Thread by @somehotelguy

Next up in the @somehotelguy series on new development… Facilities Programming!! As always, stream of consciousness (a.k.a., long).

Here we’ll talk about an incredibly iterative process – how you figure out just what you should build when you’re looking at a hotel site.

I’ll do this from my perspective as an investments guy, but CRITICAL to the actual success of this process: involve someone with deep zoning knowledge, someone with deep HOTEL architecture knowledge, and someone with deep operating knowledge.

Ideally, all four of you know the market well also (or at least reasonably well).

Before we start on planning the physical facilities, we need to get our heads around a few things: (a) how much (roughly) can I build on this site?

(b) what are the competitive hotels in the area? (c) who am I catering to? (d) what’s my market and target RevPAR?

Here’s why.

(a) Need to know how much you can build – hoping the reason was obvious here, but at this stage you’re looking for a high-level estimate of buildable SF allowed under local zoning regs. Don’t go too deep at step 1, as we’ll iterate through this a few times.

(b) Know your competitors – you’re going to be trying to steal business from local hotels; know where they are, what they offer, and on who’s back you’d like to place your target(s).

(c) Know your target audience – when you know why people are coming to your market, you can start to figure out what they need in physical product terms.

(d) Know your potential performance – this will help you know if your amenity offering, service level, room size(s), etc., etc., etc., are all supportable by the market. No sense building the quality of a Four Seasons in a market where rates resemble a Hampton Inn.

The VERY FIRST cocktail napkin test I run – knowing my (rough) buildable square feet and my targeted chainscale / service type (e.g., upper upscale full service), I’ll take a swag at how many keys I can get.

For luxury, I typically assume 1,000 gross SF per key; for higher end upper upscale 700 – 750; for upscale 550 (extended stay a bit larger); etc. on down. (varies for urban core, resort, etc.)

So, if I have 115,000 buildable and I’m targeting that full service upper upscale space, I think I can get about 155 to 165 keys. This varies CONSIDERABLY based on other offerings (most especially meeting space) but is a good Day 1 swag.

When pulling (a) through (d) above together, I’ll try and put together a grid of sorts, outlining all the facilities details on my local competitors (e.g., room count, suite mix, meeting space, etc.), how those competitors perform (if possible), and highlight which of each…

…makes the most sense for my target guest (e.g., the suite mix from the Marriott, the fitness center from the Hilton, etc.). Now I can compare that against my swag’d key count and see if I’m in the realm of being competitive!

Even better, I have some basic market intel and can use some swag’d ratios to see if a hotel of my size makes any sense!

RevPAR times room count times 365, divided by the percent of revenue I’d expect to see from rooms, times my typical NOI margin for that product type gets me a super back of napkin read on “how will this do?”.

If I then know roundly what cap rates are in that market, I can get a stabilized value. Talking to architects, GCs, industry friends, etc., I should be able to get a quick swag on construction costs also.

So, I now know how much it’s worth and what it’ll cost to build, so I can add my land value and see if the total basis makes sense in light of my estimated income and value. This whole first screen process? Maybe an hour.

If it fails this, no chance it suddenly passes with more info, so we move on. If it passes? It still probably fails with more info, but it’s worth taking the next step.

Each iterative step as we move on will also include a version of “do the math and see if the numbers work” but I’ll skip noting it from here on out. With each step you’ll have more info, so your numbers can become more refined.

Facilities Programming in the absence of economic outcomes means you’re gonna have a bad day.

So, time for iteration two. We’re not going to get more granular on what’s buildable yet, but we are going to get more granular on our programming assumption.

Here, we’ll either want to work with our architect or a brand’s architect to get a more accurate paper gross square footage estimate for our building. I say paper because it will unquestionably change when we start laying it out, but we’ll get to that later.

Here we’ll want to look at the details for our competitors and what we think about our target guest and want to craft a target for our guest facing spaces.

We’ll want to come up with key count and suite mix, how much meeting space and what size are the spaces, how much / what type (at a high level) / how big are the F+B spaces, how big is the gym, what other features are we offering, etc. That may look like this: 300 keys (350…

…net SF bay size; 10% suites – 20 one bedroom, 7 two bedroom, 2 hospitality, 1 presidential); 25,000 SF meeting space (8,000 square foot ballroom; 5,000 square foot junior; two 2,000 square foot rooms; one 1,000 square foot room; two 500 square foot boardrooms); 1,000 square…

…foot fitness center; three meal restaurant and bar with approximately 100 seats; coffee shop of about 1,000 square feet; outdoor pool with deck holding ~150 chaise lounges; pool bar with seating for 20; basement speakeasy with seating for 50. P.S. – Don’t build this hotel.

It wouldn’t go well.

Now we take that to our architect (or our brand’s architecture team) and ask them to do some work. We’ll want to translate this to gross square footages and add the spaces that are missing.

Having an experienced architect on your team do this is great, because they can draw upon other projects to see that a 100 seat three-meal will want ~25 square feet per seat front-of-house and will want ~2,000 square feet of kitchen and back-of-house, or know that a 350 net SF…

…room typically grosses up to 390 SF, or that you typically have 10 square feet of circulation per key, etc. Your architect may, and brand architects will, have an excel model for all this that is being pretty constantly refined.

The output of this model can then be compared to what you can build, and you can refine your programming to match. You’ll do another round or rounds of economics math at this stage, depending on the iterations you make.

We’ve now hit the point where we have a good program that we think is economically feasible. So, we’ll want to get a bit more granular on what we can actually build, so first it’s time to talk to that person with zoning knowledge and ask them to do some research.

You could read the local zoning code yourself, if you’re so inclined – I tend to ask the expert, but also read the code, mostly so I can learn something and ask half intelligent questions. This is time consuming, which is why we did the first screen.

We’re not looking for something exact in this phase but getting a rough building mass from parking / drainage / setbacks / etc., is going to be helpful now. Once we have this, we should do another round of iteration on our programming model to make sure things fit.

If they do, and our economics math hasn’t changed materially, time for our architect to shine.

Time for a test fit! No building is ever as efficient as an excel model, and some are way, way worse.

Your architect will want to start with a standard guestroom floor, and also hit the major public space floors, to see if we can get everything we need or if we’re going to have problems.

Here we’ll find out if we can get close to the number of keys we want, if our meeting spaces can lay out column free, if our kitchen can be anywhere near the areas it needs to service, how our arrival and loading dock interplay, etc. In short, this will be the first time we…

…can look at something that functionally resembles a hotel. Exciting!!

You’ll want to iterate a bunch here with both your architect and your ops expert, ensuring we’re not just moving boxes representing spaces around a page to make them fit (e.g., Tetris), but also doing so in a way that the interplay between the boxes makes some level of…

…operational sense. As we work through these iterations, we’ll want to also continually be refining our economic model, both on the operating performance and on the development cost estimate side.

At some point, after all that effort, we’ve programmed ourselves a hotel!

The final programming is usually settled sometime a few months after you’ve stabilized operations, so, you know, keep at it.