CDOQ

Earlier this week, at a roundtable hosted by Chargebee , a group of GenAI (AI) founders compared notes on a shared learning: building and selling AI products to businesses is not just harder than selling SaaS — it is "fundamentally different'. Many entered the market expecting to ride the SaaS (Software as a Service) playbook to success. They are now discovering it’s written in a different language. Here are five reasons why: 1. The Vanishing Customer Profile In SaaS, identifying the Ideal Customer Profile (ICP) is often straightforward: a clear job to be done, a known budget holder, and measurable ROI. In GenAI, that clarity dissolves. Everyone wants to “experiment with AI.” Few want to commit. Interest is high, intent is elusive, and trial usage often masks the absence of a real buyer. 2. The False Promise of PLG Product-Led Growth — the darling of modern SaaS — is stumbling in the AI world. While PLG drives traffic and trials, it fails to convert at scale. AI products often...

The Complexity of B2B Decision-Making B2B marketing isn’t about impulse-driven purchases—it’s about navigating a maze of stakeholders, each with different priorities. A CRM software purchase, for instance, must satisfy the CFO’s cost concerns, the CRO’s sales enablement needs, the CIO’s security requirements, and the CEO’s strategic vision. These competing perspectives make sales cycles longer and more complex, demanding that marketers craft tailored, high-impact messaging. Evolving Buyer Personas: From Static to Dynamic For years, B2B marketers have relied on buyer personas—fictional representations of ideal customers based on roles, challenges, and goals. Personas like "Strategic Sarah" (VP of Operations focused on efficiency) or "Risk-Averse Richard" (Head of Procurement prioritizing compliance) help humanize audiences. However, traditional personas are often too rigid. Today, AI-powered insights allow for real-time persona refinement, using behavioral data, sent...

When we listen to stories we have the tendency to suspend disbelief in order to be entertained, says Paulos [John Allen]. But when we evaluate statistics, we are less willing to suspend disbelief in order that we are not duped.  Storytelling vs. Statistics  Paulos goes on to describe the two types of errors in formal statistics. Type I error occurs when we observe something that is not really there. A Type II error occurs when we fail to observe something that is actually there. According to Paulos, those who like to be entertained and wish to avoid making Type II error are more likely to prefer stories over statistics. Those who do not necessarily yearn for entertainment but are desperate to avoid Type I errors are apt to prefer statistics to stories.  - From the book: Investing - The last liberal art  

"Failure to explain is caused by failure to describe." - Benoit Mandelbrot

The length of the cigar butt is influenced by how many cigars remain in the humidor.

Which is better: a person of good character with good intentions who tries to improve society but fails, or a person of questionable character with self-interest in mind who does something that benefits him and ends up making society better?

No two people live the same life. 

If you get the 'what' wrong, it's hard to get the 'why' right.

Nature dominates nurture. 

Leaders who want to do more with less have been doing less with more. 

It's ironic how caring for our home—planet Earth—seems to drift in and out of fashion.

In some clocks, the pendulum swings only to deceive—it has nothing to do with keeping time.

A good product manager says 'no' a lot and can explain why, just like a good parent. Do disciplined parents make good product managers, or is it the other way around?

A business with momentum solves many internal problems and obscures others.

A startup begins without a competitive advantage. It is the management’s responsibility to build one using market feedback.

The mind can find connections where none exist. 

When things aren’t working and you want to make a change, focus only on what will impact the outcome. Trying to improve everything at once can slow you down and distract from what needs to be done.

A hero needs a villain, but a villain can exist without a hero. 

My dear friend, Ralph Abraham , passed away on September 19th, 2024. He was 88. It is one of the greatest honors of my life that Ralph considered me his friend. In my entire life, I have never met anyone whose intellectual depth, openness, and kindness matched Ralph's. My family and my parents feel the same.  I met Ralph in November 2009 at Esalen where I attended a workshop he taught on Human Potential . Since then, I met him regularly every three to four months for lunch. I cherished my time with him. I often joked that I didn’t understand half the things Ralph said, and the half I did understand made me rethink everything I knew. There is some truth in that joke.  Ralph discussing Consciousness  Ralph profoundly influenced me. Here are some of the ways: 1. Consciousness: This was probably the most discussed topic between the two of us. He shared with me that there are three levels of consciousness :  (i) Individual (humans and other sentient beings)  (i...

Are the parents teaching the children, or are the children teaching the parents? My son teaches me many things effortlessly, while I strive to teach him a few things—with a lot of effort. It may take decades to see if I succeed. :) I asked my parents if they had a strategic plan to teach me values. Their response was that they demonstrated values through their actions, believing that children learn by observation. They didn’t worry too much about it. :) Perhaps I worry too much about making my son a good human being. While it’s not entirely in the parents’ hands, I still want to try. Observing my son, I’ve been focusing on teaching him the following: Joys of childhood 1. Actions have consequences.  Children often have no concept of this idea. They live in the moment, which may be why they’re so happy. For example, throwing a toy on the floor might break it, or not stopping at a red light could lead to an accident. These lessons might eventually be learned through experience, but I’...

New thinking requires new context.

It’s hard to have an independent mind when surrounded by many minds. 

It is harder to sell a cheaper version of an expensive product than the other way around. 

When you know a few ways of doing things, it is hard to find new way of doing things. 

Like children, employees often learn more by observation than by instruction.

Busyness can be a form of laziness. 

Fail fast a few times but don't make a habit of it. 

It is hard to develop an acquired taste when alternatives are always available. 

All art lies between Mondrian and Pollock . 

Intelligence is a function of the environment. 

In the semiconductor industry, the adage 'eat or be eaten' emphasizes the need for continuous improvement and strategic expansion, often by acquiring adjacent businesses with complementary or superior technologies. The SaaS industry operates under a similar principle. To thrive, SaaS companies must constantly innovate and enhance their product offerings, whether through internal development, strategic partnerships, or acquisitions. 

Temporariness is permanent. 

Last week, I attended the Bay Area IPO Summit , hosted by the Connor Group . The process of going public is a complex ecosystem involving a wide range of specialized participants. Below are my key insights from the summit, shedding light on the current dynamics of the IPO market. IPO: A milestone in a company's journey (image credit DALL-E) Current State  1. Improving Market Conditions : The IPO market is showing signs of recovery. Year-to-date, 2024 has seen 30% more IPOs than the entirety of 2023. Additionally, over $300 billion in venture capital "dry powder" remains available for investment in the US. 2. Revised Definitions: Investment bankers now define "high growth" as achieving over 30% year-over-year revenue growth, while "profitability" is simply a business generating more than break-even profits. 3. Supply Constraints: Contrary to popular belief, there is no shortage of investor demand for IPOs. The real issue lies on the supply side—few hi...

Earlier this week, I saw a new friend for coffee—one of the few truly wise people I’ve encountered in Silicon Valley. What follows is unadulterated wisdom from my friend:   Wisdom over coffee 1. Separate style from substance. Most people judge others based on their message delivery style rather than the content itself. To understand people better, focus on what they’re saying, not how they’re saying it. 2. Test for trustworthiness. To know if someone is trustworthy, look for contradictions in what they say. 3. Evaluate the substance-to-noise ratio. If someone speaks charmingly for a long time to convey something simple, they might have a hidden agenda.  4. It is hard to figure out in an interview if the candidate is trustworthy but you can understand their grasping power [intellect] and articulation [communication] skills.  5. Smart bosses do their best to keep their smart employees. 6. Watch for noise in data. For example, data might show that firetrucks a...

"Travelling comfortably dominates people's thinking when they should be thinking about destinations."  -Nick Sleep 

Semiconductor Business FAQ  Q1: What are the fundamental differences between the semiconductor business and a typical B2B SaaS startup? A: The semiconductor industry differs significantly from software-based startups, primarily in terms of investment scale, development cycles, and market dynamics. Here's why: i) High Capital Investment & Longer Timeframes: Semiconductor development demands significant upfront investment, often reaching tens of millions of dollars, with multi-year development cycles before a product can even be tested. This contrasts sharply with the iterative, lower-cost nature of software development. ii) Targeting Large Markets: Due to the hefty initial investments, semiconductor companies must target large, high-volume markets to achieve profitability. A single design win in a consumer device like a smartphone can translate to millions of units sold. iii) Limited Iteration: Unlike software, where updates and bug fixes can be rolled out relatively easily, mak...

A few days ago, I attended one of the first non-academic robotics conferences, Actuate , in San Francisco. One of the most fascinating insights was the early evidence suggesting that robots trained using a variety of data ( cross-embodiment training )   can outperform specialized robots at their own tasks. For example, a robot trained on diverse tasks—such as walking, assembling, and manufacturing—could perform better at picking objects than a robot trained exclusively for that task. Robots in our life (image generated by DALL-E) Following are my key observations from the conference:  1. Physical Intelligence , a silicon valley startup, raised $70M in seed round to build a general robotics model which can be used by different kinds of robots to do various tasks.  2. This idea of general model for robotics is called Robotics Foundation Model  (RFM). The concept is pretty similar to Large Language Model (LLM) which are the basis for Generative Artificial Intellige...