Feature Flag Management & Progressive Delivery3 min readUpdated September 2026

Building a Rollout Worksheet for a Tenant Portal Update

A property management company's engineering footprint is usually a tenant or resident portal, a maintenance request app, and maybe a payments integration, often maintained by a small internal team or a vendor. The rollout question that actually matters here isn't experimentation, it's whether you can change one property's experience without touching every other property at once, especially during a lease-renewal season when nobody wants surprises.

Here's a worksheet to walk through before picking LaunchDarkly or Split.

Most property management companies don't need a sophisticated platform decision here; they need a disciplined process for rolling changes out property by property, and either tool can support that once the process itself is clear.

This holds whether you manage five buildings or fifty. The process, pilot first, watch a full cycle, expand deliberately, matters more than which specific platform you choose, and it's worth getting the process right on a small portfolio before it has to work at a larger one.

Worksheet, part one: what actually varies property by property?

  • List every feature in your portal that behaves differently, or should, across your properties (late fee policy, maintenance request routing, amenity booking)
  • Note which of those differences are currently hardcoded per property versus managed through a shared setting
  • Identify which properties are mid-lease-renewal-cycle right now, since that's the worst time to introduce an untested change

This list tells you how much of your rollout problem is really a targeting problem a flag platform can solve.

Worksheet, part two: who needs to approve a change before it reaches residents?

  • Property managers who need visibility into what's changing for their building
  • Whoever handles resident support and will field the first calls if something breaks
  • Anyone responsible for the payments integration, if the change touches billing at all

If this list has more than two people, LaunchDarkly's approval workflow is worth the setup time. If it's genuinely just one person making the call, a simpler manual toggle may be all you need.

Why targeting by property, not by percentage, usually fits better here

A random percentage rollout doesn't map to how property portfolios actually operate; a change tested on two pilot properties, watched for a full billing or maintenance cycle, then expanded, matches how property managers already think about rolling out anything new. Both platforms support this kind of explicit targeting, so the deciding factor is usually cost and simplicity rather than a feature gap.

Avoiding a change during lease-renewal season

Renewal season is when residents are most sensitive to anything that feels different, so a portal change timed to land during that window, even a good one, tends to generate support calls that have nothing to do with the change's actual quality. Schedule flag rollouts around each property's renewal calendar, not a company-wide release schedule, if your portfolio spans properties with staggered renewal timing.

What to do when one property's manager pushes back on a pilot

A property manager who's skeptical of a pilot change, often because they've seen a past rollout go badly, deserves a real conversation about what will be different this time, not just a mandate from corporate. Give that manager a way to flag a problem quickly during the pilot window, and honor a rollback request from them without requiring escalation, since their day-to-day view of resident reactions is a genuinely useful signal.

Keeping the payments integration separate from everything else

If your portal's payments integration is flag-gated at all, treat it with more caution than any other feature, since a billing mistake affects residents' money directly and generates the most urgent support calls. Test payment-related changes on a longer pilot window than other features, and require a second person's review before expanding any payments-related flag beyond the initial pilot properties.

What resident feedback during a pilot actually tells you

A spike in maintenance requests or support calls during a pilot doesn't always mean the change itself is broken; sometimes it just means residents are asking questions about something new. Distinguish between confused residents who need better communication about a change and residents reporting an actual bug, since the fix for the first is a clearer notice, not a rollback, while the second genuinely needs one.

How this decision changes as the property portfolio grows

A company with five properties can manage flag targeting with a simple spreadsheet tracking which properties are on which version, but that approach breaks down well before fifty properties, when manual tracking becomes its own source of errors. Revisit this decision as the portfolio grows rather than assuming whatever worked at five properties will still work cleanly at fifty; the platform that felt like overkill early on may become the simpler option later.

Executive Capability Standard

What Good Looks Like

A property management company can roll a portal or policy change out to two or three pilot properties, watch a full cycle, and expand or roll back without touching any other property's resident experience.

Building The Capability (5-Stage Skill Ladder)

1. Learn:Complete the property-by-property variance worksheet before evaluating any platform.
2. Do Manually:Pilot one upcoming change manually on two properties with different profiles, tracking support call volume as your signal.
3. Delegate:Assign one property manager per pilot to own resident communication and feedback during the test window.
4. Automate:Move property-scoped targeting into LaunchDarkly or Split once pilots consistently show a need for faster rollback.
5. Buy:Build renewal-season timing directly into your rollout calendar so flag changes never land during a property's sensitive window.

How to Get Started

Frequently Asked Questions

Do we need Split's experimentation features to know if a portal change is working?

Probably not. Most property management portal changes are judged by support call volume and manager feedback rather than a formal metric, which is a simpler bar than what Split's experimentation layer is built for. LaunchDarkly's straightforward targeting is usually enough unless you're running a genuine A/B test on something like payment reminder timing.

How many pilot properties should we test a change on before expanding?

Two or three properties with different profiles (different sizes, different resident demographics, different amenity sets) gives you a more useful signal than testing on ten similar ones. Watch a full billing or maintenance cycle before expanding further, since many issues only surface once, not immediately.

When should we avoid rolling out a portal change to residents?

Avoid renewal season. Residents are most sensitive to anything that feels different then, so even a good portal change can generate support calls that have nothing to do with its quality. Schedule flag rollouts around each property's renewal window, and give anything touching payments a longer pilot.

About the numbers

This guide doesn't quote a sourced benchmark. Figures in it are estimates or general guidance, so check them against your own numbers.

Related Guides