Most commercial real estate owners can tell you the net operating income of each asset in their portfolio, but few can explain why one property consistently outperforms another or why maintenance costs at a building in Dallas run 30 percent higher than a nearly identical asset in Phoenix. The reason, according to a new analysis, is not bad management but the absence of a data strategy designed for the portfolio level.
In commercial real estate, data has traditionally been viewed as a property-by-property matter. Owners log into separate lease management platforms, pull reports for each building, and piece together a rough picture of portfolio performance. However, that picture rarely shows the why behind the numbers, leading to decisions about capital allocation, vendor contracts, and operational priorities made largely on instinct.
According to Bill Douglas, CEO of OpticWise, the shift to portfolio-level intelligence starts with treating each building as a data-generating node in a larger network. “You look at the property as one data point, but there’s a data lake in it,” says Douglas. “How can I compare whatever it is you want? And when you start using large language models across those data sets, you see correlations that are just astounding.”
When operational data flows across a portfolio instead of sitting siloed in individual vendor platforms, patterns emerge that no property manager would spot manually. For example, a particular brand of rooftop HVAC unit that starts failing at year 12, a class of building where lighting costs spike in winter due to improperly configured timers, or a portfolio-wide opportunity to renegotiate a vendor support contract based on actual failure frequency.
The limiting factor for most portfolios is not tools but data ownership. Most CRE owners do not actually hold their own operational data; it lives in vendor clouds—the property management platform, the leasing system, the parking software, the access control provider. Owners can log in and run reports but do not possess the raw data in a form that allows cross-system or cross-asset analysis. Douglas notes, “If all you do is take your P and L from each building and look at the bottom line, you’re missing a lot of the drivers that have impact. You’re looking at the result rather than the cause.”
For a portfolio of 50 assets, each property likely runs 12 to 15 systems generating data around the clock, creating a massive volume of operational events every month sitting in separate silos, invisible to each other and to the owner. When data is trapped, teams spend time reacting—slower work orders, more vendor finger-pointing, and a tenant experience that quietly erodes renewals.
The concept of “Champion” in the Peak Property Performance framework is written specifically for owners and asset managers operating at the portfolio level. Borrowed from a sports analogy, the best owners are in the skybox, not on the field, looking across the whole game. That skybox view means being able to answer questions like: which assets are aging into capital replacement cycles? Which properties are consuming utilities above the benchmark and why? Where is tenant satisfaction trending down before it shows up in lease renewals?
Portfolio-level intelligence starts with a data and digital infrastructure audit, property by property, to establish what data exists, where it lives, and what it would take to bring it under owner control. From there, owners can begin connecting the dots, first within a single asset, then across the portfolio, and then into predictive, cross-portfolio analysis. The properties generating the highest returns are not doing it by luck; they are doing it because someone decided to stop looking at the scoreboard and start understanding the game.
