Techno Optimists have waxed lyrical on their power and ability to positively change the world, while many are worried about the societal impact of their introduction.

I’m here to say I think their impact is underrated: if you’re thinking you’re thinking big enough, you’re not - the total addressable market (TAM) of LLMs is the entirety of white-collar work, and trillions of dollars are about to change hands.

The future is almost here

The concept of AI is, of course, not new. Works were being published in the 1920s, Turing wrote papers on it in the 1950s, and the phrase was introduced by John McCarthy in 1955. What we’re seeing now is not new; it’s just much better.

Through a confluence of compute power improvements, the development of transformer models by Google and others, and OpenAI’s public introduction of ChatGPT in mid-2020, the power of AI is at everyone’s fingertips. The worlds I follow are ablaze: Twitter, LinkedIn, and my friends all played with this thing and speak about it ad nauseam.

For me, GPT - especially GPT-4 - passes the Turing Test: it’s hard to tell if a human is typing the responses or a machine. There are many examples of ‘hallucinations’ where the model has produced something blatantly wrong, and many critics have used these instances as reasons to doubt its power. Similarly, people lament these models for having a training cut-off date, i.e., their datasets are outdated and don’t include recent information. These are genuine issues today, but they won’t be for long. I expect the recency of data issue will be fixed long before the hallucination problem.

Downplaying these technologies because of these two current issues is myopic.

My recent experience

Recently, I used GPT to build a toy Customer Relationship Management (CRM) platform. This experience and reflections of what I see in the market prompted me to write this.

For context, I started to learn to program when I was 13, have a degree in Computer Science, and worked as a software engineer at Morgan Stanley after graduating. While I’m technical and have a good foundation of knowledge, I haven’t coded professionally for years. I’ve run two startups since my last full-time engineering role and haven’t meaningfully kept up-to-date with coding. All skills atrophy over time, and I’ve certainly felt that.

Most important for this story, I used to write ‘backend’ code. I built risk systems for equity derivative traders that ran valuation models for stock positions and fed data into various systems. Another project I led automated the hedging of FX exposure when traders bought an asset denominated in a different currency than the one used to pay for it.

While highly technical behind the scenes, I didn’t build the User Interface (UI) associated with these projects. I’ve never spent time learning how to build UIs, and I don’t have the fundamental knowledge to assemble them myself.

My project - a basic CRM

CRMs are commonplace within companies I’ve founded and worked with. For those unfamiliar, they’re systems that allow users to create a list of customers, associate ‘deals’ linked against them, and perform analytics. This description is a gross oversimplification - the largest CRM companies are worth hundreds of billions, and people specialise their careers focusing on them - but will serve the purpose of this story.

I asked GPT-4 to help me build one. I’ve included the prompts I used to create the CRM at the bottom of this article. The workflow to build went something like this:

1. I asked GPT to create a Python backend, creating the concept of a Customer and Deal.

2. I iterated on this, extending the concepts, adding new functionality, and so on, copying and pasting code from my Python coding environment (PyCharm) to GPT and copying the fixes and new code back to PyCharm.

3. I cycled around this improvement loop until I had what I wanted.

4. I then asked GPT to build me a React frontend.

5. I repeated the same improvement and bug-fixing copy/paste cycle until I had something working that did what I needed.

6. I connected the frontend and backend and repeated the same copy/paste, improve/fix loop until I was happy with what I had.

After about four hours of coding - or, really, four hours of me copying and pasting text from one window to another - I had a (very basic) CRM. I could add deals, add customers, assign them deals, change the deal values, and chart them over time.

What I built was a toy: there’s no way I could release it or charge for it, but it was an incredible first step towards a usable piece of software, especially for someone who, it’s charitable to say, is rusty.

Reflections

GPT allowed me to build something in a language I’m unfamiliar with - Python - with a frontend - something I’m entirely unaccustomed to building - creating a working prototype in a couple of hours.

I had four thoughts:

1. I was surplus to requirements. I had the idea for the project, and I knew what I wanted, but other than that, I just copied and pasted things from one window to another. As soon as GPT has access to an entire desktop and as soon as it can do Robotic Process Automation (RPA), I wouldn’t need to do… anything. I’d just describe the app I was looking to build, review its progress, and give feedback and pointers.

   Humans will transition from being the creators of work to the curators - supervising, editing and improving work, not initiating it.

2. I had grossly underestimated the TAM of LLMs. For years, when thinking about technological progress and creative destruction, I thought job disruption would be at the low end: blue-collar workers would be replaced with robots, automated machines, performing their tasks 24/7, with no ‘people’ requirements - like wages, health care requirements, bathroom breaks, and so on.
   
   While that will continue to happen - Figure, Tesla and others are pursuing this vision - it’s not where the most change will happen in the next ten years. The scope, dynamics, and number of white-collar jobs, especially those relying on operating on large corpora of text - think lawyers, accountants, … even programmers, will fundamentally change.

   LLMs with full access to our computers and company file systems will take over all repeatable and explainable important-but-rote work.

3. This is unbelievably powerful. I wish this thing had been available during my programming days; it is so, so, so much better than the previous approach of Googling, reading Stack Overflow and other sites, sometimes, dare I say it, physical books, speaking to others, etc. None of these options is as powerful as this $20/month product.

   How would your role change if you had the smartest helper in the world aiding you 24/7 for $20/month?

4. It’s the best teacher of all time. This thing is: never tired, never snarky, doesn’t judge, there to help, available 24/7, highly personalisable, with the backing of nearly all information ever published online.

   How much better would you be at your job, hobby, or whatever if you had access to an always-on teacher whose only role is to help you?

Where will the value go?

Lots has been written and predicted about the future transition of money in this new economy.

Many commentators have predicted value accruing to incumbents. I expect the multinational software houses - Microsoft, Alphabet, SAP, Oracle, Databricks, Snowflake, etc. - will benefit in droves from LLMs: their reach and distribution into the major buyers, the ability to improve COGS or sell a net-new product will create a significant windfall.

From my point of view as a startup founder, whole new industries and ideas, previously impossible, have become possible due to the power of LLMs. These products will need to add significant end-user functionality above the underlying power generated by the LLM, but the market has that message, and “GPT-wrappers” are likely to become a thing of the past.

Regarding LLM providers, it’s clear to me that (1.) competition and open source will compete away the excess profits of any pure-play LLM provider1, (2.) training new LLMs will require an enormous amount of capital to be competitive with those already on the field, which are compounding additional benefits and power, and at some point - quite soon - new entrants will not be funded for fear of too much competition, and (3.) developers will want to use a ‘company safe’ provider that provides value above pure LLM functionality.

All this means, in my perspective, that a lot of the value will accrue to the hyperscalers - AWS, Azure, GCP, etc. - that’ll either build their own LLMs or provide third parties, likely through creative investment deals, al la Microsoft’s investment in OpenAI, Amazon’s in Anthropic, and others to come. Will we see another pure-play LLM provider post-Mistral? I don’t think so.

Then, of course, there’s money spent on jobs: $10.5tr in the US, according to ChatGPT2. How much of that spend will transition to LLM providers? Will companies need five people in FP&A, eight people in Legal, or ten in HR? What about the 150 programmers? Could their work be magnified through LLMs, or could the company just reduce headcount? This is an evolving field that is going to play out before us in real-time. I wouldn’t be surprised if we start seeing LLM-inspired layoffs before long, especially if the much-spoken-about-but-still-impending recession does come knocking.

Transition

I believe that roles in organisations that are currently pivotal will be substantially transitioned and potentially erased. Trillions of dollars will change hands over the next few years: from human capital to AI-backed service providers that understand the roles, apply AI - and other technologies - to them, and can execute faster, with no ‘people overhead’. People in my tech-bubble are talking about a potential future business worth $10bn run by only ten people. This, I think, is infinitely possible.

I mentioned in the subheading of this post that this is an optimistic view of the future, while I have outlined in this post that I think a lot of people are going to lose their jobs. I’m a capitalist, certainly, but also a human, and I don’t want to be seen as celebrating an enormous reduction in available roles.

What I’m describing is the next step in the creative destruction cycle: innovation leads to the obsolescence of old industries and jobs but simultaneously creates new opportunities and markets. Who knows what new roles this LLM revolution will create.

Who knew that the GPS in smartphones would create Uber, a $145bn3 juggernaut that has fundamentally changed how car drivers make money, opening new opportunities and flexibility to them at the expense of traditional cab firms? Who knew how introducing smartphone cameras would create the world of the digital influencer? Instagram, YouTube, TikTok, and other apps have created a world where teenagers can make millions from dancing, unboxing toys4, and creating makeup brands5.

Consumers still want to speak to someone when setting up their Wills; they still require a human to survey a house they’re considering buying; they want to be sure they’ve correctly submitted their tax return; they want help when choosing an expensive good in a store; etc.

I expect humans will continue this way. Perhaps the automation of some white-collar roles will create pressure on blue-collar roles - will the next wave of applicants to management roles in restaurants have multiple degrees? Will service in restaurants be improved because of the quality of applicants due to increased demand for those roles that, at this point - and even with AI improvements to come - require a human? What about call centres where humans field questions from customers? Will those be replaced with human-like voice systems?

There have been many conversations around Universal Basic Income (UBI) over the past few years. I expect these conversations will reoccur while the transition that has already started in the coastal cities of America and the central cities in Europe, China, and elsewhere proliferates.

The future is not bleak for humanity, but the introduction of LLMs marks a definite shift. Many have said before me that “AI isn’t going to take your job; someone knowing how to use AI will”. This, I do believe - it’s time to get learning.

The future is here and is relatively evenly distributed: $20/month gets you access to the most powerful resource of all time. What will you build?

Thanks for reading; if you enjoyed this, please do share. Contact me if you’re building something interesting in this area or have other ideas.

Share

Subscribe now

Disclaimer

I have invested in several AI-focused startups: Jitty, building a better home moving experience, indexing the property market with GPT, Channel, helping companies answer their burning questions by converting natural language to SQL, Quantly providing financial insights in seconds, Platformed, helping companies accelerate procurement processes, and AudioStack, generating studio-quality audio for adverts, walk-throughs, and so on, with AI. None of this article is influenced by my experience with these companies or is intended to promote them.

The prompts I used

Here are the first few prompts I used:

Write a simple CRM in python including APIs, including models for Customer and Deal

It produced a Python web app that did exactly what I needed. It included the concept of customers out of the box, with the ability to create, fetch, update, and remove them - the four horsemen of CRUD (Create, Read, Update, and Delete), a core paradigm for building web apps.

I tried to run the code it gave me in a code editor called PyCharm (think, Word for writing Python). It didn’t work, and gave me an error, so I asked:

I've run the pip install step on the command line, but pycharm can't find the module

GPT told me what to do. It fixed that error, but I got another:

now I get this error when running the app: UserWarning: Flask-SQLAlchemy integration requires marshmallow-sqlalchemy to be installed. warnings.warn( Traceback (most recent call last): File "/Users/tony/programming/python/crm/main.py", line 15, in <module> class CustomerSchema(ma.SQLAlchemyAutoSchema): ^^^^^^^^^^^^^^^^^^^^^^^ AttributeError: 'Marshmallow' object has no attribute 'SQLAlchemyAutoSchema'

It told me what to do, which fixed the error, and the app ran.

I wanted to expand it:

cool now it's running - give me some URLs I can use to call the API

It gave me the code to do so, but with some warnings:

I get python warnings saying "Shadows built-in name 'id' " in the code - what does this mean?

It told me what the problem was and how to fix it. Eventually, I added a UI:

write a simple react frontend that queries all of the APIs

This web app worked perfectly, so I asked to extend it:

add a text box to the front end to allow the user to specify the first and last names

And so it went on.

Tim Ferriss has a way of thinking about reading (non-fiction) I really like: focus on just-in-time information rather than just-in-case information. There are a lot of resources below; it can be useful to snack on content, but I suggest focusing on what really moves the needle at the relevant time.

For founders

• Founding Sales - a free-to-access 400+ page book on founder-driven sales. Especially useful for founders starting to create a repeatable sales process for the first time. If you want to buy the Kindle or physical version, link here.

• The 7 Powers, by Hamilton Helmer. This is the best book I've read on describing, creating, and maintaining moats with your business. Like all good concepts, these are obvious in retrospect, and you can quickly start describing them in businesses you observe. Hamilton's description of persistent differential returns is extremely valuable. If you don't have at least one of these Powers, stop what you're doing and reconsider how you might achieve one.

• - introduces many topics on financial documentation and thought processes in a very easy-to-access way that especially first-time founders might benefit from.

• by , former CEO of Outlier.ai and Flurry, Sean writes of his experience building companies at scale and shares his mental models for solving problems. I often learn something through reading.

• Mochary Method Curriculum and The CEO Within by Matt Mochary. The Google Doc later became a fully-fledged book. Both very accessible and easy to consume; they help you think about framing difficult conversations.

• by , former COO of PayPal, CEO of Yammer, now Partner of Craft Ventures. Infrequent posts, but excellent content on building, creating, and growing SaaS startups.

• Jason Lemkin - if you’re working in SaaS, and not following Jason “godfather of SaaS” Lemkin on Twitter, what are you doing?

• Amp it Up by Frank Slootman, and the ensuing book by the same name - Frank is an execution machine, has taken three companies public, and is currently CEO of Snowflake. You can do more, faster, with fewer resources, if you have the right people on the team, and push very hard.

• Good to Great by Jim Collins. Jim describes complex principles in a simple way. Two of his phrases I think about regularly - “First who, Then what”, and “Confront the Brutal Facts”. You can read pieces of the book here.

• Trillion Dollar Coach by Eric Schmidt, Jonathan Rosenberg, and Alan Eagle. An easy to read account of Bill Campbell, an impressive executive in his own right, and a coach to many of Silicon Valley’s high-performance CEOs.

• The Goal by Eliyahu M. Goldratt - teaches the theory of constraints in business. Written in a fiction style about a factory manager in a troubled manufacturing plant, Alex, the protagonist, continually runs into issues (bottlenecks) that he sequentially solves to make the business move faster, and ultimately make more money. Everyone needs a Jonah in their life.

• High Output Management by Andy Grove - a classic in management, and a must-read for those thinking about building high-performance organisations.

For business nerds

• Acquisitions Anonymous by Michael Girdley, Bill D'Alessandro, and Mills Snell - a fun and entertaining guide through the small business M&A space. Each episode the hosts run through a prospective deal. A great way to learn about a topic you unlikely know that much about.

• Every - a great collection of newsletters. I especially like Divinations, and read almost all of their writing.

• Invest Like The Best by Patrick O'Shaughnessy - amazing guests, and Patrick is a really talented interviewer; he often asks the questions I’m thinking of.

• The Tim Ferriss Show by Tim Ferriss. The first podcast I subscribed to, Tim is a hero. Outside of my parents, Tim is the longest-term teacher I’ve had in my life; I owe him a great deal.

• BG2 by Brad Gerstner and Bill Gurley. A great podcast covering multiple topics from two seasoned investors.

For engineers

• by - an excellent Substack on Engineering growth.

• Lethain by Will Larson - just piece after piece of gold.

For product people

• - would recommend to more than just product teams.

• Inspired by Marty Cagan - one of the OGs of Product Management, Inspired is a Product Managers classic. It’s highly aspirational, and frankly good luck implementing all the ideas, but it’s great… inspiration.

• Intercom on Product hosted by Des Traynor and Paul Adams. Covers a lot more than product (strategy, team building, scaling, innovation). “Fast gets good quicker than good will ever get fast” is the best quote.

That’s it for now. I’ll continue to update this article as I read (or remember) more. Let me know if I’ve missed anything obvious!

Please note, where I’ve linked to Amazon I’ve used my Amazon affiliate link: I may receive royalties from purchases. Regardless, I genuinely recommend these resources and have linked where possible to free-to-access versions also.

During the walk, I had to constantly stop, rewind, pause, consider the implications, rewind, and repeat. An hour interview took me three to complete. In the end, I listened to it twice.

The innovation David and Jason discussed has the potential to disrupt a large subset of very valuable, entrenched companies. At this point, public response to the podcast seems to be minimal, but I have no doubt that it will be truly transformational for many, and devastating for some companies that won’t see this change coming. At first they’ll belittle it, then it’ll be a nuisance when it starts rolling out, and eventually these companies won’t be able to react to it when it scales.

First they ignore you, then they laugh at you, then they fight you, then you win.

I’ve been lucky to meet all kinds of business builders, doing all kinds of things, throughout my career. But it’s rare for me to leave any kind of interaction and think that my counterpart is building something truly revolutionary. A pronounced improvement in some parts of the Internet, perhaps, but a revolutionary change to billions of people and major parts of the economy? Generally not. 

On the episode Jason brought on David, who runs The Product Board (TPB), a platform for launching businesses focused on hard-tech problems that require serious R&D investment to validate. The idea David covered is a machine from Cana, a TBP company, that uses 80 compounds which, when mixed together in the correct quantities and with the right amount of water, sugar, and ethanol, can reproduce the smell, taste, and mouth feel of virtually any drink on the planet. 

An inkjet printer for drinks. 

Using Cana’s touchscreen you can tap in your order, choose what you want to drink, and voila - it’ll be poured for you. The device has a water cooler inside so it’ll be served at the perfect temperature. The compounds come in a cartridge, which you replace as you run out.

A cola? You got it. You’d like a Duckhorn Cabernet Sauvignon 2004? No problem. Coffee? Exactly how caffeinated would you like it? We’ve got it.

Many analysts talk about the future of the world after the internal combustion engine and the impact of driverless, electric cars. There are the first order changes, the obvious shifts, like long-haul coast-to-coast trucking jobs will decrease. Millions of people are employed directly by the industry, and millions more in restaurants, garages, service stations, and so on, that service them1. But then there are the second order changes. Humanless cars shuttling themselves from coast-to-coast will require fewer pit stops in middle America, which leads to falling demand in traditional servicing like gas stations. This means a reduction in jobs, of course, but also trickles down to less obvious outcomes like a reduction in cigarette sales; half of those sold in the USA are transacted in gas stations2.

What led to the rewind-pause-consider-repeat process of listening to the podcast was my fascination with the second order effects of the Cana product. During the episode, Jason asks the classic Venture Capitalist question of “what happens if this works?”. The following presupposes that Cana (or another company) does get this thing to work. David mentions that the initial machine will be prohibitively expensive, but by version three they'll be more mass market. Let’s assume that version three happens. 

This is such a big idea I can only cover a tiny fraction of the impact it’ll have, but below I consider eight factors: retail, transportation, venues, the environment, Coca-Cola, Amazon, creators, and general life. 

Let’s imagine Cana - or a version of it - works. What happens now? 

Retail: the at-home disintermediation

While people will buy drinks for on-the-go consumption, a huge reduction in purchasing for the large family weekly/fortnightly shop will occur.

Stores will be able to reuse the real estate they currently dedicate to drinks - often several aisles - for other goods. The amount of warehouse space required will reduce. Both impacts will allow them to reduce their physical footprints.

Maybe retailers start having a Cana in-store, and consumers buy empty two litre bottles, self-serving their designer drinks; this is already starting to happen, with Coca-Cola introducing a bring-your-own-bottle vending machine in Japan, and the wine service Vagabond trialling similar ideas34. Perhaps micro stores will start to appear, unstaffed, giving consumers access to a Cana machine - a vending machine for tailor made drinks.

Stores will lose revenue because of the machines - although they may sell replacement cartridges, they’ll lose out on high margin drinks and alcohol sales. In this new reality, 6,000 square foot off licences will be replaced with a single machine5. Worth mentioning that I’m writing this as a Brit, where alcohol is sold in grocery stores, which isn’t the case in the US/Canada6.

Transportation: lightweight, cleaner, and cheaper

Planes that currently take large amounts of beverage stock into the sky just need to take a water tank and a Cana machine, allowing them to convert the water into whatever drink the customer desires. The amount of plastic used to wrap the drinks, and the amount of aluminium and cardboard used to house them, won’t be required. At the scale of flights internationally, this will be an enormous weight and waste saving.

This means that the waste produced by a flight will decrease impressively, and turning planes around at airports will take less time. Not to mention that fewer staff will be required to process the rubbish. 

American Airlines famously saved $40,000/year by removing an olive from each salad plate7. Imagine how much an airline could save in kerosene with fewer refrigerators needed to keep drinks cool? 

Will airline tickets become cheaper because of Cana?

Venues: more choice, smaller space, fewer staff

Mixer-taps in bars, nightclubs, bowling alleys, and cinemas will continue to exist, but rather than connecting to cardboard boxes containing plastic bags of syrup, they’ll draw directly from a machine that’s using the right compounds. 

If venues have the choice of which route to invest in - hundreds of cans of different drinks, wrapped in cellophane, with large refrigerators to keep them cool, and staff to load/unload delivery trucks - or a single machine connected to a water line and a cartridge of spare compound, they’re definitely going to choose the Cana.

And with such a rightful focus on the environment, regulation of new bars, pubs, and nightclubs may require venues to use the machine. Councils might even raise rates if a venue doesn’t use one.

Environment: dramatic reduction in waste

The data is staggering; 400 million tonnes of plastic are produced each year, 40% of which is single use89. Coca-Cola alone produces three million tonnes of plastic packaging per year, nearly one percent of the global plastic production10.

While many organisations, including those in the beverage industry, have been pressured into reducing the usage of plastic in packaging, we’re far from a plastic-free reality. And the continued increase in demand for plastic-packed products - driven by population increases - will create upward pressure on their usage of plastic.

Much of the plastic used by the drinks industry has to do with transporting the drinks to consumers; the branded plastic-wrapped crates of drinks come in, cellophane wrap around bottles and cases of drinks, etc. And drinks are heavy and take up space. They require large diesel-fueled trucks and trains to transport. Cana’s device leapfrogs this - it’s already in the consumer’s home; it has direct access to buyers, completely skipping the transportation step.

Presuming that the compounds are easy to retrieve or create, the environmental benefits of this device are mind boggling.

Coca-Cola: ceasing all bottling operations

Coca-Cola operates through two corporate identities. The Coca-Cola Company owns all the recipes, and produces syrup concentrate, which bottlers - the other constituent - are responsible for marketing, producing, and distributing. Given their global reach, The Coca-Cola Company owns various bottling companies throughout the world11. 

Through these operations, The Coca-Cola Company has become the world’s number one producer of plastic waste12.

Cana offers a radical alternative to classic production models. They’ve already ‘done’ the work of shipping and logistics, making any drink under the sun readily available in people’s homes and venues. All that Coca-Cola and other companies like it would need to do in a Cana-filled future is digitally distribute recipes via a website, right to people’s homes.

What does innovation look like for a brand like Coca-Cola in this new potential world? Do they become ultra high margin formula companies? Do they shed their logistical bottling and shipping operations, or at least pivot and focus only on marketing activities? 

Consumer sentiment continues to shift to be more focused on environmental issues13. Companies that continue to deal in plastic bottling and distribution will be severely disrupted. What do they do in a future where consumer choice will drive down usage of disposable drink containers?

While I’ve focused on Coca-Cola throughout this section, the lessons are applicable for all large drinks producers and bottlers; hundreds of billions of dollars of enterprise value is ready to be disrupted.

Amazon: the category owner?

Here’s my prediction: Amazon will buy Cana. They’ll go vertical on beverage production, already owning the relationship with the customer. The device will come free with a Prime subscription and because Amazon already operates a huge third-party marketplace, they can leverage their existing brand partners - who already know how to produce formulas, lotions, and potions. In this new reality, Amazon partners just shift their focus to this new consumer channel Amazon controls. It’s a slam dunk. There are only a few markets large enough to really stretch Amazon to the next trillion dollars in enterprise value. By purchasing Cana, Amazon becomes the largest distributor of drinks overnight; nearly a $2 trillion market including alcohol14. It’s a big enough offering for Amazon execs to jump out of bed for.

Amazon doesn’t like shipping heavy, low margin products - they’re harder to ship, take up a lot of room, and are relatively low value; drinks are the opposite of what Amazon likes to distribute. Shipping high-value lightweight cartridges full of compounds is right up Amazon’s ally. And of course, they know how to build hardware: the Kindle, Echo, Ring, Fire TV, Eero. Yes, this is a more complex product, and one that is based on significantly more detailed science, but they have the capabilities and human and investment capital to quickly get to grips with it.

If Amazon doesn’t try to purchase Cana, I predict that they’d try to create the product themselves. Though you’d imagine there’d be some considerable IP protection work going on inside TPB/Cana my pretty comprehensive search on LinkedIn only showed one person with a legal background working at TPB - and seemingly no one at Cana. I suppose science is Cana’s current core competency, but I expect IP (protection and defence, and attack from incumbents) will become their long-term protector of value.

Creators: everyone becomes a drink brand

In the episode, Jason and David speak about what’s clearly to come, but hasn’t been announced yet: an App Store for drinks running on the Cana touchscreen. Companies will be able to move out of the physical ingredient and drink logistics business and focus solely on brand and marketing execution, pushing their product to consumers through bits and bytes, not trucks and cellophane wrap. Presumably Cana will take an App Store-like clip of sales; 30% of the gross cost.

With this in mind, drinks brands will need to continue to attain the appropriate regional and worldwide licences and certificates to sell soft/hard drinks, and I’d imagine that some kind of approval process will be required to be featured on the Cana platform. But wow… the distribution channel potential with the low setup costs is incredible.

Imagine being able to change the composition of your drink with no retooling (albeit, likely requiring licensing and certificate reapproval). With Cana, companies will be able to launch new drinks brands with just a formula and zero inventory. It’ll radically change the landscape of available drinks.

Famously, consumer goods companies like Coca-Cola, Unilever, and P&G do not have access to the end consumer; they’re disintermediated by the retailer (supermarkets, convenience stores, pubs/clubs etc.). Despite the fact that Unilever have 2.5 billion daily users of their products, they really don’t know who the vast majority of them are or what drives them to purchase15. If Cana passed back more information about the end user to the brand after purchase, they could rapidly become the preferred sales channel for brands.

As a data nerd, the new reality could be hugely beneficial to disintermediated brands: in a future state, Cana might collect demographic data on their users to build algorithms that provide recommendations to their customers - “gym goers of your age like XYZ beverage after working out”. Frequency and preference of drinking, number of people in the household, rough ages (maybe age bands), gender? Data driven brands could leverage this data in ways that are currently impossible. Imagine the targeting capabilities - it’s the targeting capability of Facebook from 2010-2020 all over again.

Brands will move from earning thin margins - especially when selling to wholesale - to earning software-like margins. Their Costs Of Goods Sold (COGS) transferred to the customer who pays for the compound cartridge. Presumably we see their other main contributor to COGS, the Cana App Store take-rate reduce with time, scale, and competition. 

People will have profiles of how they like to drink coffee. They’ll store them in the Cana network, and share them with their friends when they visit their houses. Their friends will love their specific recipe and ask if they can have the formula. Soon they’ll be wondering, “Maybe other people would like my coffee formular too? … Maybe they’d pay for it?”. Soon enough, your dad will have his own branded coffee on the Cana network.

Life: altered 

I wonder if some brands with huge existing bottling operations will start innovating in new ways. Maybe they’ll start creating drinks with fresh fruit in them, something that can’t be replicated through the machine. Will Orange Juice with Bits become the top selling juice at retail because other juices are made through the machine, at home?

Will drinks as we know them now, bottled in plastic, aluminium, glass and otherwise, be relegated to something people buy because of their novelty? A bit like how, in a future world of electric-only vehicles, internal combustion engine cars will be a toy for the rich and the reminiscent, only accessible to those with the means who go out of their way to purchase them. Or the way records are treasured for their crisp sound and nostalgic reference to a world gone by.

Will you go to friend’s houses and marvel at the fact they offer you a Coke in a glass bottle, bought from a shop? “It’s more expensive, and I had to go to three different stores before I found it, but I think it just tastes more authentic” they’ll say, while trying to remember how to use a bottle opener.

In the episode that sparked this article, David mentioned Cana’s product roadmap and the fact that this initial version of the product is very much version one, and that in the future lots more will be added. Something that wasn’t discussed, and I think will have a massive impact on the adoption of this product and it’s beverages, is packaging: how can I take my lemon ice tea with me on the go? A disposable container won’t cut it: it won’t make this invention better than the status quo. We’ll need a sustainable way to transport the drinks themselves.

Homes will produce less rubbish - fewer plastic and glass bottles, cans, and so on. Less fridge space will be required to store drinks. Palettes will be widened, and people will try drinks they’d never usually reach for, just because they can. 

I can imagine a world where new homes come with these machines, in the same way they come with air conditioning or solar panels or built-in speakers right now. Home builders will become a sales channel for Cana. 

This machine has the power to reset our expectations of what products can do. It’s straight out of science fiction - The Replicator in Star Trek16 - and it’s going to pull the future forward. Our children’s children won’t know what it’s like to go to a store and see hundreds of plastic and glass bottle options.

All in, this is an invention that will radically, and I think rapidly, change the world. The next challenge for Cana - aside from shipping this science fiction device and reducing the cost to make it affordable for the majority of the world - is of course food, an even larger market.

PS. Just as I finished this article, Cana launched their product - there’s a new This Week In Startups episode to watch, and I can’t wait.

Disclaimer: these are of course only ideas of things that may happen, projected from various readings and learnings. All rights belong to their respective owners.

One of the memories I do have from being young is sitting in a classroom in year three, aged about seven or eight. A supply teacher was covering the lesson. We had to write a story, something I found intensely challenging. I looked at my page and willed words to come out. But they didn’t, not in any great volume. At the end of the lesson we had to show the teacher our work.

The boy sitting next to me went first. I sat looking at my half page of rubbish that had taken the whole lesson and had made my brain ache, listening to the teacher discuss my table-mate’s story. Much to my horror, I heard her turn the page. “He’s written more than a page!”. I panicked.

The teacher, very impressed with my colleague’s Ulysses, beamed and invited me up to be judged. I walked the three meters to her desk in what seemed like 20 minutes, and presented my workbook, full - or rather lacking - in tripe to the teacher. It was quickly obvious that her style wasn’t soft and encouraging.

“Is this it?” she enquired. It was, I reassured her. She invited my classmate back to the stage. My classmate, let’s call him James, returned carrying his heavy tome.

“Do you think yours is enough compared to James’?” she asked me as she flicked between the three pages of his modernist literature masterpiece. At this point the whole class was looking at me, while the teacher looked at my half page, and then back to James’ great work.

I can’t remember quite how I replied. I do remember 30 sets of eyes looking at me when I took my book back and sat down, and prayed for some kind of cataclysmic event to take place to remove me from the room.

Although an event I remember well for bad reasons, this episode taught me an important lesson about stage order: be first, or last, but never in-between if you can help it.

In year eight, when I was 12 or 13, my class took a spelling test. I looked forward to spelling tests in much the same way that a confectionary connoisseur looks forward to visiting the dentist.

This was a surprise test; a surprise to us all, and to my cortisol levels. “Just 40 words”. My English teacher walked around handing out pieces of paper for us to enumerate our great wisdom upon. I received my blank A4 and thought silently about how it knew as much as I did about how the words should be spelled (spelt?).

We began the test and it was clear that I’d end up with a “clean sheet”, in that I wouldn’t know how any of the damn words were spelled and my piece of paper would only contain the numbers 1-40.

The test included words that I was very familiar with. I have always had a great vocabulary,  but I have no idea how to turn spoken words into text. People often say “spell it like it sounds”. This line of advice doesn’t work for me, at all. Any word I know how to spell is from muscle memory and practice, not from first principles.

We get to the end of the test. It’s over, I think. I’ve actually managed to write a lot of “words” down, but I knew many of them were letter soup thrown at the page. I get ready to give my test to the teacher, for her to mark and give me the terrible feedback another day. It was over.

“Swap your test with the person next to you for them to mark”, my teacher instructed.

Fantastic.

My neighbour marked my test. There was a 40 on the page, but sadly it was only the denominator. The number above it was 1. It quite quickly got out that “Tony got 1/40 on the test”. I was sitting at the back of the room, and the news travelled to my teacher at the front; I remember her  saying “no, no, he can’t of, that must be a mistake?”. It wasn’t.

As I’ve grown up, various professionals, and my mother, realised I have symptoms of dyslexia (a word that I always have to Google to spell correctly, including just now as I typed this).

I don’t exhibit all the common symptoms, and some I do have are only mild, which I believe prevented a formal diagnosis when I was younger.

There is an enormous power to dyslexia that isn’t generally recognised. I definitely see things differently to other people, and make connections where others wouldn’t. I think the most important thing by far in life is relationships and connections, not knowledge and data retention. The phrase “it’s not what you know, it’s who you know” feels very natural to me. The creativity that stems from dyslexia has enabled people like Richard Branson see through what people often think is entirely impossible.

My inability to write at an early age, issues with spelling, and numerous other delights of being an undiagnosed dyslexic at school, prevented me from growing up to be the sort of entrepreneur who chronicles their journey in a public blog. It’s not all bad: my lack of expertise in the written word pushed me towards computer programming, the greatest skill I’ve ever learned. 

It wasn’t until last year when I read Stephen King’s On Writing, from where I took the name of this post, that I realised writing, like many skills, is easy. Simply practise everyday, work really hard at improving it for an extended period of time, and just like that you too could be an ok writer.

During the pandemic I started writing for work to distribute ideas and plans, that previously I’d communicate in-person. It’s an essential skill for any leader, and one I want to continue to hone, covering many topics here. Despite the fact that it takes me longer than others to create the same output, I’ve found I enjoy it.

I’m moving to a new chapter in my professional life, and I’m looking for a way to explore what I will do next. I’ve decided to try writing in this blog as a way to publicly learn more about myself, and share ideas.

Thanks for following along.