Best Fits for Field & Central Teams
Every real estate services business starts with one guy or gal in a vehicle meeting customers where they are, delivering service where they need to, and writing down just enough to get by on a yellow legal pad. When the operating businesses grows, to achieve scale with consistent performance across individuals and markets, successful organizations shift some functions to centralized, desk-based teams with multiple screens, and let field team members focus on the things that let them deliver the most value.
Redfin, the technology-powered real estate brokerage, has spent 15 years building their infrastructure to deliver a robust and consistent sales experience for customers in 90+ markets. The big idea in Redfin’s core business is to hire very good real estate agents as full-time employees, and have them focus as much of their time as possible on salesmanship and customer experience. Redfin has tried to extract away most administrative and standard work to pair a Lead Agent with an army of support staff:
Listing coordinators who handle the paperwork and emails to get the home on the market);
Transaction coordinators who take over to handle the paperwork of closing on a buy-side or sell-side deal;
Support agents, real estate licensees who can help answer property and business questions over the phone;
Tour coordinators, non-licensed team members who triage new potential customers and route them on for touring; and
Associate agents, real estate licensees who conduct property tours and other field activities so the lead agent doesn’t have to do all the legwork.
To facilitate the interactions between these six roles, Redfin has invested heavily in a custom CRM called Agent Tools that lets a Lead Agent and her per-deal coordinator(s) run the same ~100-step playbook for every home they help a customer buy or sell. Agent Tools is incredibly robust, and it’s organized around the principle of making all the tasks and deadlines visible on a given deal, and clarifying who is responsible for each one.
American Homes 4 Rent (now rebranded as AMH, matching their stock ticker) built a property management organization as it scaled to 50K+ single-family rental homes. We had portfolios in 21 markets, but with different levels of density; we built functions that needed to run similarly in the 4000-home Dallas market and the 250-home Portland market, and we spent years refining the set of functions that should be done centrally–things like tenant underwriting and auditing marketing photos for web syndication–and how we could best perform the set of functions that should be done locally–in-person move-in and move-out appointments and renovation project management. For maintenance, utilities, property management, collections, and other functions, we experimented with central and local capabilities until we had a winning formula.
Take Aways
The key to how we landed on the “right” set of responsibilities was to document the workflow thoroughly. We then assessed the traits and value-creation in each step to see where field employees could make the biggest difference, and where central employees were needed to deliver the most consistency in execution. Here are some principals I’ve followed in setting responsibilities for field and central roles:
Empower the person in the field who is closest to the problem to solve it, and support that person with policy and infrastructure.
If it takes less time to solve the in-person problem now (because the person in the field can see it and fix it), just fix it and don’t write it down – this is the Getting Things Done philosophy in action.
Don’t build too much specialization in the field until you’re sure the new specialty can support two people in a market. Without volume to support the adequate utilization of two team members, you’re better off cross-training so you’re not left in the lurch when the one person whose job it is to put a yard sign in the ground and add a lockbox goes on vacation.
If you need a quick estimate with high tolerance for variability, build it in the field; if you need a detailed output that is work-ready, build it centrally.
If you need to deliver on a consistent SLA for turnaround time, centralize it to someone whose day is less randomized than a field technician.
If documentation must be done by a field technician, make sure it’s straightforward enough, and that you set the expectation that he must submit it in the driveway before driving to the next site or it will never be done well.
As you set speed SLAs, consider the frequency you’re trying to hit–delivering within the target speed SLA at 90% is going to be much, much cheaper than staffing to be able to hit it 99% of the time.
Set speed SLAs in days rather than business hours until you’ve solved most of your other operational problems. Systems like Zendesk don’t support tracking performance down to +/-4 business hours very well; you may need to invest in different infrastructure if knowing your SLA to the hour is important.
Be consistent in your metrics: when you build reporting, pick business days or calendar days for everything; making stakeholders switch between 5- and 7-day weeks leaves lots of room for confusion and misalignment.
Never put a ticking clock on screen to show a team member how long they’ve been on a given task. It's much more valuable to provide prioritized queues to let folks claim and fulfill new tickets, and to surface “time till SLA expires,” which lets them focus on a clear goal (“don’t blow the SLA”). Focusing on “time within performance goal” gives teams a goal to beat, instead of a vaguely big brother style “I’m just curious how long it takes you to do this–no particular reason.” When you realize it’s the wrong goal based on your time study or client needs, fix the goal, but keep it foremost in the system.
For central functions, it’s important to measure speed and quality–if people only care about speed, it’s pretty easy to hit the goal (especially if you make that goal visible). Without measuring the quality of the output, the field team may not trust it and may spend time reworking the product, costing you more than if they’d just done it in the first place.