t is reminiscent of the Tech Boom in the 1990's. New tech startups were popping up every day and every major corporation was figuring out their IT strategy. When the boom was over, it resulted in a tech graveyard of companies that were not able to adapt to the changing times.
On the buyers end, the tech crash resulted in a number of failed multi-million dollar IT projects that crushed careers. The key to success was the ability to turn new technology capabilities into new and unprecedented business value.
There are many parallels between the tech boom of the 1990's and the current AI race and, like the 1990's, there will be winners and losers in this AI race to success.
We At WatrHub Inc., are an AI-powered Digital Intelligence company uncovering market insights for water infrastructure. In just a few years we have amassed the world’s largest data warehouse in the U.S. water and wastewater utilities market. We enable our clients to answer their burning questions on their markets and sales opportunities, surfacing new insights that were not possible to get before.
Most of the largest players in our industry use WatrHub to bring their sales teams into the era of analytics and precision sales. We’ve not only built a powerful data engine, but figured out how to embed the data into our clients’ workflows for them to achieve compelling results, like 300% increase in their sales pipeline conversions.
Through this journey we have learned a lot about the realities of building a business that commercializes AI technology.
Below are 5 unique views we hold on building disruptive business models based on AI.
1. You need deep Domain Expertise
The challenge with data is that unless it’s actionable and actually actioned on, it’s useless. You can spend millions of dollars on a Data Analytics project but if it doesn’t generate 3 to 5 times the ROI that most executives look for, then it’s meaningless. And the only way to really fine-tune the data and understand what’s valuable and actionable and what’s not is to have deep domain expertise into customers problems.
In WatrHub’s case, we are aggregating data from 70,000+ entities in our market, hundreds of government data sources, and hundreds of millions of public documents. There are enough data points to keep us busy for a lifetime of analyzing and ingesting this data. No standards exist for this type of data reporting and each state and each jurisdiction does it differently. However, not all the data out there actually creates value for clients, or is represented in ways that our clients can interpret. How we bridge that gap is by having deep expertise into our client’s business and markets.
2. You need the ability to see things End to End
On one end of the spectrum you have the critical insights that users are looking for. On the other end of the spectrum you have the data and tools available in order to generate the insights. Data is comparable to the role of raw materials and the insights that are generated are the final product. While you are defining the insights to be developed, it is also important to simultaneously be thinking about the data that is available to generate those insights.
In WatrHub’s case, we have to understand the customer problems, what is possible from a data and AI perspective, what’s the best way to integrate into user workflows, and bring it all together into a winning solution for our clients. If we are not thinking about all things at the same time, we miss key pieces of the puzzle in solving our clients’ business needs. Specifically, if we overcommit on ‘understanding the customer problems’ — we will be uncovering lots of issues that can’t actually be solved by data that’s available in a feasible way. If we focused too much time on building the data engine, we spend a lot of resources capturing data that doesn’t create actual value for our clients.
3. You need to build Multi-Disciplinary teams to succeed.
Generally, industry experts are not going to be very data-savvy since they have been operating for decades without these new data capabilities, and the tech world is constantly changing. However, the experts will have a deep understanding of the problem space and the existing workflows. Different skills are required in sourcing data, managing data, and generating insights. The data scientists will bias towards building cool tools and immersing themselves in the data, and it can be difficult for them to step back and see the customer’s needs. Having a multi-disciplinary team bringing different expertise ensures every base is being covered.
WatrHub’s team includes a unique blend of water experts, client success managers, data analysts, software engineers, product managers, and AI engineers. Each bring a different perspective and skill, and they are all necessary for any of our products to succeed in the market.
4. AI-powered Insights have to (initially) match Expert Instincts.
One of the challenges with AI and most Data Analytics projects is that by definition, they are disrupting business processes and workflows that have been entrenched for many years and working at a certain, predictable speed and accuracy. Users that are comfortable with the previous process will naturally push back at change that creates risks and uncertainties. They will find holes in the data because, by nature of AI, it is not going to be perfect or accurate in the beginning. Also, these AI-powered insights have to generate conclusions that the users can make sense of. The reality is that the experts are likely to be right quite often, early on in your Data Journey. So, initially the goal is not to actually get accurate results and prove the experts wrong. The goal is to get the experts on-board with using data and offering their expertise to use the data in their workflow and decision-making.
In WatrHub’s case, we are disrupting the sales process in the water infrastructure market, which has been set in its way for decades. We have learned that you cannot possibly take the expert out of the equation. Their instincts are invaluable and AI is far from being able to replicate or replace it. So, we are simply an accelerator for them. Our job is to extract critical market insights that our users do not have the tools or resources to do, so that our users can do their job more effectively.
5. AI-powered Insights and Human Judgement will co-exist for years.
We took on a number of R&D projects to ‘fully-automate’ certain functions and what we have found is that based on available technology today, we are not at a point where we can replace human judgement for critical business processes. It is one thing to build algorithms to recommend your next movie on Netflix or your next purchase on Amazon. Those predictions will not require human intervention in every case. The impact of a 10% error rate is inconsequential.
However, for most critical business processes, a 10% error rate is fatal. WatrHub is essentially delivering insights, predictions, and recommendations on where client’s should focus their sales & marketing efforts. The cost of sales is high in the water sector and our clients are spending anywhere from tens of millions to hundreds of millions of dollars on sales every year. Therefore, the cost of inaccurate or irrelevant data is high, which is why we have built a team of analysts to curate the data and make sure it is as actionable as possible before it goes to a client. And we have built feedback loops to understand what data is actionable and is not actionable so we can refine.
The Journey To Success In This AI Race
Like many others, we are still in the nascent stages of the AI race, developing and growing with the aspirations to be a winner in this AI race. The 5 views that I have developed over the years leading WatrHub provides a great starting point into your journey of the AI race. If you are already in this exciting AI race, I hope the 5 viewpoints were insightful. We welcome the opportunity to exchange notes with you. Because the world of AI in business is still relatively small it allows us to learn and succeed together in this exciting journey. Let’s build an AI future!
What tips have you learned on your AI journey? Please share by commenting to this blog or emailing us at email@example.com!
WatrHub Inc, a Digital Intelligence Company, is pioneering data sets inspired by the needs of our clients, and we are eager to find new ways to shed light on the water industry.