The Applications layer in AI is getting brutal

This story can play out in many ways.

Each use case has tens of funded companies. Each is churning out features rapidly, getting to parity faster than customers can imagine. Each has early traction and a worthy claim to win.

What will it take to eventually win the game?

1) Will it be about surviving the multiple shakeouts that each vertical/ use case will eventually see? Letting capital-bloated companies implode and letting the “tourists” give up…

2) Will continued product obsession be the key? Essentially refining the product beyond where others give up…

3) Will choosing non-obvious wedges/ ICPs be the way to differentiate & survive? Serve markets that others are choosing to ignore/ finding unviable to serve…

The technology is still so early, and we clearly have a few decades of upside left. Yet, there is a gold rush going on right now, which I am sure will push people to optimize for the short term.

In that case, will founders who are truly playing the long game ultimately win? Or is it more important to “surf the wave” in the present?

The former will look unattractive in current times and hence, will be undervalued and “contrarian”. The latter will appear to be imminent winners, yet could flame out.

Just some thoughts running through my head!

The futility of Plan B

Image Source: LinkedIn

Growing up in India, where inherent chaos makes sure most things don’t go according to plan, I got organically trained to always have a Plan B. The classic fallback option – the bylane you take when the main road is clogged because a minister is scheduled to pass through, the backup college seat you block in case you ranked low in the entrance exam for your top preference, or the autorickshaw you hail when the car refuses to start.

Look, I get it! Now that I am a father to 2 boys, I see the instinct parents have to ensure their children are tangibly & emotionally “safe” in all situations. So, I can appreciate why my middle-class upbringing was designed this way. To top it up, my technical education & early analytical jobs further pushed me into the world of scenario analysis & fail-safes.

Down the road, as I entered the risky world of startups, I naturally brought this instinct with me. While building, operating & investing in high-risk-high-reward endeavors, my animal brain would always push me to have a Plan B in my backpocket:

  • If this startup doesn’t work, I can always go back to Company X.
  • What if this investment fails? Let me spread my resources & take a smaller bet.
  • If I don’t like living in Country Y, I can always go back to India.

A few years into taking these asymmetric bets (presumably backed by Plan Bs), I expectedly started encountering failures, both big & small, one after another. They ranged everything from major projects going South & unforeseen external risks coming to the party to unexpected company restructurings & gross misjudgment of certain people’s skills & intent.

During a recent introspection of these adverse experiences, something interesting jumped out – every time I attempted to call on a Plan B for a specific situation, more often than not, it wasn’t really there. In some cases, the “backup” companies had changed their strategy & weren’t a fit anymore. In others, I had grown in a different direction & going to a fall-back option would be a negative step. Many times, people I was relying on to help materialize a certain Plan B had either fallen out of touch, were themselves dealing with adversity, or had changed their context & therefore, relevance.

So this was my lightbulb moment that inspired this post – in high-risk-high-reward situations, Plan Bs are….fictitious. The very nature of extremely risky situations is that they take you in unpredictable directions, change your context in unimaginable ways & leave you with baggage that’s hard to foresee. And all this happens in parallel to a rapidly-changing external environment that in most cases, becomes increasingly incongruent with your endeavor (most asymmetric projects are by definition, contrarian in relation to established rules of the game that the majority operates by).

This complex system renders even the most thought-through Plan Bs useless. Given asymmetric bets are driven by power laws (a few will drive a majority of the total outcome) & compounding (need a long enough timeline for ideas to mature, which is when outcomes start growing exponentially), positioning yourself to be on the right side of these rules requires going all-in for a significant period of time.

While having a Plan B provides the initial psychological space to initiate a risk, in my experience, it unfortunately also creates a mental mechanism to cop out of it, & even worse, often doesn’t provide the safe landing space it initially promised.

Going forward, my aim is to ditch the “Plan B” mindset in all asymmetric bets. A fall-back instinct comes from a place of fear, and while controlled fear can be a useful tool to drive alertness & urgency, it becomes adverse when acting as a roadblock to going all-in & persevering on a thoughtfully-chosen path.

It’s important to add here that while ditching the Plan B outlook, I will still proactively focus on avoiding the Risk of Ruin at an overall life level. Asymmetric bets require multiple shots at the goal & therefore, safeguarding the ability to keep playing is paramount.

On a related note, a mental heuristic I have recently started using while making asymmetric decisions I am 50-50 on – “which option is the fear side of my brain asking me to choose?”. In most cases, I then lean towards the other option!

I have found the following quote by Swami Vivekananda to be hugely inspiring in driving this mental transformation:

Take up one idea. Make that one idea your life – think of it, dream of it, live on that idea. Let the brain, muscles, nerves, every part of your body, be full of that idea, and just leave every other idea alone. This is the way to success.

Swami Vivekananda

As you consider this approach, I want to leave you with this outstanding scene from Christopher Nolan’s ‘The Dark Knight Rises’. As a frustrated Bruce Wayne is trying to catch his breath after yet another failed attempt at climbing out of the pit (he was using a rope each time), an old & wise prisoner gives him the mantra for successfully making the climb:

You do not fear death. You think this makes you strong. It makes you weak.

How can you move faster than possible, fight longer than possible, without the most powerful impulse of experience – the fear of death!

Make the climb…as the child did. Without a rope!

The Dark Knight Rises (2012)

Subscribe

to my weekly newsletter where in addition to my long-form posts, I will also share a weekly recap of all my social posts & writings, what I loved to read & watch that week + other useful insights & analysis exclusively for my subscribers.

David (Microsoft) vs Goliath (Google)

Image Source: Channel Futures

By coming out & saying “I want people to know we made them dance” in this clip, Satya Nadella has officially announced the beginning of an AI war with Google, who in turn, has also accepted the challenge by launching its own version of ChatGPT called Bard (unfortunately, the launch was botched, wiping out ~$100Bn from its market cap in a day).

Btw, how awesome was this clip? Just the look in Satya’s eyes & the intent behind the statement fired me up, & I don’t even work at Microsoft.

The first battleground of this war is Search. And it’s expected to see classic “David vs Goliath” type asymmetric warfare (Google has 90%+ market share in Search, as opposed to single digit % for Bing). Goliath has everything to lose while David has relatively fewer resources (existing Search distribution in this case).

So, how would each of them be thinking about war strategy? There are clues in asymmetric military wars that have unfolded historically eg. the US in Vietnam.

David’s view (Microsoft):

David can’t beat Goliath in conventional warfare due to the sheer gap in resources. So, it doesn’t make sense for him to engage Goliath by following standard rules in the open. David’s best bet is to engage with Goliath unconventionally, perhaps playing by a new set of rules ‘cos that’s when existing resources will mean less.

Real-world examples of this include (many of these ideas are covered in Sun Tzu’s Art of War, & can be seen in historical military confrontations):

  • Attack Goliath when he least expects it.
  • Target areas where Goliath has more to lose than David (eg. a classic nuclear threat).
  • Avoid a battleground that Goliath is familiar with. Take the battle to unfamiliar territories.
  • Prefer guerrilla warfare over all-out confrontation.
  • Use new modes of warfare wherein there is more parity with Goliath eg. economic warfare, communications warfare, strategic diplomacy etc.
  • Engage in indirect conflict by leveraging third parties that have some edge over Goliath.

If one closely observes how Microsoft is approaching the AI war in Search, it’s using many of the above elements.

First, under Satya’s leadership, Microsoft made itself stronger as a software conglomerate (Teams winning over Slack, LinkedIn’s massive moat, Azure taking a significant lead over GCP etc.). This has brought it more parity with Google at a group level.

Second, while Microsoft has increasingly become an agile & aggressive war machine, Google’s unthreatened monopoly in Search has eroded both the rate of innovation & sense of urgency from its operating culture. In a way, Microsoft is attacking Google when it is at its weakest culturally, while itself being at its strongest in a decade.

Third, the rise of AI is fast changing the rules of the game and as OpenAI’s ChatGPT has shown, Search is likely to look very different in the future. This change is being organically driven by a technology inflection, making Google’s existing dominant position in Search potentially less meaningful going forward.

Fourth, Search is a battleground where Google has much more to lose than Microsoft – the classic Innovator’s Dilemma. Microsoft can afford to take bolder bets, while Google has to fend it off while also protecting its existing business.

Fifth & final, Microsoft is leveraging a third party (OpenAI) as a main actor in this war. Unencumbered, unpredictable, agile & brave – third parties like OpenAI are hard to figure out & gameplan against by large incumbents, similar to how large military machines often struggle against guerrilla warfare.

So, how can Goliath counter these tactics?

Goliath’s view (Google):

While David’s main aim is to use his “brain” & make the battle as unconventional as possible, it makes sense for Goliath to use his “brawn” & exploit David’s vulnerabilities, in particular the disparity of resources.

Some ways he can do this include:

  • Attempt to drag the war back to familiar territory.
  • Open multiple fronts against David so he is forced to spread his resources thin.
  • Drag the war out for as long as possible, to drain David’s resources.
  • Cut off any access points that David can use to replenish.
  • David’s key strength is his morale so think of ways to destroy it.
  • Focus on de-throning the general & the army will automatically collapse.

So, while Microsoft’s challenge appears stiff, Google can use many strategies to counter it.

First, Google shouldn’t be deterred by the first punch. It can strategically prepare itself for a long drawn-out war & leverage its Search distribution might to outlast the competitor.

Second, it can open up multiple fronts against Microsoft to distract it. Potential areas include Cloud, enterprise workflow (GSuite) etc.

Third, given AI is so early, there isn’t likely to be any first-mover advantage. As we speak, many high-quality teams are already working on OpenAI competitors, providing Google with a valuable opportunity to partner with them & make up for lost ground.

Fourth, one of Microsoft’s major strengths is its leader. Google should be open to making moves in the market that distracts Satya or puts him under pressure.

Fifth & final, Google should use this rare competitive pressure to revitalize its execution culture. Perhaps one of the founders returning to the helm is a possibility? If taken in the right spirit, this is a valuable opportunity for the company to reset itself for the next 2 decades.

These are just game-theory conjectures at this point. Given the resources at the disposal of both companies, this AI war in Search is likely to unfold over several years. We will see many of the above tactics get played out in each scene, which will be tremendous learning for lifelong students of strategy like myself.

Subscribe

to my weekly newsletter where in addition to my long-form posts, I will also share a weekly recap of all my social posts & writings, what I loved to read & watch that week + other useful insights & analysis exclusively for my subscribers.

Using the Focus Canvas to Cut Burn

As the fundraising environment continues to get harder in 2023, founders & investors are deep into rationalizing business plans & finding ways to cut burn. The first temptation is to follow what I call an “excel sheet” approach – starting with the largest expense items without enough strategic thinking around the revised set of goals, business constraints in this new environment, what is working well right now, & how capital should be most efficiently allocated going-forward.

As opposed to big companies, startups operate with a finite runway, trying to address significant customer problems that remain unaddressed by large incumbents. This requires constant innovation – essentially doing hard, non-consensus things across the stack, everything from technology & design to customer experience & business model, that incumbents aren’t doing.

While big companies can afford to be relatively unscientific in cutting costs & still tide through tough macros with the help of their existing PMF, startups unfortunately, have no option but to play offense at all times in order to continue innovating & thereby, give themselves a chance to survive & succeed. In financial terms, this implies investing incremental $$ into innovation that drives more revenue (& profit), which is what will ultimately save a startup, not investor cash sitting in the bank.

So how should founders think about playing offense while being capital-constrained? I would like to propose a thinking tool called the Focus Canvas:

  • As a first step, rather than focusing on P&L line items, break down your business into specific buckets. These could include customer segments (eg. Individual, SMB, Enterprise etc.), product lines (eg. shrink-wrapped, custom deployment, pure services etc.), platforms (eg. desktop app, iOS, Android, browser extensions etc.), distribution channels (eg. self-serve, inside sales, direct sales, channel partners etc.), geographies (eg. US, EU, India etc.), teams by function/ type (eg. engg., product, design, sales, marketing, offshore contractors, agencies etc.) & other buckets that are relevant for your business.
  • Arrange all the relevant buckets & their constituent elements on a single page. This is your Focus Canvas.
  • On the top-left corner, list the most updated business goals for this year that all stakeholders in the company have aligned on. These could be things like “hit $1Mn ARR”, “show x% retention”, “start fundraising in Q4” etc.
  • On the top-right corner, list all the business constraints you expect to face this year. These could be things like “12 months runway left”, “only 2 backend engineers”, “sales cycle taking 6+ months to close” etc.
  • Now, as you are looking at this Focus Canvas, try and answer the question “what is working well right now?”*. You need to define “working well” for each bucket as per your specific context, also taking into account the above goals & constraints. It could be driven by one or more of revenue growth, most profitable, highest ROI, generating the most valuable feedback, creating the most differentiation, highest team productivity etc. *Note: this step is well-suited for a team workshop/ brainstorming session.
  • The most important step – for each bucket, put a ✅ in front of the element(s) you believe is your best bet to achieve this year’s business goals while navigating expected constraints. Then, ❌ out all other elements in the bucket. This is where ruthless focus is extremely important for the Canvas to do its job well – ideally, force yourself to ✅ only your #1 focus element. In the case of most startups, that’s probably all you can afford to execute anyway.
  • Finally, take the ✅ element from each bucket & weave them into a simple, 1-2 paragraph Focus Narrative. An example to illustrate this – “In 2023, we will focus on the Enterprise customer segment & offer them the standard product suite with a billable custom deployment services wrapper. The product roadmap will focus on the desktop app. We will double down on the internal sales model for distribution, with founders pitching in for strategic logos. To increase our team’s efficiency, we will significantly reduce contractor headcount & re-allocate them to full-time hires in engineering & internal sales.”

This Focus Canvas now provides a clear & strategic view of opportunities to both cut burn & re-allocate resources, while staying on track to achieve business goals & making progress toward PMF. The Focus Narrative can be used to socialize the going-forward strategy across teams in an easy-to-remember way. If used well (read: with ruthless focus), this approach can help startups in playing offense even in a tough economic environment.

PS: sharing a Focus Canvas template that you can use as a starting point. Feel free to download a copy & modify it as per your requirements.

Subscribe

to my weekly newsletter where in addition to my long-form posts, I will also share a weekly recap of all my social posts & writings, what I loved to read & watch that week + other useful insights & analysis exclusively for my subscribers.

True competitive advantage is a “Mesh”

Today, I was remembering my strategy professor from ISB Prof. Prashant Kale. All my batch-mates will agree that he was probably one of the best, if not the best, professors that year for Class of 2010. In particular, one of his classes where he taught the legendary Southwest Airlines strategy case is imprinted in my consciousness — I even remember exact drawings he created in the class.

The key takeaway from that class was that the competitive advantage of Southwest Airlines wasn’t a single linear element; rather, it was a “Mesh of inter-connected, inter-dependent, self-reinforcing activities” that was almost impossible to replicate by competition. Eg., turning around a plane in 15 mins (fastest in the industry), which required the gate staff to operate at a certain cadence, which in-turn, required the check-in staff to do certain activities at a certain speed etc. Essentially, executing any precedent element well made the next dependent element even stronger; conversely, if one of the elements was poorly executed on, the entire mesh advantage ceased to exist. This Mesh model has been a transformative strategy concept for me, and has been a key foundational element of my thinking.

Illustrating the proverbial “Mesh of Competitive Advantage”

I recently got reminded of this concept again after a decade, this time in an entirely different context of public market investing. I was reading the Q1 2019 Quarterly Investor Letter by O’Shaughnessy Asset Management (OSAM), a top quant asset management firm founded by the legendary public markets investor Jim O’Shaughnessy. This letter is particularly fascinating, as it talks about how to cultivate real edge as an investment firm.

OSAM defines the following framework for real investing edge (quoting the letter):

  1. “Real investing edge should (instead) be cultivated at the organizational level.”
  2. “Properly built, an edge should be very difficult or impossible for others to replicate.”
  3. “Ideally, the edge naturally increases over time — something venture capital investor Keith Rabois calls an “accumulating advantage.”

Specifically, OSAM does 2 things that drive the edge — 1) consciously building a Research Graveyard, which essentially means doing lot of research, data analysis and number crunching projects that don’t necessarily lead to immediately improved investing outcomes but increase the overall ideation & knowledge of the firm in a compounded way over the long term; and 2) building tools (data-sets, software and combos thereof) and then deliberately opening them up publicly for other researchers to use (like the way Amazon opened up its cloud infra to developers, creating AWS), whose usage, in turn, has generated some of the best research insights for OSAM.

As a finance and investing person, while both these elements are individually interesting to me, the real deal was this sentence (again quoting the letter):

“These things — software, data, research partners, our graveyard, even the podcast and our twitter activity — all link into and depend on each other, which makes each more valuable and harder to copy.”

“Think back to the ownership data project. Without other connected tools, that would have just been a dead idea — time wasted. But now the ownership data set has become a critical piece of another software tool we use for clients called Portfolio X-Ray. It is now also a new data set available to research partners, who may find something interesting that we did not.”

This is the “Mesh of Competitive Advantage” all over again. A set of activities that, while look replicable in isolation, are almost impossible to replicate by competition as an inter-dependent, inter-connected, self-reinforcing system. This, my friends, is where true competitive advantage comes from!

This is the same reason why, despite having an incredibly transparent investing strategy, framework, terms, processes & activities, other venture firms are unable to replicate the Y Combinator model. This is the same reason why I saw Alibaba winning in China eCommerce (a rhythmic mesh of commerce, payments, logistics, cloud and advertising that is perhaps, impossible to replicate even with infinite capital). It’s the same reason why, as Prof. Kale told us in 2009, Southwest won in the US airline market.

As I think more about this concept in the context of tech startups & Silicon Valley, I believe the following execution elements are important drivers of on-ground success:

  1. Mesh creation has to be deliberate — it’s really hard to defend individual, linear advantage elements in the long run (eg. just having more capital than competition).
  2. Constituent elements of the Mesh need to flow from an authentic place residing inside founders/ leadership teams — what we call as DNA, else it’s hard to sustain.
  3. The Mesh strength compounds over time — demands consistent execution over a long-enough period of time.
  4. This is why true diversity in the team is important — underlying this Mesh of Competitive Advantage, is really, a Mesh of diverse people, each contributing a uniqueness that, combined as a whole, is a super-power. Like the YC Founders or Paypal Mafia.

Would love to hear how you have created competitive advantage for yourself/ your companies.

Side-note: the “Mesh of Competitive Advantage” can also be used to differentiate yourself as an individual professional. Instead of being linear in your career, try to create your own cross-functional, cross-sector, cross-cultural & cross-market Mesh of skills & experiences that, while individually might not look compelling enough, combine together to give you a truly differentiated world-view and approach to life. In today’s age of automation & tech-driven leverage, having this type of Mesh is worth its weight in gold as it can’t be replicated by software; rather, software & tech tools can be used to leverage it up & further magnify its impact.

Related note for parents raising kids in Silicon Valley: given we live in an echo-chamber, with template approaches to pretty much everything (from hiking at the same spots, wearing similar Patagonia vests, to starting up and listening to the same podcasts), it’s important we consciously expose our kids to the non-Silicon Valley world. We would do well to nurture their authentic qualities and original habits, whether they fit with the Valley way of doing things or not. In fact, I would argue that the more contrarian or differentiated these intrinsic personal qualities are, the more we as parents, should encourage them. This will set them up as adults to create their own, authentic “Mesh of Competitive Advantage” that stands the test of time and disruption.