What if you could have your own personal data analyst that thinks like your business, and is available 24/7? We are excited to announce that today this is possible for every organization with the launch of Numbers Station Cloud.
This product release marks a significant milestone for our company and customers. Our talented team, bolstered by the expertise of ML PhDs from Stanford and the University of Washington, in close partnership with our enterprise customers, has worked tirelessly to bring this vision to life.
With Numbers Station Cloud, we bring our enterprise product to users from businesses of all sizes to discover and verify business impacting hypotheses using data through a trusted and engaging interface. We are excited to open up our wait list to users for early access to Numbers Station Cloud today.
The All Too Familiar “Data Problem”
Aspiring to be data-driven, businesses are sitting on a goldmine of data, yet up to a staggering 73% of companies struggle to harness it to enhance business outcomes effectively. This likely comes as no surprise to those who are working first hand on answering business questions with data.
The core issue lies in the cumbersome process business users face when attempting to leverage their data. From the initial hypothesis generation during investigative phases to the painstaking wait for data teams to respond to queries, the journey is full of inefficiencies. This not only burdens the data teams, overwhelmed with requests, but also stifles the ability of businesses and stakeholders to derive value from their data.
Imagine a transformative scenario. What if every business user had an expert data analyst at their fingertips, ready to provide immediate data-driven answers? This would unlock unprecedented value from company data, fostering a robust data culture and significantly increasing the company’s overall value. This is not just a dream—it’s the future we’re creating with Numbers Station.
With LLMs gaining popularity, some AI solutions attempt quick fixes to this data problem that is decades old, but our database and ML experts with years of research know the problem is more complex than that. So, we adopted a full stack approach to address this problem at multiple levels of the analytics stack. After working closely with our early enterprise customers, we are tackling the barriers to truly apply this technology to business problems.
Introducing Numbers Station: Your Data Analyst in the Cloud
Numbers Station is a full stack platform designed to empower business users to discover relevant business-impacting hypotheses through data, and verify them. Through a natural and engaging interface, business users and data analysts are able to have a contextual conversation until they reach the answer to their specific question(s). Our new product introduces key capabilities leveraging LLMs with the human-in-the loop that redefines how you interact with data.
- Contextual Conversations: ChatGPT from OpenAI showed that a large segment of users enjoy acquiring information through a natural chat interface. We realized from our early users that it is no different in the realm of analytics. Specifically, business users who understand the metrics which drive their businesses but do not have a deep knowledge of the underlying data model really enjoy the chat interface. Numbers Station engages users in its natural interface, retains the context of prior questions while responding, and synthesizes the knowledge about the data and the business metrics. The example below illustrates these conversational capabilities.
2. Transparency Built In: AI-generated responses which will likely impact critical business decisions must be accurate and trusted. We ensure accuracy and users’ trust of responses in two ways.
First, our product ensures we leverage curated metrics (and definitions) to answer a user’s question before dynamically generating new metrics. When our AI model generates new metrics, they are flagged accordingly.
Second, we clearly explain how the response was derived to the user so they are able to evaluate correctness easily. Below you can see the explanation of the answer as well as the metric and expression for CLV defined to maintain transparency for the user.
3. Imprecision and Intent: With computing languages such as SQL there is no room for ambiguity as to what the result should be. However, when it comes to natural interfaces such as chat, humans tend to be imprecise and vague in specifying their intent. For example, a user may just say ROI and not specify how to measure the investment cost and more importantly how to measure the return. Numbers Station handles such vagueness by learning from other users as well as by asking users for relevant clarification. This capability enables users to continue their natural behavior rather than trying to adapt to the quirks of the chat interface.
4. Continuous Evolution of Semantics: The data environments, metrics, and user patterns continue to evolve. Business metrics evolve, existing datasets change and new datasets are added continuously. Numbers Station recognizes and adapts to these changes by updating the semantic catalog and learning from users’ questions and feedback. Thus, the semantics auto-created by Numbers Station evolve with the data environment and usage.
Configure Numbers Station on Your Data with Ease
Deploying Numbers Station on your database can be done in a handful of straight-forward steps.
1. Connect to a data source and set up a scratch database for Numbers Station. We support most common database engines such as Snowflake, Databricks, Redshift, BigQuery and Azure SQL.
2. Optionally, users add existing dashboards from products like Tableau to enrich the knowledge that Numbers Station auto-creates about the dataset.
3. Numbers Station generates a set of sample questions and responses. User(s) review them for accuracy and provide feedback.
4. Users can then start conversations with their data to verify relevant hypotheses.
At Numbers Station we’ve built our platform with security as a top priority, and we take it seriously. We do not store any of our customer data and connect directly to their data warehouses to run queries, and we are SOC-2 Type II certified.
Join the Waitlist for Numbers Station Cloud
Our journey has been driven by a passion for innovation and a commitment to our customers, and now you don’t have to be a F500 company to leverage it for your business. We’re excited to offer users the opportunity to join the waitlist for the Numbers Station Cloud. We welcome you to join us in this exciting new chapter and experience data analytics redefined.
Sign up here to join our early access waitlist or send an email to info@numbersstation.ai with any other questions.