Sesame Part 1: Field 3



Sesame came out of the Field 3 class at Harvard Business School (HBS). The field curriculum was developed to complement the core case-method with opportunities for hands-on experience. For Field 3, groups of five MBA students have a semester to launch a micro-business. Students are allowed to form their own teams and work on whatever business they choose within certain boundaries.


The inspiration behind our selected project came from my own experience. For years I had used blackout blinds to make my room completely dark. I also convinced my employers to let me come in whenever I woke up (as long as there wasn’t an early meeting).

Then I got married and started grad school. My wife hated waking up for work in the dark- it left her feeling groggy for at least an hour. Since my classes wouldn’t wait to start until I was fully rested, I was in the same boat. But without the blackout blinds, I struggled to fall asleep and sleep soundly with the street light shining through the windows.

My initial idea was an automated blackout blind that would keep out light at night, but gradually open in the morning to ease you awake.

Customer Research

During the class, our team conducted an online survey of customers, in-person focus groups, and solicited feedback within retailers.

A total of 283 non-HBS, non-Harvard people responded to our online survey; 194 opted-in for future follow up and 111 people (~40%) signed up to be a beta testers. 

24% of respondents said that the quality of their sleep negatively impacts their day. Renters were less satisfied with the aesthetics of their window treatments (avg score 3.15/5.0) than owners (3.79/5.0). Renters were also less satisfied with the light blockage of their current window treatment (3.02/5.0) compared to owners (3.34/5.0).

Of people who reported having problems sleeping, 41% had tried blackout blinds, 25% had tried sleep cycle apps, 45% had tried melatonin pills, 36% use earplugs, and 32% use prescription or OTC sleep aids/drugs/alcohol.

We identified four key pain points in creating a good sleep environment in an apartment:

  • cost barriers: “I want this product, but I can’t afford to pay $500-$700 per window,” “sometimes I come just to look at things and dream about my future home, which I can’t afford right now.”
  • installation barriers: “I ended up having to duck tape [the curtain] to my wall. I didn’t know what else to do,” “I’m a mechanical engineer and it took me an hour and a half to put in a single blind!”
  • aesthetic trade offs: “I wanted to buy the pretty curtains, but I knew they weren’t going to block out the sun so I had to find black out material to line the curtains with and sew it in…a real labor of love.”
  • effectiveness: “The sun may rise at 5:30 am, but I want my sun to rise at 7:30”

Within the category of young, urban renters, we identified three user groups that were especially excited about our idea: problem sleepers, home automation enthusiasts, and aspirational optimizers.


We wanted to solve the following three problems for urban renters who don’t have a good sleep environment:

  1. struggling to fall asleep from nighttime light pollution outside their windows;
  2. being awakened earlier than intended by sunlight streaming into their bedrooms; and/or
  3. implementing a blackout solution only to struggle to fight off the grogginess that comes from waking up from an alarm in the pitch dark.  


In order to tackle these problems and address the most important pain points for our customers, we identified eight key features. The product will:

  1. include a blind that covers an entire window.
  2. include magnets and brackets to prevent excess leakage
  3. include hardware for mounting the blind outside of the window
  4. include a motor capable of opening and closing the blind
  5. be powered by a rechargeable battery capable of lasting 1 year between charges
  6. include a wifi chip capable of communicating with a mobile phone
  7. include a smart phone application capable of programming the opening time of the blind
  8. be offered in 3 sizes to fit most customers’ windows


The team worked together during Field 3 to build the first two MVPs for presentations and interacting with customers.  These prototypes had the ability to automatically raise and lower, but did not yet have any programmability.   

The first prototype was a horizontal-opening, motorized curtain system that was built with a DC motor, belt, gears, wood, a table clamp and a curtain rod. The system was battery operated and could be opened and closed with a physical switch. This prototype allowed us to test out ideas around flexibility and ease of installation. 


The second prototype was a vertical-opening blind built with RollerTrol parts. It was a roller shade that went up and down with an RF remote. The second prototype was a great demonstration tool that communicated the idea to interviewees and helped elicit more informed feedback.

The second prototype also allowed us to test using a magnet in the base of the blind in order to seal the blind to the window. This feature reduced the amount of light that leaked around the blind, which created a better sleep environment.

Together, these two prototypes helped us to make the tradeoffs of a vertical-opening vs. horizontal-opening prototype tangible to the team and customers. The vertical opening blind was a simpler mechanism and made less of an aesthetic statement. The horizontal-opening curtain was more complicated but allowed for a flexible design that could fit different window sizes. 

Business Model


At the time of this project, off-the-shelf offerings were designed with convenience, not sleep, in mind and cost $500-$700 per window plus custom installation fees. These options are produced by Lutron, Bali and Hunter Douglas.

At the low end of the market, the alternative is to hack together and program a solution yourself using a kit sold by electronic component companies like RollerTrol. There were also a few substitute products including face masks and wake up lamps, which offer only a partial solution.

Bill of Materials

We developed an estimated bill of materials to help us understand what our approximate unit economics would be. Many of the indirect costs were derived from Ben Einstein’s posts on hardware development which can be found here and here.


Unit Economics

We estimated our unit economics for selling and fulfilling through Amazon. The COGS was chosen within the range of values given by our BOM. However, it was on the lower end because affordability would be a design constraint going forward.

Based on this post by Dragon Innovation, the 42% gross margin was just below the target gross margin of 50% that leads to a business that can fund it’s own growth. 


5 Year Cash Flow Projections

There are a lot of assumptions built into this 5 year cash flow projection. If you are curious where the numbers came from, feel free to reach out and I’d be happy to explain our process.

Assuming these numbers are mostly correct, we would need to get $500k in funding to get to cash flow positive. However, we would likely target $1M in funding because it can be difficult to get additional funding in your time of need if your projections were too low.




Next Steps

A teammate and I got $6,000 each through the Rock Center at HBS to pursue this idea over the 2015 summer. Our goals for the summer were to:

  1. Design and build a “works-like” prototype of the full Sesame system
  2. Conduct in-depth customer interviews to understand whether there is a market for our product vision
  3. Test and refine our assumptions around WTP
  4. Develop technical specifications to transition from the prototype stage to the DFM (design for manufacture) stage


You can download a version of our final presentation with identities removed below.

Sesame Presentation