Three Massive Market Opportunities in the 2020s: Clean Energy, Global Labor, Senior Living.

I intended for this to be a short article about three different business growth areas that I am interested in and have spent some of the past year thinking about; Clean Energy, Global Labor Market, Senior Care. I had more enthusiasm bottled up than I realized so this article got a little bit longer than I had anticipated. I assume there are a lot of other people who, like me, are curious about a lot of topics but don’t have the time in their day to dedicate to each one of them. I hope this article can offer a gateway into a new intellectual curiosity or catalyze a desire to learn more. At a minimum, ideally when you are talking to anyone about these topics, you’ll have a baseline understanding.

Clean Energy

I have secretly harbored a great deal of shame since my high school physics days for not having a strong grasp on how electricity and energy systems work. Physics was not intuitive for me but recently I bought an electric vehicle and it has been a powerful inspiration for me to start reading books/articles and watching videos about energy, electricity, motors, power plants, etc.

A few things are clear:

1) I am never going to become an electrical engineer.

2) Creating a simple mental model for understanding energy systems is the best way to start to understand this domain.

3) You do not need to be a physicist to put together a logical high level picture of what is going on in the energy world and to form opinions.

I found the book “How to Avoid a Climate Disaster” by Bill Gates to be immensely helpful in getting foundational knowledge and then going down Google rabbit holes from there. I’d also suggest checking out any book written by Vacliv Smil. Here is the gist from ‘How to Avoid a Climate Disaster’:

The book breaks the future of clean energy down into 5 buckets:

1) How much of the 51 billion tons of emissions (annual emissions today) does any individual solution impact?

2) What is your plan for cement? (The world uses a boatload of cement and it accounts for 10% of all emissions).

3) How much power can any individual solution produce if scaled? How much would that cost to implement?

4) How much space does it need? (ie wind needs more than nuclear for the equivalent amount of power).

5) What is the green premium? (How much more does it cost than the non-green alternative).

If we put these questions in our pocket for a second, let’s think about the different types of energy and how energy works at a very very high level (like 1,000,000 ft).

Energy is described (very unsatisfyingly) as “the ability to do work”. Ok what the fuck? Basically that just means you need it to do literally anything. I don’t know all the laws of Thermodynamics offhand but you can Google them, they will give you scoop on the properties of energy and then if you drop “The Second Law of Thermodynamics” at a cocktail party people will probably think you are super smart and know a lot about energy. But maybe after COVID, cocktail parties won’t exist anymore and then your newfound knowledge won’t be as impressive on Zoom. TBD.

Back to the point, there are many different types of energy (nuclear, chemical, solar, heat, electrical, motion, light, gravitational) and energy can convert from one form (ie chemical) to another (ie heat). I don’t think you need to understand the specific details of each conversion process or else you will drive yourself crazy. Just know that energy can convert from one form to another.

Power is the metric we use to measure how much energy can be used or converted in a given time period and this is one of the primary ways we can compare energy sources to one another.

This is what (I think) you need to know in order to have a good mental model.

Power, power, power. Power is the concept that literally makes the world work (no pun intended).

Here is a very specific framework you can use to plug in details and assess if an energy concept sounds promising. (NOTE: This is incomplete as it ignores the other half of the puzzle which is removing greenhouse gases from the atmosphere i.e. direct air capture…but it’s still a useful thinking tool).

The power source starts as <type of energy> and each <unit of volume> produces <unit of power> which costs <dollars to produce> and has a <green premium of X> and the method of conversion is <insert method of conversion> which is <percent efficiency> and the byproduct is <what emission & heat>.

Once you have a good grasp of these high level ideas then you can start to think of energy systems as sequences of energy conversion methods and start to imagine creative ways to get a type of energy captured and converted to power in a clean way. Kind of like energy legos.

Here is an example:

If I want to charge an electric vehicle by the side of the highway, I need electricity to put into its batteries. So how can I get that electricity? There are a lot of different ways.

1) I can be connected to an electricity grid powered by a local coal plant.

2) The charging station can use solar energy to generate electricity.

3) I could have 100,000 smurfs ride stationary bicycles to generate electricity.

4) If I am by a river, I could generate electricity from a hydroelectric dam.

5) I could collect shit from thousands of cows in the area and convert it to chemical energy and heat, then power a motor that generates electricity which then charges the car.

6) If I am by a hill with a reservoir at the top, I could pump water up the hill into the reservoir and then let it fall back down to produce hydroelectric energy to create electricity to charge the car. This is similar to #4 but illustrates how you can effectively ‘arbitrage’ energy (ie the ‘energy cost’ of pumping the water up to the reservoir is less than the ‘energy output’ of the resulting hydroelectric system).

As you can see there are a lot of different chains of energy conversion that lead to the desired outcome of power that is needed. So at the most fundamental level what we are evaluating is what technologies, energy sources and chains of connecting them together results in the power (usually electricity) we need and how much more that is going to cost than the way we do it today. If a technology, energy source or method of connecting them together appears to answer any of the 5 “Bill Gates questions” in a favorable way, then you’d be more likely to consider it a promising idea.

The city of Shenzhen, China is a very interesting and cool example. They managed to convert their 16,000 city buses to electric vehicles. So if we use this as an example, the questions that you might consider in this case:

1) How much is the conversion cost from internal combustion to electric vehicle?

2) Who did the conversion? How did they do it? What electric components are they using? What batteries? How long did it take to convert each vehicle?

3) What is the emissions reduction?

4) What government policy is needed to make this happen?

5) What is the green premium (total cost of ownership)?

6) What are the power sources of charging the buses? Are they green or fossil fuel based? If green, what is the green premium?

7) How much cheaper is the servicing of the vehicle?

8) What is the battery replacement cost?

#6 and #7 are sub-questions related to the ‘green premium’ but worth laying out explicitly for illustrative purposes.

One last example that I found particularly creative I saw firsthand when I was at Kakuma Refugee Camp in northern Kenya in the summer of 2019. I met these Dutch NGO workers who were introducing a (for profit) very inexpensive solar panel system that generates enough light for one household to have at night and is cost effective enough for a refugee to afford (on a payment plan…like $80 total). Why do I find this interesting? It shows that not every green energy solution needs to be a massive infrastructure project or energy system. This is a great example of how a creative energy solution can unlock a huge amount of value in terms of human capital which I will discuss in the next section (more light == more time to learn & read when it’s dark == more educated global population == new labor opportunities with higher wages == more economic development in these places).

My eyes have become wide open to the massive energy transformation that is taking place in the world since purchasing an electric vehicle and its an area I am trying to actively learn more about. I hope this brief summary will help to demystify it or make it more approachable for anyone else who wasn’t that stoked during high school physics class.

Global Labor Market

Let me just start by saying I am LONG the global market for skilled labor. If I had a personal north star, this is it. Developing human capital is the oil of the 21st century. Not data. Data is cool too, but figuring out how to educate a billion more people in a highly pragmatic way so they are able to do the work needed in the “information economy” is the greatest value unlock of the next 10 years.

I have spent a good amount of time in Ukraine over the past 6 years and have seen up close what a growing IT industry does to an economy and all of the peripheral new opportunities that emerge once people in an economic system start to have more money. Once a labor market reaches a critical “activation energy” it produces a few really important byproducts- role models, mentors, new career tracks, community groups (ie Kyiv product manager meetup), knowledge sharing, more demand for talent and specialized training programs. This makes continued growth organic and sustainable. Reaching this ‘activation energy’ is the key precursor to private investment.

There are a lot of factors that go into ‘activating’ a knowledge economy. Some of which I took for granted until I went to Rwanda and Kenya in 2019 so let’s start there. That trip taught me a lot about my blindspots. I attempted to start a coding bootcamp in Rwanda (unsuccessfully, but maybe I will try again in the future) and a few prerequisites for a successful information economy I had never considered became very apparent to me; cost effective internet, cost effective electricity and reliable devices (…I know). Without those three elements, you are basically fu$ked when it comes to participating in the global ‘information economy’. I hope that the innovation in clean energy can help solve the electricity part of this puzzle (ie that solar panel example I mentioned).

Between my experiences in Ukraine and Rwanda, I have thought a lot about the “lifecycle” of how any particular place goes from where it is today to reaching the ‘activation energy’ needed for sustainable private sector growth and participation in the ‘information economy’. In my view, the first part of the lifecycle requires government funding, smart policy and/or non-profit funding to get a large enough percentage of the population to the starting line for investors interested in developing a skilled workforce.

Here are some macro thoughts:

Part 1: Good primary education for poor kids is a prerequisite. This is not something that private investors are going to pay for (unless you want people betting on kindergartners and taking 20% of their lifetime earnings which is a dystopian hunger games future we probably don’t want). Without this step, the gap from where an apprentice or student starts their training to the time and cost it takes them to become a productive, specialized employee is too high for private investors to fund.

This is part of why Ukraine, Belarus, India, Russia, Brazil, Argentina, etc are booming ‘information economy’ labor markets but Rwanda, Kenya, Nigeria, etc much harder. The first set of countries I mentioned also already has all of the positive ‘byproducts’ for sustainable growth but none of that would have been possible without a strong foundation of good primary education for millions of kids.

Andela raised over $100M but that wasn’t enough. It requires billions of dollars in public investment to get a large enough portion of society primed for specialized training that can yield higher wage jobs. A cynical (but realistic) way of looking at this is that if a company wants to open an office for developers, they know in Ukraine they can find 100,000 software developers who will be competent or trainable, but in Rwanda today they might be able to find 200. So the shortage of economically viable, trainable labor supply is the primary problem to be solved. Until economies of scale for professional training programs can be achieved, an “information economy” is not going to “activate”. The profit incentives aren’t strong enough.

That being said, whoever can unlock the intellectual capital of Africa, Central America, Indonesia, etc has a huge market opportunity. There are a lot of smart people in all of these places who just need the right conditions and they will thrive.

You can dramatically reduce the amount of public or non profit funding required to make new populations of people ‘investable’ as future knowledge workers by getting computers and the internet into their hands at a young age. Online learning isn’t restricted to college courses and above. There is plenty of great content for younger kids. If you solve the challenges and cost barriers to getting reliable internet as I observed in Rwanda then large segments of the population will definitely be industrious enough to self-learn from there. Computers & the internet are the raw material for knowledge. Conventional wisdom in the US over values formal education and undervalues motivation for a better life plus the internet.

Part 2: Once you have a population with enough people who have prerequisite knowledge and infrastructure, then the market dynamics are aligned for private companies and education innovators to train hundreds of thousands of specialized workers. Platforms like Cousera, Udacity, Udemy, EdX, YouTube, etc are massively empowering and when combined with mentorship and real world opportunity, this will lead to an explosion of new global talent. These platforms have so much high quality learning content that if I am company that wants to train 500 people, I don’t need to invest in creating my own university or hiring a costly team of professors. Instead I can use these existing ‘pseudo-public common’’ education platforms and then focus on rich apprenticeship experiences. Scaling apprenticeship programs, real world work projects and verifying skill quality is a big business opportunity.

My mental model for the future global labor market is broken down into four areas:

1) Affordable and accessible infrastructure. Is satellite internet going to be the dawn of an education renaissance in Africa? Will inexpensive monitors & keyboards that can connect to AndroidOS so people who have an Android can do work that requires a PC without bearing a cost for a laptop they can’t afford? Will $100 computers get good enough for serious work? Affordable alternatives to commonly used software?

2) Closing the Culture Gap. Here is an example.

“How is the project going so far?”…”It’s ok”.

A person in Ukraine heard “Oh ok cool, everything is ok. All good.”

A person in the US heard “Oh man so we are doing a really bad job…everything is NOT ok.”

This is a much bigger hurdle than the accrual of technical knowledge. There is a very interesting book called The Culture Map that speaks to the communication differences between cultures around the world. I’d suggest reading it if you work with a global team. So much of the innovation, creativity and problem solving that happens in today’s workplace takes places between people, not inside of one person’s brain. Even if a team is all speaking to one another in English, they are not necessarily speaking the same language.

Part of this is going to naturally solve itself through social media because millions of young people are consuming the same content so culture itself is increasingly becoming borderless.

The other part of this problem will be solved through soft skills training courses and an investment in cultural and contextual education within organizations. 99% of problems I see within a cross cultural team in the software industry are rooted in communication failures, not technical shortcomings.

The communication gap is also why if we look at the pattern of jobs that have become global we see that sales, marketing and customer success, copywriting are the hardest to move to another culture but as the world continues to assimilate this is going to be next. To put it bluntly, EQ is harder to outsource than IQ.

As an American myself, I want to point out that I am not suggesting cost-effective alternatives to US workers are a better solution, but simply that the global competition that designers and developers face is coming for soft skill specializations too.

Programs like SV Academy (in the US) are interesting to me because they are amongst the first to be systematically approaching the issue of specialized soft skills training and I expect the same things will happen in other countries. This is an area of particular interest for me.

3) Inefficient labor markets in law & other professional services. I had to pay $300/hr to talk with one of the leading neurosurgeons for Deep Brain Stimulation surgeries in Canada. I can talk to him for as many hours as I want at $300/hr. If I want to get one hour of time from an associate at a corporate law firm it is going to cost me $400–600/hr. Why is the expertise of an experienced brain surgeon less expensive than a legal associate with little experience and knowledge that is not particularly complex? Because of regulatory barriers and higher education costs. Why can’t a very smart young person in India do the same work as an associate at DLA Piper? It’s not like you can’t download the US constitution or US legal textbooks in another country.

The point of this story is to illustrate the skill to value paradox that gets created by conventional wisdom, fear of the unknown and regulatory barriers which result in unnatural labor markets. It just doesn’t make sense to me that an inexperienced legal associate costs more than a specialized brain surgeon. This is a big inefficiency and my bet is that a lot more professions with regulatory barriers are going to see an onslaught of global competition as savvy entrepreneurs and education investors find regulatory workarounds and the right messaging to alleviate fears.

4) US higher education is fucked and for profit apprenticeship & education programs are going to continue growing in popularity. The formats and subject matters of these programs are going to expand well beyond the technology sphere. It just doesn’t make sense for many students to incur the debt required for a four year college degree. The economics don’t work. Somewhere along the line, the US higher education model has become a dick swinging contest of 1) endowment size (no pun intended), 2) US News and World Report ranking and 3) having as many administrators as possible while keeping the number of educators static. This is stupid and there are increasingly high quality alternative programs that are lower in cost, faster to complete and result in higher economic yield. It just occurred to me that I should be more specific about what I mean by “for profit apprenticeship & education programs”. I don’t mean criminal enterprises like University of Phoenix or other Private Equity owned ‘accredited universities” that they converted into predatory lending schemes…I mean companies like Udacity, Udemy, LambdaSchool, HackReactor, SVAcademy.

US higher education obviously has a place in the future, and some schools are showing initiative in the right direction — Georgia Tech, Arizona State, UPenn, University of Illinois & University of Michigan, for example, are leveraging platforms like Coursera to offer lower cost online degrees and certificate programs. I periodically go on Coursera and look at which Universities are offering degrees just to keep my finger on the pulse of the state of the migration to MOOC model or if I come across any interesting innovations or programs.

Lastly, I think undergraduate education and advanced research can be decoupled. Before the internet, streaming content, etc, if young people wanted to collaborate with or learn from an expert they would physically need to be in the same place. So if I wanted to learn chemistry from an expert, I would need to go to the same place he was doing research and hope he was willing (or was forced) to teach my course. But today, this isn’t necessary. Because of the fact that undergraduate training is no longer dependent on physical co-location with experts, Universities have lost their monopoly on higher education. We are in a transition period where University brand values are higher than their practical value to most students but it seems like this is starting to correct itself.

This section got longer than I had hoped for. It’s a domain I am passionate about. I believe deeply that there are millions and millions of talented people around the world and as innovators discover more and more efficient ways to make them productive workers (ie do this faster and more cost effectively), this will change the makeup of the global labor force at an increasing rate. If someone in Somalia can learn how to be a DevOps engineer on AWS, they will have plenty of time to read Catcher and the Rye & Socrates and get a liberal arts education if they so choose on Coursera for $300 rather than $200k.

Senior Care & AI Diagnostics in Healthcare

There were a few things I never thought I’d say as a 5 year old. One of them was that I find the senior living industry to be extremely interesting and dynamic. But here we are. 36, thinning hair, getting sunburns on the top of my head and burning the midnight oil thinking about senior living. Life is so unpredictable.

The silver tsunami is coming. There are already over 50 million people in the US over the age of 65 and that number is skyrocketing because of the baby boomer generation. So what does that mean?

1) The tightasses from the 50s had more kids than their ‘free love’ baby boomers kids. Seems like coming back from a war gets your more in the child rearing mood than smoking some low end marijuana.

2) The number of older Americans in need of acute health care services is going to outstrip the supply of health care workers. This is a problem that needs solving.

3) The number of Americans suffering from cognitive diseases like Alzheimers, Dementia & Parkinson’s is going to continue to rise and is very sad (10M+ people are expected to suffer from Alzheimers and Dementia in the US in the next 5–10 years). How are we going to handle all these people in need of memory care services?

4) There is an ongoing loneliness pandemic amongst seniors that started well before COVID and is going to continue well after it’s over. When my grandmother was 90, she told me that when she was in a room with younger people she felt invisible. That’s very disheartening but I don’t think it’s uncommon. Maintaining a high standard of living for seniors is just as much about social interaction as it is about medical care.

So where are the opportunities in senior living?

1) People want to live at home. How do we make that viable for a larger segment of seniors including those who require some level of ongoing care? Will the future include smart home companies like that do customized implementations specifically focused on seniors with care needs, monitoring of vitals, etc.

2) Train more care workers and enable them to get paid more. This is connected to #1 in that if the ratio of care workers to seniors can be increased, then one could imagine solutions combining caretakers and smart technology that enable a greater proportion of people to live at home.

Becoming a CNA (Certified Nursing Assistant) requires getting a certification that you pay for and then earning the honor of potentially cleaning feces, being treated poorly, receiving verbal abuse by people with dementia and more positively growing close to people who then die. And for all of these things, today you get paid less than a Starbucks barista. This is unreasonable and it’s why there is such high turnover in this field and a labor shortage in the senior living industry.

I could write an entire article about this one topic, but suffice it to say the model of ownership for large senior living operators and debt financing used in the industry makes it challenging to solve this in senior living facilities. An older person living at home does not have the same debt financing constraints and IRR targets as a senior living facility.

3) Caretaker + tech hybrid at home care. This is very closely related to #2. If the average monthly cost of an assisted living facility is $5000/mo (this assumes private pay, non-medicaid) which includes an hour or two of daily care for “Activities of Daily Living”, an apartment/room, meals and activities, then we have high level criteria for comparison. We can ask ourselves if there is an alternative model in the same cost range that lets someone stay at home, pays a CNA $20/hr (or a livable wage depending on where they live).

If we do the raw math problem, $5000/$20, we get 250 hours of available care if all the money were being spent on a CNA with someone staying in their home. Just for the sake of comparison, a full time professional job is 160 hours per month and there are ~720 total hours in a month if we wanted to understand what a 24/7 picture looks like. Let’s say a senior opted for 160 hours of home care. That leaves $1800/mo for food, entertainment, etc. Not bad.

So a solution that keeps cost equivalent for a senior who wants to stay at home will have some trade offs. They’d lose access to 24/7 care, but gain the comfort of being in their own home and would still be able to have a full time caretaker (160 hours per month). This begs the question of what technologies can mitigate health risks and send real time alerts when a caretaker is not there.

Can a model like this with innovative smart home & health monitoring technology & access to a direct and trustworthy caretaker marketplace offer a new, equivalent cost way to keep people in their homes, keep them happier and help raise the economic opportunity for caretakers?

4) AI, telehealth & preventive care innovations. It’s no secret that healthcare costs in the US are very high. I’ve learned a lot more about this over the past two years and feel a mix of disgust for the system and empathy for practitioners who want to do a good job. There are a few big problems.

a. Billing codes create strange incentives for provider visits and care plans. I don’t know a solution to this. I just wanted to point out how zany I’ve found our billing code system to be the more I’ve learned about it. This one requires a policy solution and based on the current Washington gridlock, it doesn’t seem like its happening anytime soon.

b. Specialized doctors (neurologists, cardiologists, etc) are very difficult to access and often do not have enough time to dedicate to each patient. I have had the opportunity to chat with a lot of neurologists over the past year because of a project I am working on. I am much more in awe of the human brain now after having learned more about it. But that can be the topic of another post. One of the recurring themes is how many patients each neurologist has to see and how little time they get to spend with each patient. Maybe 20 minutes every six months? Thats not ideal for anyone.

Innovations in telehealth are starting to solve this. One very cool company that I have worked with a bit is Sevaro Health. They make it easy for anyone to get connected for a virtual visit with a neurologist in a few minutes. Providing this type of online connectivity to a network of specialists can be a very efficient way to solve the scarcity issue and help more people with very specific issues get treated faster, earlier and with more attention.

c. AI for diagnostics and real time alerts. This is closely related to “b”. Another negative effect stemming from scarce access to specialists is that major problems often are not detected until a catastrophic event or well after preventative care measures could have been taken to resolve or improve the underlying issue. This produces two bad outcomes. Bad things happen to more people and because bad things happen to more people, care costs skyrocket.

Here is an example. The starting point for the diagnosis of Alzheimer’s is to take a short test called MoCA or MMSE that takes 15 minutes and is administered by a doctor if they suspect a person might have a cognitive disease. These tests are so basic that someone would only fail years after they’ into their cognitive decline. Taken together, if an older person sees a neurologist once a year for 20 minutes and the diagnostic tests used are so basic, we are diagnosing cognitive diseases years after their onset so any potential preventative measures are moot. Could we use data collected from their every day life or their performance of cognitive tasks to provide alerts when there appears to be a pattern moving in a concerning direction? Or how else can we automate the diagnostic work of specialists with data collection?

A company based in Israeli called K Health is taking an intriguing approach to diagnostics using AI (but not focused on senior living). Their AI technology uses data from your peer group to suggest the potential issues you are facing so you can get feedback without always needing to incur the costs and delays of seeing a doctor.

d. Clinical trial tools for geriatric conditions. There are platforms which connect patients with trials for specific conditions or diseases, but how about building an opt-in database of all people over 65 who are interested in access to clinical trials for any disease that disproportionately impacts older people. The cost of facilitating clinical trials is very high and part of that is finding qualified participants, especially older participants, since they are not accessible through as many communication channels.

All things startup and technology. Founder of— A product development studio for high growth startups and leading brands.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store