Issue #3: Innovation Nation (7/20/20)
Topics: When will we reach peak innovation? According to one scientist - we already did - back in 1873. Also, Open AI launches GPT-3: is the Turing Test dead? Lastly, jobs & pitches for July.
1. Hot Off the Presses: How to Target High-Incomer Households with Facebook Ads
My friend John Jabara teaches entrepreneurship at the University of Maryland Clark School of Engineering and is the author of the book “Mastering Customer Interviews for Entrepreneurs: The Fastest Way to Great Customer Feedback.” He ran a program this summer with Marc Steren called University Startups that taught entrepreneurship to high school students from around the country who attended elite private schools (tuition runs at $50,000/year).
This year, they ran the program mostly through word of mouth and got a huge turnout. He asked me if it was possible to use Facebook or Google Ads to target similar households. I ended up doing a write up about it on Medium. What’s great is that after I wrote it, two of my friends ended up giving additional insights which I incorporated into the article. Thanks Serign Jobe (Manager, Solutions Engineering at Facebook) and Adam Shlomi (Founder, SoFlo SAT Tutoring) for the additional insights!
P.S. Serign also told me that he’s hiring for engineers at Facebook so if any of you have engineering backgrounds or know someone who does, I’d be happy to connect you to him.
Article Preview: [Tutorial] How to Target High Income Parents Using Facebook Ads
The top 10% of American households have an out-sized effect on the U.S. economy. They contribute to nearly one-third of U.S. GDP ($7 trillion a quarter) in terms of spending. (Business Insider) If you are an entrepreneur who wants to sell a product or service to high-income households, the following tutorial will walk you through how to create audiences for them in Facebook Ads.
Anyhow, you can read the Full Article on Medium.
2. Five Things I Learned This Week
The rate of innovation peaked in 1873 and has been declining ever since then. (Study: “A Possible Declining Trend for Worldwide Innovation” by Jonathan Huebner [2005]). You can read refutations of his argument by Ray Kurzweill at New Scientist.
Due to COVID19 related issues, state and local governments will need up to $1 trillion in Federal aid just to stay alive. (Economic Policy Institute)
The Altman Z-Score was developed by NYU Stern Finance Professor Edward Altman to help predict whether a company has a high probability of going bankrupt and looks at five main financial ratios: profitability, leverage, liquidity, solvency, and activity. (Investopedia) (He’s predicting a higher wave of bankruptcies this year compared to last year, including some mega-bankruptcies. His biggest fear is that if we can’t get bankruptcy courts to scale up, many more companies will go into liquidation than necessary, driving up unemployment. This could be catastrophic to say the least. (Bloomberg)
“Humans are depleting more than half of the world’s 37 largest aquifers at unsustainable rates.” (Mother Jones)
How much would it cost for the government to effectively eliminate homelessness in the United States? About $20 billion. (The Huffington Post) (As a comparison, the US government spent $2 Trillion in Afghanistan according to the New York Times 2019)
3. Demos, Demos, Demos: GPT-3 vs the Turing Test
OpenAI is a research lab founded by the likes of Elon Musk and Sam Altman (Y Combinator) who collectively pledged $1 Billion. They recently published Generative Pretrained Transformer 3 (aka GPT-3) which is basically the largest language model ever trained. Why is is it such a big deal? Some are basically arguing that the Turing Test is essentially dead now and we need new ways of evaluating how good AI is.
How does it work? From the Open AI website:
Given any text prompt, the API will return a text completion, attempting to match the pattern you gave it. You can “program” it by showing it just a few examples of what you’d like it to do; its success generally varies depending on how complex the task is. The API also allows you to hone performance on specific tasks by training on a dataset (small or large) of examples you provide, or by learning from human feedback provided by users or labelers.
Here’s a video of GPT-3 in action by Sharif Shameem where he shows how simply describing a website to GPT-3 is enough for it to build the front-end of it.
Here’s a link to another video by Cassidy that shows how easy it is to turn text into a website button.
Wanna mess around with GPT-3? They’ve got a waitlist to get access to their API here.
4. Cool Tools
Framer: For some reason, there’s always another prototyping tool. Framer is the latest one and it seems pretty promising.
Beamer: We build a lot of software products for our clients at Fifth Tribe. One of the most painful points is letting everyone know about the latest release notes. It sounds super-simple but it’s really tedious. We’ve tried emails, power point presentations, giving access to our confluence, etc. The problem with emails and powerpoints is that they get lost and no one really uses confluence if they aren’t from your team. So instead of sending clients to a third party site, we’ve bought the third party site into the apps through Beamer. It’s a bit on the pricey side but it’s so easy to manage. H/T to my business partner Adam Motiwala for finding this gem.
Pear Deck: Given that a lot of education is going to be moving online over the next 12 months, educators need to find a way to create online learning that is natural and engaging. I saw this cool demo by Stacy Roshan who is the Director of Innovation and Educational Technology at the Bullis School. Basically, you can use Google Slides and use Pear Deck to get feedback from your students, answer quizzes, etc. It’s pretty nifty.
5. What I’m Reading: The Scientific-Industrial Complex
This past week, I read this great research article called “The Changing Structure of American Innovation: Some Cautionary Remarks for Economic Growth” (June 20, 2019) by professors from the Duke University Fuqua School of Business Ashish Arora, Sharon Belenzon, Andrea Patacconi, and Jungkyu Suh.
The authors argue that when it comes to R&D, Research in America has actually been going up, it’s the Development part that’s gone down. What’s the main reason behind Development going down? According to their research, the decoupling of commercialization by large companies from university research has resulted in less ideas going to market, resulting in a decline of innovation in America.
They divide the history of the Scientific-Industrial Complex into four distinct phases:
Phase 1: Agricultural Tinkerers (1850-1940)
America was academically behind Europe when it came to pursuing research. During this period, invention was largely the domain of independent inventors in agriculture and mechanical engineering.
It’s important to note that the big companies during that time didn’t invent technologies like the steam engine, braking systems, or telegraphy, they merely commercialized it by acquiring these inventions from inventors.
Phase 2: The Rise of Corporate Research (1900-1940)
During this phase, America began accelerating in scientific and industrial development and began imitating the German industrial practice of building in-house R&D labs.
The technologies developed during this period was far more complex then the technologies developed in previous periods so having more resources (people and funding) gave corporations the advantage when it came to innovation.
Moreover, antitrust legislation during this period made mergers and acquisitions a bit trickier, so instead of buying firms, it would be easier to invest in recruiting scientists for in-house research.
Phase 3: The Golden Age of the Corporate Lab aka the Scientific-Industrial Complex (1950-1980)
During the period following World War II, Federal funding for university research increased substantially resulting in a nice spillover effect for corporations. The rise of the university research institution helped create a nice assembly line between university students, professors, and recruitment into corporate labs.
Phase 4: Decoupling of Corporate Labs and University Research: The Rise of Startups (1980-2016)
However, starting in the 1980s, this assembly line between universities to corporate R&D labs began breaking down. While the research university sector continued to grow, corporate research began to decline. Most of the major companies either shut down their R&D labs or significantly downsized them.
According to the authors, the rise of startups siphoned off university research away from big corporations into smaller, more nimble startups. This has more to do with the specific branch of technology.
After all, venture capital has been around since the 1950s but it didn’t really start influencing the creation and acceleration of the startup ecosystem until the technology tree shifted to computer-based systems such as with companies like Microsoft or Apple.
In addition to the rise of startups, the authors also attribute the decline of corporate R&D to: competition, the 1980s rise of the Wall Street Profit Motive, outsourcing manufacturing, and a more relaxed anti-trust environment.
6. What Should I Ask?
I’ll be interviewing Clare Burnham, an innovation leader whose worked at Amazon, Starbucks, and Marriot, for the Innovation Nation podcast. What should I ask her?
7. Opportunities
For Investors
Investing in hedge funds isn’t just for the rich. There are stocks like Greenlight Capital Reinsurance (GLRE) and Third Point Reinsurance (TPRE) which function like a Hedge fund in terms of performance. (Barron’s)
Interesting thread on how to evaluate a specific stock: Roper Technologies (Industrial technology, radio frequency technology, instrumentation, energy systems & controls, and now software and network businesses) [Note: I’m not endorsing this stock, just thought the methodology was interesting]
For Entrepreneurs
7/23/20: Registration deadline for the 2030 Climate Challenge ($10 million in total available funds)
7/26/20: Application Deadline for the UNICEF Innovation Fund for Blockchain Startups (Up to $100,000)
8/1/20: Application Deadline for F3 (Farm Fish Food) Tech Accelerator. ($150,000 in funding + a shot to win an additional $25,000 during the demo day at the end of the program)
For Professionals: Here are some job posts from people in my network if you’d like me to make an introduction for you.
As I mentioned above, my friend Serign Jobe is always on the look out for fresh Engineering talent at Facebook. They are looking to hire Solutions Engineers. If you are interested, let me know and I’ll make an introduction.
Ogilvy is hiring a digital analyst.
Resman Property Management Software is hiring a digital marketing manager.
Arcadia Power is hiring for Director of Community Solar Business Development.
FedEx Services is hiring for a Full Stack Developer (Angular).
National Audubon Society is hiring for a Senior Manager for Donor Engagement and Stewardship (NYC, NY)
Here are some jobs posts that are not from my network but I came across them through social media and various email lists and thought they might be relevant to my network:
The University of Maryland Dingman Center for Entrepreneurship is hiring an events and marketing manager.
Saffron Road Foods is hiring an eCommerce Manager (NYC, NY).
Closing: Photo Story
I'll be teaching my first course at Georgetown University this Fall ("MGMT 224 Launching Entrepreneurial Ventures"). I’ve been working on updating the course syllabus which currently provides a great overview of design thinking & lean startup. However, I’ll be supplementing it with agile scrum so our up & coming entrepreneurs also learn how to manage tech projects.
To my fellow business leaders, what else do you think the next generation of entrepreneurs should know?
Also H/T to my business partner Adam Motiwala for finding this great image during a presentation on design thinking he did with staff at Oxfam and PBS.