The world seems to have developed ‘internet brain’ over the last few years and headline investing is all the craze these days. But what if we spent some time looking at the dispossessed, unloved stocks? That’s my job. I’d like to shed some light on companies that are being overlooked in some way or another by bringing my perspective to the table, and I thought there would be no better way to kick things off than to start with Asana Inc. It does start with an ‘A’ after all.
Part 1
Company Overview
Asana is a cloud-based work management and coordination platform that helps teams organize projects, workflows, goals, approvals, and cross-functional operations. The company sits in the broader category of “enterprise work management” software — somewhere between project management, workflow orchestration, and operational collaboration.
It was founded in 2008 by Dustin Moskovitz and Justin Rosenstein, both former engineers at Facebook, Inc. and went public on September 30, 2020, via a direct listing on the New York Stock Exchange (NYSE). Dustin Moskovitz is the co-founder of Facebook and fits the founder, owner, operator profile very well. He also happens to be one of the richest people in America and I believe that is important when considering the risk profile of a company and the cash it may need to weather bad times. The alignment of a founder and what they choose to spend their time on is important and high signal.
The stack and why Asana’s position in it matters
If you imagine technological development as a multi-layered cake, then the following holds true. The lower layers provide capabilities, and the upper layers organize and abstract those capabilities. You’ll see this popping up everywhere. The industrial revolution for example, freed up time to move away from agricultural work and towards more abstract forms of work like project management. Agricultural work is still just as important as it has been, its just that there’s the opportunity for some part of a population to perform other work, enabled by the automation of lower layers; ultimately creating more for less.
Let’s spend some time outlining the AI stack.
Layer 1 – Physical substrate. It answers the question: “Can computation happen?”
Examples include power grids, data centres and GPUs. Companies operating at this level include Micron for DRAM, NVIDIA for GPUs and Super Micro Computer for servers.
Layer 2 – Compute and model layer. It answers the question: “Can intelligence emerge?”
This powers intelligence and is where the foundation models exist. Think companies like OpenAI, Anthropic and Google. It also includes inference systems, training, embeddings and open-source models.
Layer 3 – Tooling and API layer. This includes things like APIs, databases, communication tools, search as well as payment systems. This is where agents get to do things.
Examples of companies functioning within this layer are Stripe, GitHub, Slack and Salesforce.
Layer 4 – Coordination Layer. This is where workflows, tasks and permissions are organised in a structured way, creating an auditable trail as well as a system for human agent accountability.
It answers the questions: “Who does what, when, and under what rules?”. This is where companies like Asana are trying to live. A coordination and orchestration platform is essential for utilising structured AI within an enterprise. Unstructured AI would not be allowed.
Layer 5 – Strategy/ executive layer. This is the human layer and where we find ourselves. It includes things like the culture, goals and values of an organisation or an individual for that matter.
Imagining an agentic future
When I’m thinking about the shape of enterprises over the next decade, the future I foresee is one in which human beings and AI agents collaborate on these workflow platforms, with people ultimately having the final say as there has to be some form of accountability within in a corporate structure and how would you hold a virtual agent accountable? I also believe that companies would want to call up multiple LLMs (big and small) onto these platforms using an API, as having a sole LLM provider would give it too much leverage. So, you could imagine LLM companies providing packaged agentic products which are then purchased and called up to the orchestration layer, all the while leaving an auditable trail along the way.
The value inside of the company
Asana Inc. reached highs of over $142 per share and then subsequently fell to lows of under $6 over the last 5 years. That’s roughly a 96% decline in value for investors and a BUY signal for me. The topline revenue base has also been increasing with larger contracts in enterprises as well as US government with Asana Gov, rising from $227 mil in 2021 to almost $800 mil in 2026. Its cash position is also very strong with over $430 mil in cash and equivalents. Now compare that to ~ $250 mil in debt. It’s important to note that Revenue Retention is also very strong at 96%. In SaaS and subscription businesses, this is one of the most important metrics as it’s a measure of stickiness, pricing power and ultimately platform embeddedness. Coordination platforms are sticky in that they need to be adopted widely within an enterprise to gain full usefulness, and it is my belief that they will become more relevant in an agentic era, not less. This should give the company some pricing power over time as it captures more market share with its recurring revenue model. Bottom line is improving too and will continue to do so as the company’s expansion matures and it turns towards operating efficiency, streamlining itself and creating large amounts of net profit for its shareholders. To put things into perspective, Asana had a market capitalisation of less than $1.4 billion this year, trading at less than 1.5 times this year’s revenue.
Final thoughts
Typically, value flows up the stack over time as the lower layers get commoditised and the upper layers become network platforms that interface with people in a useful manner. This creates strong tailwinds for Asana as we move towards an agentic future.
Invest Today 