Announcing Delve: Discovering radically better urban designs

The traditional urban development process struggles to design for complex and competing priorities. Delve uses machine learning to reveal optimal design options — and deliver better neighborhoods.

By Okalo Ikhena


Late last year, Sidewalk Labs released a prototype generative design tool with the goal of helping urban development teams — including developers, architects, planners, and urban designers — identify far better neighborhood design options than time and cost constraints typically allow. Today, we’re excited to introduce Delve, a product that marks the next step in realizing this vision.

Over the life of any development project, teams invest countless hours in getting the little things right. Yet the traditional urban development process struggles to design for a complex and competing set of priorities and constraints. The decisions made early in the process, especially in the master-planning stage, can have unforeseen impacts later on — and can make or break a project’s viability.

Delve uses machine learning to help development teams discover the best neighborhood design for their project, based on the priority outcomes that matter most. By revealing the optimal design option, Delve helps development teams exceed their project economic goals while improving quality-of-life outcomes for residents and businesses.

Development teams around the world are actively using Delve now, and we’re looking to engage others, so we’d like to share a little more about how Delve works and how it was recently applied to a large-scale development project near London. If you’re a developer interested in working with us, reach out through our website.

How it works

City neighborhoods have unique personalities, but they also share a lot of the same core components. Our team has built a model of these core components that includes buildings, open spaces, amenities, streets, and energy infrastructure. By applying machine learning to that model, Delve explores millions of design possibilities for a given project, measuring the impact of these designs to help development teams arrive at the one that’s right for them.

This discovery process begins with a development team using Delve to identify a project’s planning inputs (such as gross floor area for commercial or residential use), site constraints (such as height limits), and priority outcomes (such as cost or daylight access). Delve’s intuitive interface makes this initial step easy, with support available from a computational designer who is a trained architect.

From there, Delve empowers a development team to get from endless possibilities to one optimal design. Teams can update their priorities and planning inputs as stakeholder feedback changes, exploring new possibilities in minutes instead of months. Delve then automatically finds the highest-performing options based on a project’s priority outcomes — leading to approvable and financially feasible neighborhood designs that exceed stakeholder expectations.

Delve can be used at various stages throughout the development process: from pre-feasibility phases, as developers scope out potential sites, all the way to having a permitted masterplan, which can serve as a benchmark to improve upon further.


By leveraging machine learning, Delve explores millions of design possibilities for a given project, measuring the impact of these designs to help development teams arrive at the one that’s right for them. Sidewalk Labs

Latest features

Since introducing our initial generative design tool, we’ve refined Delve based on real-world engagements to make it easier for urban development teams to get fast, clear answers to their toughest design challenges. Here are some of Delve’s core features:

Ranked options. While Delve generates all kinds of original neighborhood designs, too many options can be overwhelming. Delve makes it easy to identify the highest-performing design options by ranking them according to how well they perform on the priority metrics defined by the development team, and showing these scores in a clear report.

High-fidelity designs. Delve generates high-fidelity designs across a range of areas. It allocates individual housing units across a project site to help find more residential density. It also evaluates daylight access for each unit — not just whole buildings — to help inform facade treatments and energy load. Other design elements include open space access, walkability, views, and more.


Delve generates high-fidelity designs across a range of areas including daylight access and sunlight hours (shown above), ranking them according to how well they score on the priority metrics defined by the development team. Sidewalk Labs

Integrated financial models. A big part of identifying better design options is creating extremely precise cost estimates. Delve runs an intricate financial model for individual segments of every design option it generates — modeling cost per building, street section, open space, and more. These cost models also integrate with a customer’s own financial spreadsheets to create a seamless experience.

Built-in utility models. Delve features a built-in utility demand tool — designed alongside experts from Sidewalk’s sustainability team — to provide an ongoing estimate of key energy infrastructure measures, such as electricity use, waste and water, and rooftop solar intake. That way, as a project evolves, developers always know whether they’re meeting their own sustainability goals as well as local standards.

Case study: Adding housing units while preserving daylight

So how does this all come together in practice? As an example, we’ll share some details from a recent engagement with the award-winning developer Quintain, which used Delve to inform its work on delivering thousands of multi-family, Build to Rent homes at a 12-acre mixed-use development for the North East Lands site in Wembley Park, near London.

At North East Lands, Quintain faced a common development challenge: providing greater housing density while preserving open space and meeting local daylight standards. The project’s benchmark designs either didn’t provide enough housing units to achieve the financial objectives or added these units by creating taller buildings, which reduced daylight and open space access for residents and workers.

Based on the project’s planning inputs, constraints, and key priorities, Delve identified 24 high-performing design options that exceeded the benchmark designs on priority outcomes, including housing units, daylight access, and open space. One high-performing Delve variant added nearly 200 units (while also increasing the average unit size by 13 square feet), improved daylight access, and expanded open space by 11 percent.


During one recent engagement, Delve identified 24 high-performing design options — including the one shown here — that exceeded the benchmark designs on priority outcomes.
Sidewalk Labs

Overall, this engagement showed Delve’s ability to reveal new value for a major development project in a global city, enabling all project stakeholders to see shared gains. For the development team, Delve found new design options that met all project priorities, including cost and return. For would-be residents and businesses, Delve uncovered designs that preserve or expand key quality-of-life outcomes. And for the London real estate market as a whole, Delve discovered an expanded supply of housing units and office spaces.

(You can read the complete case study here.)

Supercharging the development process

Delve is designed to integrate into the work that developers, architects, planners, and urban designers already do — supercharging the process in a few key ways. It heightens a team’s ability to weigh the many complex (and often competing) elements of new urban development. It unlocks options a team didn’t know about to inform decision-making. And it enables meaningful, rapid adjustments that accelerate project timelines.

We also expect Delve’s impact to reach beyond the needs of a development team into those of the neighborhoods they serve. Residents and businesses want more jobs; more housing, office, and retail options at more affordable price points; higher standards for green buildings and clean air; more accessible outdoor space; or all of the above. Delve alone can’t solve all those concerns, but it can help projects move forward by discovering design options that balance the priorities of development teams and community groups alike.

Any path to greener, more affordable cities involves building more dense, mixed-use developments that expand housing options while preserving quality-of-life outcomes. By revealing designs that meet or exceed the competing goals of many stakeholders, Delve takes a step toward building stronger consensus — and stronger cities.

If you’re part of a development team interested in using Delve, reach out through our website or email us at