A platform that reports millions of units has not necessarily reported millions of customers or millions of tenants. A customer can be an organization managing many properties. A property can contain many units. A unit can be vacant or house more than one person. These differences are easy to explain in words, yet they are frequently lost when software vendors are placed into a single largest-platform chart.
The TenantPlatform.com scale section separates each type of operating metric and preserves the original definition and date. It does not convert units to people or sum overlapping vendors into a market total. This guide explains how to read the data and design a more defensible comparison. It also shows why leaving a number blank can be more informative than filling the gap with a superficially plausible estimate.
Start by defining the customer
Customer is a business-specific term. It may mean a paying management organization, a client account, or a group-wide relationship across several products. It does not automatically refer to individual residents. Ask what event makes an organization count as a customer and whether different accounts within the same organization can be counted separately. Those questions are important whenever customer figures are used to compare companies with different product portfolios.
Read the customer definition
The AppFolio profile identifies property-management customers as the reported population. MRI's supplied figure refers to global group clients across a wider range of real-estate products. RealPage's figure is similarly group-wide. These metrics can all be useful context, but their definitions are not identical. A comparison should preserve those labels instead of simplifying every column heading to users.
Distinguish properties from units
A property or community can contain one rental home or many apartments, while a unit describes a rentable space under the vendor's definition. The relationship between them depends on the portfolio. Multiplying a community count by an assumed average number of apartments would introduce a new estimate that the source has not supplied. Such a calculation should not appear as a verified vendor footprint.
Advertising properties add another distinction. The Apartments.com record describes paying advertising properties rather than an independently verified count of buildings operationally managed through Rental Manager. Advertising participation and software management are different activities. The marketplace comparison therefore shows marketplace-scale measures separately from management-unit measures. A building's appearance in a marketplace does not establish that its accounting or tenant services run on that platform.
Treat listings as inventory, not occupancy
An active rental listing represents inventory offered to prospective renters under the source's definition. It is not necessarily a unique occupied unit, a signed tenancy, or a current resident. Listing figures may also be measured as averages over a period rather than a point-in-time count. The date and measurement method must travel with the number if it is reproduced in an article, dashboard, or comparison card.
The Zillow profile records an average monthly active-listing measure alongside a separately described multifamily-property measure. Neither is relabeled as managed tenants. A useful marketplace evaluation may consider discovery and inventory context, but those measures answer a different question from how many core accounting units a management platform serves. Comparing them directly would conceal the difference in both activity and measurement.
Keep leases and residents separate
A lease is a contract record, not a person. A resident count refers to people only when the source explicitly reports such a population. One lease can involve multiple occupants, and multiple records can relate to one resident over time. The supplied dataset does not assume a household size or occupancy rate to manufacture tenant counts. Where an actual resident figure was not verified, the numeric cell remains empty.
Entrata's profile illustrates why even a published resident measure needs a scope note. Its broader ecosystem resident figure is recorded separately from a more narrowly defined billed core-unit figure. Those numbers have different dates and populations. They should not be divided to infer average household size or combined to infer unique active app users. The resident-access directory focuses on functionality rather than pretending every platform publishes a comparable tenant population.
Inspect whether the count is core or group-wide
A broad software company can serve properties through several products that do not all use the same operating system. A units-served claim may include ancillary products, separate acquired platforms, or different property types. Ask whether the measure describes the proposed core product or the wider corporate ecosystem. A large group footprint is not necessarily a count of accounts on the exact software configuration being evaluated.
RealPage's recorded housing-unit footprint is group-wide and may overlap with Buildium. MRI's recorded figures span global product and property categories. Yardi and Rent Manager have unit figures relayed through a third-party source from vendor statements. Each source can provide useful context, but each has a different evidentiary boundary. The source register keeps those limitations visible rather than turning every figure into an independently audited core-system count.
Do not erase the measurement date
A precise historical number can be less current than a broader recent statement. Buildium's acquisition-era customer and unit benchmarks in the dataset are explicitly dated to 2019. They are not presented as a current estimate simply because the profile was published or reviewed in 2026. A research cutoff tells the reader when the collection was assembled; it does not refresh every underlying measurement.
For a filing-related example, AppFolio's official second-quarter 2026 results are among the source records associated with its recent reporting. The dataset links the relevant filings and supporting unit evidence separately. When a source supplies no explicit measurement date, state that uncertainty. An undated marketing figure should not silently acquire the website's current date.
Preserve qualifiers and avoid false precision
More than ten thousand is not the same statement as exactly ten thousand. The CSV uses separate operator fields to preserve qualifiers such as more than or approximately. Its website displays those operators where relevant. This is especially important when figures are turned into visual bars or sorted lists, because a neat integer can conceal that the original source described only a lower bound.
Qualitative descriptions should remain qualitative when an exact number is not established. DoorLoop's record does not manufacture a precise customer or unit count from broad company language. Likewise, blank values elsewhere do not imply that a company has no customers. A data-quality note should explain whether a field is undisclosed, not applicable, historical, or not verified. Those statuses are evidence, not defects to conceal for visual consistency.
Use scale for the questions it can answer
Scale can indicate the breadth of a company's reported presence, but it does not directly measure customer satisfaction, implementation quality, product reliability, or the experience of one resident. Those require different evidence. A buyer can reasonably ask for references from portfolios with similar requirements, while keeping vendor-wide scale as contextual information. Avoid assigning a performance score based only on the largest published number.
A credible comparison starts with the metric, not the ranking. Use the scale glossary to identify the population, then inspect date, scope, qualifier, and source quality. Keep corporate overlap in view and resist the temptation to calculate a total across heterogeneous counts. The result may be less dramatic than a single biggest-platform chart, but it gives a reader a much more accurate understanding of what the research actually establishes.
Before sharing a scale chart, check whether a reader can understand every bar without opening another document. The label should identify the metric and scope, while a nearby note should identify the date and any important qualification. When those requirements make a chart too crowded, a well-labeled table is the better format. Presentation should preserve the evidence rather than compress different populations into a misleading visual hierarchy.
