The World is Changed
The world is changed. I feel it in the water. I feel it in the earth. I smell it in the air. Much that once was is lost…
And yet as the tide washes away the failed startups of the past, a new era has already begun. An era where venture capital is scarce, and artificial intelligence is everywhere. If they want to succeed, companies must embrace this new reality. Once a mere whisper in the wind, it now roars like the mighty Anduin, full of potential.
Root System was forged in the zero interest rate environment, and the business model was well suited for that condition. At the time of our inception in the spring of 2020, sub-one percent interest rates made venture capital funds an attractive option for large institutional investors. This abundance of capital flowing into venture firms meant that it was easier than ever for early stage startups to get funded. Since then, two major shifts have altered the landscape. Interest rates have risen to 6%, causing venture capital to dry up. And generative AI promises to change everything about the way that software is built and used.
As interest rates increased, the overall levels of early-stage venture capital investments began to decline. The snapshot below represents the relationship between higher interest rates and early-stage venture capital investment. This data is just from the last two decades, and the interest rate is based on the 10-year U.S. Treasury rate. And yet, these circumstances are not entirely bleak — investments in new early-stage companies continue to be the lifeblood of breakthrough technologies and innovation. When everything dried up, it was early-stage that continued to get funding.
In the crucible of this new world, startups with software products must respond with strength and resolve to stay competitive. Revenue growth, customer traction, and profitability are crucial in this new environment, even for venture-funded companies. The startups who survive will do so by leveraging generative AI — not just for building their products, but also for teaching customers how to use their products and for running their own companies effectively.
The most effective founders will deploy AI anywhere in their operation where there is unstructured data that needs to be structured. When people need to be trained to use new tools, AI will be used to streamline the process, whether those people are employees or customers. Generative AI makes it significantly easier to build software. This allows companies to launch with fewer resources, but it also results in increased competition and reduced defensibility. Startups must now differentiate themselves by best understanding their customers’ problems.
While the rise of generative AI delineates an exciting shift that reconstructs how, what, and who builds products, other environmental changes intersect with these technological advancements. It’s vital to consider the following questions in connection with a company’s brand —
- What is your ICP (ideal customer profile)? Are you selling to a particularly complex market, and does the sales cycle length create a barrier to new entrants?
- Can you legitimately refer to your company as a leader in a ‘soft’ but crucial aspect (such as compliance or security)?
When how you build software gets easier, what you choose to develop becomes more critical. In terms of a company’s product, it’s vital to grapple with —
- Are you leveraging frameworks, such as Jobs to Be Done, to ensure that you’re constantly working to perfect the alignment between the scope of your product and the needs of your customers? To deepen your product/market fit?
The art of product management remains the most challenging aspect of early-stage startups. Even with reduced R&D costs, building features still carry other costs, such as negative user experience impact. Competition that doesn’t turn on technology turns on UX and overall value benefit. Including too many features is often worse for your product’s usability than too few.
If generative AI is used purposefully, startups can lower their burn rate by reducing the cost of coding and software development. For example, natural language processing (NLP) can interpret human language requests and convert them into functional code. This can be particularly useful for startups with limited technical expertise as it allows them to create or modify simple applications without a full development team. Additionally, instead of developing certain functionalities from scratch, startups can integrate existing AI-driven APIs and services. This approach can save considerable development time and resources. However, it’s important to understand that these use cases are predicated on quality assurance and human oversight to ensure that the code generated meets the required standards; and functions as intended.
When AI is used intentionally and in conjunction with human oversight, it has the ability to bridge the gap between idea generation/validation and actionable steps. Advanced algorithms can analyze market trends, user preferences, and past success stories to generate unique and viable app ideas. Further, AI can assist in validating these ideas by conducting sentiment analysis, competitor analysis, and customer surveys, which still help entrepreneurs make decisions in a data-informed way but without an army of analysts to source the intelligence (or a large payment to Statista).
AI also has the potential to transform certain aspects of the product development process, such as rapid prototyping. Startups can use AI-powered, low-code development platforms that offer pre-built templates and components. This can help companies leverage natural language understanding to interpret ideas into features automatically, which may substantially speed up the validation, if not the development processes. However, any discussion of the opportunities that AI brings forth should also address the caveats that come with it.
Companies should be advised against using LLMs and models to be their innovation brain — because LLMs are trained on the past! This makes them far more useful for suggesting research topics and gathering background data than generating truly new ideas. Sometimes Generative AI can “hallucinate” or create false content, or content that is not in line with the intended purpose of the model. Consequently, having LLMs write your copy rather than edit and clarify and achieve tone is not recommended. Over-reliance on the native model output will likely produce generic and not targeted product features, and companies run the risk of falling into legacy patterns (because LLMs are trained on the past! They are incapable of true innovation). Ultimately, AI technology like LLMs is best used to help humans do their jobs faster and more efficiently — not replace them.
The possibilities of generative AI are endless. But it’s important not to become so hyper-focused on the technology that you lose sight of the value your product delivers. It has never been easier than it is now to collect enormous amounts of customer information and perform sophisticated analysis on it; but it’s also easy to become seduced by the patterns, or correlations these numbers show. The article in Harvard Business Review, Know Your Customers’ “Jobs to Be Done,” unpacks this phenomenon when the authors’ write “that the focus on correlation — and on knowing more and more about customers — is taking firms in the wrong direction. What they really need to home in on is the progress that the customer is trying to make in a given circumstance — what the customer hopes to accomplish.” Only when a company gets curious about the story behind what causes or drives the purchase can it better understand the social and emotional factors at play.
Indeed, the world has changed. The new VC landscape intertwined with the emergence of artificial intelligence heralds a transformation profound — one who’s currents are deep and unknown, bearing both the promise of bountiful harvests and the peril of unforeseen storms. With great power comes great responsibility; those who shape and harness its gifts must also navigate all future endeavors with care.