Today’s businesses have more data than ever before, easily amassing terabytes–even petabytes–of information. It’s no wonder why data is hailed as the oil of today’s digital era.
But while most organizations understand that their data has the power to enable them to make better and faster business decisions and, consequently, boost their profitability, not all are able to derive true value from the overwhelming amount of data they gather on a daily basis. The fact is, businesses often have so much data that knowing where to begin is one of–if not–the greatest challenges of working with data.
To address this problem, Cinchapi, an Atlanta-based data startup and winner of the 2017 Atlanta Startup Battle pitch competition, offers an intelligent platform that leverages machine learning to simplify data preparation, exploration and development.
SourceForge had the chance to speak with Jeff Nelson, the CEO and Founder of Atlanta-based data startup Cinchapi, to discuss the biggest obstacles today’s enterprises face when tackling data and how the Cinchapi Data Platform enables organizations to harness the full potential of their data more effectively.
Q: First and foremost, please tell us a bit more about Cinchapi. What’s your story? When was the company established and what challenges does it seek to address?
Cinchapi CEO Jeff Nelson
A: Cinchapi is on a mission to make data make sense. I founded Cinchapi in 2016 with a broader vision of radically simplifying the interaction between humans and machines.
Since our founding in 2016, Cinchapi’s core vision has been to deliver “computing without complexity”.
To that end, Cinchapi combines machine learning with human intelligence to make it easy to explore disparate data with natural language questions. Users see real-time results presented as visualizations along with descriptive text.
Cinchapi can also dynamically trigger, halt, or adjust workflow automation of connected systems based upon the data, or a human can be alerted to make a judgement call.
The platform radically reduces the need to spend valuable time doing data prep and cleanup. Instead, users can begin to ask ad hoc questions within minutes of connecting to their data sources. Moreover, machine learning proactively reveals otherwise obscured patterns, anomalies, and relationships across all of the connected data.
Q: Data is now viewed as a strategic asset and the fuel of the digital economy. Because of this, many organizations aspire to become data-driven companies. But what does it mean to be data-driven, and what does it require?
A: To truly be a data-driven company, a business must do more than just collect data – they need to become a mathematical corporation. The business must also be able to explore, analyze, and react to high-velocity data changes in real time. But conventional methods of doing this, like creating a custom data pipeline using a bunch of ill-fitting parts, are costly and time consuming.
As IoT data becomes more prevalent, the ability to keep up with rapidly evolving data becomes critical. Imagine being a Data Analyst at a trucking company, and seeing that temperatures in refrigerated trailers are fluctuating. That’s critical to resolve, because the entire load could be at risk. It’s not enough to do forensic analytics after the fact. It’s essential to be able to take action in real-time before it is too late.
Q: Businesses of all sizes are looking to capitalize on the increasing amount of data available to them, and yet not all are getting the most value from their data. What do you think are the biggest obstacles companies are facing today?
A: Cost and complexity are primarily to blame here. Conventional methods of working with disparate data tend to require an investment in data pipelining, which is not inexpensive. Then there is the need to invest in ETL tools and processes to get all of the data to share a similar structure.
Beyond that, there is the costs of the tools for analytics and visualizations, and of course someone is going to have to do all of the work. Because working with data is kind of complicated, that could mean hiring one or more Data Analysts or working with outside consultants.
Add all of that together, and you can understand why some businesses are unable or unwilling to allocate the budget needed to get started, or decide that it’s too much of a headache to deal with now.
Q: Tell us about the Cinchapi Platform. What makes it stand out over other similar solutions in the market?
A: To do what Cinchapi does as a single comprehensive solution, you would likely have to invest in series of tools from different vendors. Some tools play better with others. Certainly Tableau is moving in the right direction with the addition of Tableau Prep to their visualization tool, but that’s still not an end to end solution.
Instead of using an ETL and a data warehouse and conforming all of your data to a common format, Cinchapi instead uses machine learning to examine connected data to understand what it means. It doesn’t force any schema on the user, so feel free to mix and match any group of data sources, including real-time data.
The Cinchapi workflow is comprehensive, but it’s also easy to master. Once your data sources are connected, it’s really just three steps: Ask, See, and Act.
Ask a Question: Using common English phrases, users literally can ask questions of their data. Even better, Cinchapi learns with use, so the user can use industry jargon or company lingo in the questions. To drill deeper, simply ask a context-aware follow up question and see the analytics update in real-time.
- See Results: Speaking of those analytics, visualizations are included in Cinchapi. Based upon the data and the results in question, the platform presents recommended visualizations which best convey the meaning of the data. These are presented along with descriptive text which indicates what makes the result of interest. Don’t like the recommended visualizations? No problem, users can choose another which better suits their needs.
- Act On the Results: Analytics and visualizations are great, but if you can’t take action easily and when it is relevant, what’s the point?
Cinchapi can connect to business systems and dynamically trigger, halt, or adjust workflows based upon the data. Not comfortable letting that happen automatically? No problem. Alerts and messages can be sent to company staff via #Slack, text messaging, or similar apps, so that they can make the final call.
The takeaway is that the business can do something in the here and now – when it really matters.
Q: How is Cinchapi empowering businesses to extract more value and harness the full potential of their data more effectively?
A: By removing or mitigating the roadblocks that make data analysis cost prohibitive, Cinchapi provides a return on investment within minutes of connecting it to the data.
It’s common for Data Analysts to spend as much as 80% of their time doing the tedious data prep and cleanup normally required before they can even begin to explore data. Moreover, they often have to use ETL (Extract, Transform, and Load) tolls to get the data into a common format and then into a data warehouse.
All of that takes time and can be cost prohibitive. Beyond the warehousing and tool costs, factor in the potential need to hire additional Data Analysts as well as additional support personnel. All of that adds up.
To be clear, we’re not suggesting that you won’t need a Data Analyst to get the most out of Cinchapi. What we are saying that is that with Cinchapi, a Data Analyst can begin exploring data with ad hoc questions almost immediately. Additionally, team members with a basic understanding of data can also make use of the platform without the need to create cryptic queries, or waiting on the Data Analyst to create the queries for them.
Q: What types of clients do you currently serve? Can you share some of the best use cases for the Cinchapi platform?
A: While virtually any organization trying to get value out of their data can get real value from the platform, we are finding that a few Cinchapi use cases are really standing out from the others.
We recently did a pilot with a major logistics company in the UK. They were having difficulties in handling real-time data from their vehicles, as well as responding to issues beyond their immediate control, like heavy traffic, weather, and related events that were slowing freight.
With Cinchapi, they found that they were able to plan more efficient routes for their fleets, as well as to respond proactively to maintenance issues before they impacted their loads.
Another interesting use for Cinchapi is with municipalities seeking to establish Smart City initiatives. As you likely know, the Smart City concept uses sensors to collect data that can then be used to better allocate resources. It’s not just IoT, but a robust Smart City also uses data collected from citizens who might report a pothole or a water leak.
Cinchapi’s machine learning makes it easy to collect disparate data from different devices and sources, while also proactively looking for patterns, anomalies, and relationships which warrant further investigation. Moreover the platform’s ability to understand natural language queries as well as jargon and local lingo make it ideal to present as a front end interface to non-technical users.
Q: Looking ahead, do you see any emerging trends or technologies that will shape the future of business data discovery and analytics?
A: As much as we are proud of our natural language interface, within a few years we feel we’ll be seeing a demand for more voice recognition as an interface. While Siri and Alexa are becoming more commonplace, the challenge will be dealing with a workforce where different languages, dialects, and even accents are part of the mix.
Similarly, analytics and visualizations are what people expect now, but we can see a future where augmented reality could be driven by data. Think about a Smart City promised future where the data not only dispatches a repair crew to fix a failing pump, but which would provide instructions via an augmented reality interface that shows the worker exactly where the problem is. It could create a packing list with all the parts and tool required for the repair, and then it could guide the worker through the steps needed to effect the repair.
The value in data is always going to be in improving efficiencies, reducing losses, and to provide new opportunities for a business. Connecting and collecting data from these emerging technologies seems like a natural progression.
Q: What can customers expect from Cinchapi in the future? Are there any new developments or product offerings customers should look forward to?
A: As we prepare to launch the platform, we are looking for ways to expand the scope of what we can offer. Our engineering team is already looking at voice recognition as a potential interface.
Additionally, while we are currently focused on an English speaking target market, we are fully aware that we will need to provide a solution for other languages if we want to be seen as a globally relevant solution.
About Cinchapi
Cinchapi is an Atlanta-based startup that seeks to transform how data analysts, data scientists, and business leaders explore and work with data. Cinchapi offers its namesake codefree platform that leverages the power of machine learning for business data discovery, analytics, and automation. The platform uses a 3-step workflow–Ask, See, and Act–to make working with data more conversational, efficient, and intuitive.