterns. A study that starts observations at 8am misses early-morning demand. A study that ends at 6pm misses evening utilization in mixed-use environments.
Rules of thumb that are generally defensible:
- Observe at minimum a full standard business week (Monday through Friday) plus one weekend day
- Cover the full operating window of the facility, from opening to closing
- Include at least one observation during any known peak period (lunch for retail-adjacent, evening for entertainment-adjacent, etc.)
- If demand is seasonal, note that a single-season study may not represent annual patterns
For facilities with highly variable demand (event parking, transit park-and-ride), a single-week study is rarely sufficient. Multiple observation periods capturing different demand conditions are needed.
Observation Frequency
How often you count determines how precisely you can characterize the occupancy curve across the day. The right frequency depends on how quickly occupancy changes.
In facilities with high turnover — retail, transit station park-and-ride — occupancy can change substantially in 15 to 30 minutes. Hourly snapshots will miss peaks and understate utilization variability.
In facilities with low turnover — commuter parking, residential — hourly or even two-hour snapshots may be sufficient because occupancy is relatively stable during the core occupied period.
As a practical matter, manual observation at 30-minute intervals across a full day is labor-intensive. Automated counting technology — either entry/exit counters or space-level sensors — makes high-frequency data collection tractable and much cheaper at scale. Facility parking managers who have implemented automated counting systems consistently report that the data quality improvement over manual studies justifies the investment within a few years.
Counting Methodology
For manual studies, consistent methodology across observers is critical. Key issues:
What counts as occupied? A vehicle in a space is occupied. A vehicle partially blocking a space? A maintenance vehicle? A reserved space with no vehicle? Define these cases in advance and train all observers consistently.
Space inventory accuracy. The occupancy percentage is only meaningful against an accurate denominator. Verify the space count in the inventory against the actual physical count before the study begins. Spaces that are temporarily blocked by construction, signage changes, or repurposing affect the denominator.
Spatial granularity. Aggregate counts at the facility level are useful for high-level planning but mask local distribution effects. A 70 percent average occupancy facility may have one section consistently at 95 percent and another at 40 percent — information that’s invisible at the aggregate level but operationally significant.
Technology-Assisted Studies
Automated sensor systems change what’s possible in utilization analysis. Rather than periodic snapshots, continuous occupancy data captures the full demand curve with no sampling gaps.
Space-level sensors provide occupancy state for individual spaces. Entry/exit counters provide aggregate counts but not spatial distribution within a facility. License plate recognition (LPR) systems can reconstruct occupancy from entry and exit events and, with appropriate data handling, provide duration distributions as well.
The tradeoff is upfront cost. For a one-time planning study, the cost of installing sensors may not be justified. For ongoing operational management, the sensor data pays dividends far beyond any single study.
For technology-assisted data collection, attention shifts to data quality management: sensor calibration, handling of edge cases (double-parking, tailgating at entry lanes), and validation against periodic manual counts to confirm the automated count is accurate.
Interpreting the Numbers
The 85 Percent Rule of Thumb
A commonly cited benchmark is that parking facilities operating above 85 percent peak occupancy are effectively full from a user experience perspective — circulating drivers can’t reliably find spaces, generating congestion and frustration. This benchmark has a reasonable empirical basis and is widely used in planning contexts.
But it’s a rule of thumb, not a law. It assumes random search behavior among drivers who don’t know in advance where vacant spaces are. In facilities with good wayfinding, guidance systems, or reservation systems, effective capacity extends higher. In facilities with poor layout or wayfinding, the practical constraint may hit lower.
Apply the 85 percent threshold as a starting point for investigation, not a hard conclusion.
Duration and Turnover
Occupancy snapshots don’t tell you how long vehicles are staying, which matters for both revenue and capacity management. A facility at 70 percent occupancy with average stays of 45 minutes is generating far more transactions — and likely more revenue — than one at 70 percent occupancy with average stays of 8 hours.
Duration data requires either LPR reconstruction, transactional data from paid parking systems, or dedicated duration studies (tracking individual vehicles across multiple observation passes). It’s worth the additional effort for any study informing pricing or supply decisions.
Comparing Across Studies
When comparing utilization data across time periods or facilities, verify that the methodology is consistent. A study using 30-minute snapshot intervals isn’t directly comparable to one using hourly intervals. Studies conducted at different seasons, or before and after significant neighborhood changes, aren’t comparable without controlling for those differences.
Parking Professional has published guidance on standardizing utilization study methodologies for portfolio-level comparison — useful for operators managing multiple facilities who want consistent data across the portfolio.
What Utilization Studies Don’t Tell You
Utilization data describes what happened, not why. A facility with lower-than-expected utilization might have inadequate signage, poor location relative to demand generators, pricing that’s too high, competition from nearby alternatives, or simply less demand than anticipated. Utilization data identifies that a problem exists; diagnosing its cause requires additional investigation.
Similarly, utilization data doesn’t predict future demand. Development patterns, transit investments, pricing changes, and land use shifts all affect parking demand in ways that historical occupancy data can’t anticipate. Study results should be framed accordingly when used in planning documents.
Reporting and Documentation
A utilization study that produces good data but presents it poorly is almost as useless as a badly designed study. Effective reporting includes:
- Clear documentation of methodology, including observation windows, counting frequency, and observer training
- Presentation of data at multiple temporal and spatial scales — aggregate facility level, section level, and time-of-day profiles
- Statistical characterization of variability, not just average or peak figures
- Explicit statement of limitations and conditions under which the findings apply
- Comparison to relevant benchmarks or standards where applicable
The study methodology should be reproducible — another team following the same documentation should be able to replicate the study and get comparable results.
Further reading on parking utilization methodology and industry benchmarks is available through parkingprofessional.com and through the International Parking & Mobility Institute.