Skip to content

Site Selection and Project Programming in Real Estate Development

Social Media Artwork Template2 (66)

As part of learning the real estate development business, the process of site selection and initial project programming (product selection, unit types, unit mix, unit yields, etc) will likely be one of the first things you come across as you build your career. It's also part of the daily activity of any real estate development or home building company, and the land acquisition team and manager, are on the front lines in this battle.

Part of the developers' "value add" is to select sites that are feasible for a given development "product" i.e. single family, duplex, triplex, multi-family, etc. We are always thinking about how many units can fit on a given size piece of land (yield analysis).

Now, you will say to yourself: doesn't that depend on what the architect can fit when he draws plans? And the answer would be yes, but.... we as developers must already know what the product is to be. Usually you look at sites based on what you know, are comfortable with, what you know works, or a new trend that you know from research is viable. Maybe it's SFR, duplex, or whatever makes sense. And, you'll always be evolving your thinking and action in the offers you present to the markets in terms of "what works now".

An example, in SoCal and California generally, podium projects (concrete garages below grade/on grade, with 2-5 floors of wood framed apartments above) have been a "new" trend since around 2000 ("new" because it's not really new, but a rehash of product has been around LA and many metros since the 60's). Podium has been the thing in the last decade plus. The reason for this, podium is the product that fits most units given parking and other zoning requirements on a given piece of land, to increase economies (i.e. most units on given land), to max economic yields to capital.

In the year 2000, urban infill was just coming into it's own, and we were designing units that ranged from 800 s.f to 1,200 s.f. and mix was composed of 1, 2, and 3 bedrooms units. Today, given demographics related to the Millennial population, we are developing mostly studios (efficiency), 1's, and a small amount of 2's.

So the podium style building is the same, but the unit mix based on demographic and economic trends changes. As the developer, you adjust the unit program (mix of bedrooms types, and sizes) to adjust to the market at the time of the deal (or what you think the trend will be when units are delivered). As professional developers, we always know to anticipate changing trends and use those to our advantage (and sometime to get the heck out of a deal or market if trends are against us)

So for all this explanation, the short story is to select sites that:

- are in neighborhoods where your tenant profile wants to live and pay the rents you want (really: need) to charge

- can accommodate unit types that renters want

- fit the maximum amount of units that are needed given economics and zoning of your deal

- can be built cost effectively

- capital wants to finance

- can be bought for land prices that work for your deal economics

Regarding the process of finding sites. I won't go into great detail, but do build your networks of deal finders: brokers, planning departments, other developers, bird dogs, etc. Also, I have found many great sites by driving around neighborhoods that I have identified as a place where we want to be. Obviously, for those who don't live in the markets you want to develop in, this presents a logistical challenge to be there to drive neighborhoods. Basically, I am looking for sites everywhere all the time, and I challenge all those on my team to do the same. It is amazing what happens when a team's eyes and ears are peeled looking for sites. Also, don't always be looking for vacant sites, look for empty buildings, parking lots, junk houses for tear down. In any given urban market, most of the best sites won't be obvious.

I also say that if a site is vacant in the middle of a built out urban area, there's some story there: difficult owners, environmental issues, zoning and planning issues, etc.

Once we find a site that we think fits the "model" - in this example we'll use a 3-story town home product, and we do some basic math to see what fits on the site. From our others projects we know we can fit one unit on 1,700 s.f. of land (with driveway), we then find a 7,500 s.f. site as an example. Divide 7,500 by 1,700, and we get 4.41 units, we then round down (always round down) and we generate 4 units on this 7,500 s.f. site.

You get the 1,700 by determining what product type you want, fit the necessary parking, bedrooms, set backs, driveways, and common area, then measure the footprint of the unit. This may require some design iterations on your part when you first design a new product offer for the market. We prototype very rapidly these days as our team has been doing this for a while. Hiring a good architect will help in this process greatly.

Another way to think of this calculation is in terms of dwelling units/acre (du/a). So 1,700 s.f. per unit of land area, divided by 43,560 s.f. (1 acre) gives us 25 du/a or dwelling units per acre. If you had a two acre site, that's 50 units (2x25).

These methods can be used interchangeably.

Of course, your zoning density will always dictate your du/a, sometimes effected by Floor Area Ratios (FAR's), and always by setbacks and parking requirements.

Subscribe to our regular newsletter and get exclusive access to our next investment opportunity.

Recent Posts

Step by Step Real Estate

Step by Step Real Estate “What we cover in this podcast episode with Scott Choppin:   Updates on what has happened…


The Diesel Crunch is Finally Causing Demand Destruction

The Diesel Crunch is Finally Causing Demand Destruction “Now signs have emerged that weaker demand in the past weeks may…


Housing Market Obliterated: Pending Home Sales Post Record Drop

Housing Market Obliterated: Pending Home Sales Post Record Drop “Well, slowly but surely this epic collapse in sentiment is spreading…