How to adopt a product-led approach when building on top of AI?

Crazy news guys, we’ve just launched a startup program for AI founders (the perks are crazy).

You can get $2000 worth of credits (Anthropic, Mistral, OpenAI, and Phospho) + a call with our amazing team to guide you in your product-market-fit journey.

You can apply here.

In recent years, there has been a growing buzz around the term “Product-led Growth” (PLG), and for good reason. This approach to building and scaling businesses has been highly effective for SaaS companies in particular.

Product Benchmark Reports have found that companies with a product led approach are more than twice as likely (100%+ year over year revenue growth) to outgrow sales-led companies, especially those with a freemium model.

It also found that PLG businesses are valued on average 30% higher than the public-market SaaS Index Fund. Which makes PLG companies significantly more attractive investment opportunities.

But what exactly is Product-led Growth? In this article, we’ll provide an introduction to the product led approach and explore how to adopt it effectively by using analytics tools when developing AI products such as LLM apps.

What is a product led approach?

At its core, Product-led Growth is a business strategy that prioritises the product experience as the primary driver of growth. This idea started in 2016 and has changed the tech industry, especially in the SaaS sector.

In this model, by putting the product as the primary driver, the focus moves from old sales ways to making a great product experience. Companies put more effort into making their products better and use data to make smart choices. This leads to spending less on acquiring new customers and making more from each customer over time.

Each of these approaches has its merits and ideal use cases, but the shift towards product-led strategies is indicative of a broader market trend towards autonomy and value-driven user experiences. Companies that understand and implement a robust PLG strategy are well-positioned to outperform competitors.

How Slack and Zoom thrive with a product led approach

Slack and Zoom are great examples of success with this approach. Slack's easy-to-use design and word-of-mouth helped 77% of Fortune 100 companies pay for it. Zoom grew fast, going from 3 million users in 2013 to 300 million in 2022, making over $1.37 billion in profits by 2022.

Amidst the shift towards remote work and virtual meetings, Zoom's intuitive and user-friendly platform catalysed its growth. They focused on creating a product so seamless and efficient that users naturally preferred it over other options.

For both Slack and Zoom the product experience became the primary driver of their growth, showcasing how a well-designed product can organically attract, retain, and expand its user base.

But Slack and Zoom aren’t the only success stories from a product led approach. With 89% of consumers valuing ease, companies are making products easy to use and easy to start with. Self-help features are key, as most B2B buyers like to learn by themselves through apps rather than talking to salespeople.

Many startups and companies have experienced similar success by sticking to the core principles of the product led approach…

1. Focus on User Experience: In a PLG model, the user experience is the primary driver of growth. Simply put, the product needs to be easy enough for users to be able to explore and use the product independently.
2. Data-Driven Decision Making: Data is essential in a product led approach. Companies need to track and analyse user behaviour to optimise products.
3. Continuous Iteration: Product led growth is an iterative process, therefore we must be continuously experimenting to optimise the product experience with evolving demands.

Steps to Adopt a Product-Led Approach in AI Development

For AI products, this approach brings lends itself to helping users start using products faster, cutting down sales time, and gathering data for better products. By focusing on the product and using AI product analytics, companies can tackle the rising costs of getting new customers which have only continued to go up year after year.

The core principles of the product led approach can be boiled down to 4 simple steps:

1. Initial setup and integration:The first step is defining end user success and aligning it with your vision for the product.End-user success is made up of three elements:Start by making sure AI works smoothly in your product and make sure it's intuitive enough to use and navigate through user testing. Also, set up AI analytics capabilities to find bottlenecks and learn what users like early on by integrating platforms like Phospho for seamless AI product analytics.
   1. Understand your value.
   2. Communicate the perceived value of your product.
   3. Deliver on what you promise.
2. Building a data-driven culture:Having a data-driven culture is key. Companies that use data well make 83% more revenue and 66% more profit. Encourage teams to make choices based on what users say and do. This way, decisions are quicker and stronger. This is where utilising data from Phospho to inform product decisions are key.
3. Developing self-service features:Self-service features let users see value fast - focus on important numbers like how many users move from free to paid plans. Use AI analytics tools like Phospho to monitor usage across your LLM app to make informed product decisions to improve engagement and satisfaction. For example, implementing in-product messaging and tutorials based on user data.
4. Iterative development and feedback loops:This comes as no surprise - regularly update the product based on analytics and user feedback. While running your experiments, document everything to help you understand what happened - like whether users found what they're looking for faster or not.Use A/B testing to validate changes and enhancements. Through automated A/B testing, Phospho can swiftly dictate the optimal versions of your product that resonate most with users and significantly shortens the feedback loop for development cycles.Experiments alone won’t give you all the answers. But when combined with data, you will gain meaningful insight into how to turn the casual user into lifetime champions of your product.

It was always challenging to accomplish effective data gathering and insights extraction from AI products with traditional tools. But today, AI product analytics tools like Phospho help teams centralize the monitoring, learning, and evaluating of text data users share in LLM apps.

The best platforms collect feedback in app, which makes it easier for customers to share their ideas and requests within the context of their product experience. Using Phospho, teams can then view this data holistically and identify any common themes or trends to inform decisions for rapid iteration based on real time user data.

Using a product-led approach, AI startups can turn users into fans, bringing in new customers on their own. This has worked well for companies like Notion who listened to their users and added features they genuinely wanted in order to grow its customer base.

Key metrics to track:

As we’ve mentioned before, one of the key differences between product-led growth and sales-led growth is the way all teams leverage the product to meet their goals.

That’s why the metrics outlined below should not be siloed. They should be reported on and affected by cross-functional teams who can leverage the data to make more informed decisions and enact coordinated changes across your business.

Those metrics are:

1. User engagement: This ****refers to the degree to which users interact with and are actively involved with your product.
2. Active users: Active users log in, interact, or perform actions within your product over a specified period, typically daily, weekly, or monthly. For example, active users on a social media platform might be counted as those who log in and engage (like, comment, share) at least once a month or more frequently.
3. Customer acquisition cost: CAC measures the cost incurred by a company to acquire a new customer, usually by dividing your aggregate acquisition costs for a bucket of users by the number of accounts created.

With AI’s availability to find and analyse data it’s a no brainer to incorporate its capabilities as soon as you can. By looking at the above metrics with AI, we can make better decisions to improve our products and ultimately our users’ experience with our LLM apps.

How Phospho facilitates AI product analytics:

Phospho is an open source text analytics platform that enables AI startups building LLM apps to accelerate product adoption and enhance user engagement through:

• Real-time monitoring of user interactions lets you track and log user inputs to identify issues or trends as well as continuously fine tune the performance of your LLM app.
• Automated insights extraction and KPI detection so you can create your own KPIs and custom criterias to ‘flag’ for, and you can label if it was a successful or unsuccessful interaction.
• Continuous evaluation and iteration support. You can use our automatic evaluation pipeline that runs continuously to keep improving your AI model’s performance.

With Phospho's AI for product analytics, companies can speed up making changes with automated A/B testing and real-time data analysis. This way, they cut down on guessing when improving features and staying in line with what users want and the market needs.

How the future looks by combining AI and a Product led approach

As we go forward with AI, combining product-led strategies and AI is a strong move. Product-led growth and AI work together to boost innovation in tech. AI helps make better user experiences and improve products with data to ultimately drive lower acquisition costs and let the product sell itself.

But the key takeaway here is that the future of business is clearly product led. No matter your company's primary go to market strategy, adopting a product led approach will be pivotal for both your customer experiences and keeping a competitive edge in the market.

In order to do so, AI startups will need to leverage AI product analytics. If you’re creating an LLM app and want to gain untapped insights from your text data, sign up here!