Taking the Pain Out of Buying and Selling Data
We’re well into big data’s second decade, and we’ve made a ton of progress on many fronts. We have cloud-based systems with infinite storage capacity, sophisticated machine learning software that improves by the month, and powerful clusters turbo-charged with GPUs. Increasingly, what differentiates big data outcomes is the quantity and quality of the data we use, which typically means going offsite for additional sources. So why does buying and selling data have to be so hard?
That’s the question that Narrative founder and CEO Nick Jordan found himself asking after going through the rigmarole of the data buying process with a company that bought and sold millions of dollars’ worth of data every year.
“I actually looked for a solution to my problem, because it was a giant pain in the ass on both sides, frankly,” Jordan says. “What I found at the time was a bunch of data brokers, a bunch of people who would sit between buyers and sellers and say ‘Don’t worry, well make it easy.’”
But instead of improving life for buyers and sellers of data, these data brokers mostly were out for themselves, Jordan says. The brokers marked up the cost of data, created opacity between the buy and sell sides, and generally did their best to put themselves in the driver’s seat as the digital middlemen – to the detriment of the actual buyers and the sellers.
“When somebody can basically sit in the middle and say, hey this stuff is so hard, we can make a business entirely out of almost arbitraging the underlying asset–to me, that just shows what an immature and inefficient market it was,” Jordan says.
It also presented a business opportunity, which Jordan is hoping to fill with Narrative, the data streaming platform that he founded five years ago. Like the data brokers, Narrative sits between the buyers and the sellers of data. But unlike typical data brokers, Narrative’s goal is to build transparency, automation, and trust on both sides of the data equation.
“The idea is to build a system that takes all the manual and inefficient bits of buying and selling data and make them more efficient through automation, standardization, and normalization,” Jordan says.
Narrative’s SaaS-based application provides a platform to connect buyers and sellers. On the buy side, it helps companies acquire and integrate second- and third-party data, typically for the purpose of AI or analytics. On the sell side, companies that license Narrative’s software have a mechanism for reaching multiple buyers in an orderly and streamlined fashion.
There’s a lot of work that goes into buying and using, on both sides of the equation, according to Jordan. There are all the usual questions about the format that the data takes (CSV, Parquet, JSON, etc.), the units of measurement (Imperial or metric, GPS or UTM) and the frequency at which it will be refreshed (monthly, daily, hourly, etc.).
Once data scientists or analysts have studied a sample of the outside data and decided that it will work for their particular activity, then data engineers are called in to build the ETL pipelines to move the data, which can often take months.
On top of the logistical questions, there are legalities that must be taken into account. Buyers and sellers both must take measures to assure that they’re not violating regulations for their particular geography. Finance teams typically gets involved to obtain usage data and make the payments. And if anything changes to the data or the contract, all the engineers, analysts, data scientists, lawyers, and finance folks get to drop whatever they’re doing and revisit the matter.
Anybody who says that buying data is easy “are clearly people who have never bought data,” Jordan says. “It takes a village to buy data, so the goal is to give tools to all of those personas to make all of their jobs easier, so it’s not the cluster that it is today.”
Companies that buy data through Narrative can build data pipelines that combine datasets from multiple sources into a single, unique stream of data, which can be pushed to a data lake, data warehouse, or specific application. The company has a range of sellers on its platform today, offering many types of consumer data, including demographic data, geo-location data, behavioral data, and device data, among others.
The New York City-based company is an AWS shop, and leverages a host of big data technologies, including Kinesis, Dynamo, and Spark, to facilitate the movement of large amounts of data. Besides the big data tech, the real advantage of Narrative is the taming of all the other questions, including the management of the relationship with data providers.
One Narrative customer is a hotel chain that needs to know the wine preferences of its clients, so that they can find their preferred variety waiting when they get to their room, Jordan says. The company currently has around 40 data partners selling data, and many more clients buying data, in some cases into the millions of dollars.
Narrative recently added the credit bureau TransUnion to the sell-side of the equation. TransUnion has a long history of providing high-quality data about nearly every American consumer, and that data is now on tap in the Narrative platform, which is targeted at Fortune 2000 firms.
Narrative, of course, isn’t the only company striving to simplify the data buying and selling process. Data marketplaces are popping up all over the place, which is a good thing. The more standardization that comes to this market, the better off we will all be.
August 4, 2020
- cnvrg.io AI OS Delivers Accelerated ML Workloads with Support of NVIDIA A100 Multi-Instance GPU
- Domo Enhances its COVID-19 Global Tracker with Google and Apple Mobility Trends
- Quantum ActiveScale Software Verified as Veeam Ready Object Solution
- Striim Expands Cloud Support with the Release of Version 3.10.1
- Azure NetApp Files Now Available to Government Agencies in Microsoft Azure
- Duality Technologies, NumFOCUS to Develop Privacy-Preserving Analysis Platform
- KDD 2020 Announces Special Diversity & Inclusion Track
- Micro Focus Announces the General Availability of ArcSight 2020
- Baffle Unveils New Tableau Integration to Protect Data Analytics Pipeline
- Cyxtera Launches AI/ML Compute-as-a-Service Powered by NVIDIA DGX A100 Systems
August 3, 2020
- Wejo Announces £10M in New Funds to Support Next Stage of Growth
- Kyber Data Science Announces $10M Series A Financing Round
July 31, 2020
- Quantida Launches Complete Data Management and Analytics Solution
- New Tool to Guide Decisions on Social Distancing Uses Hospital Data and Emphasizes Protecting the Vulnerable
- Acrisure Acquires Tulco’s AI Insurance Business
- Talend Named a Leader in Gartner Magic Quadrant for Data Quality Solutions
- Hasura Releases Hasura Cloud
July 30, 2020
- Qumulo’s Cloud File Offerings Now Available in AWS Marketplace for AWS GovCloud (US)
- Entel Chooses OmniSci to Improve Customer Satisfaction
- USAF Selects Tamr to Accelerate Aircraft Configuration Evaluations with Data Mastering Solutions
Most Read Features
- Big Data File Formats Demystified
- How to Build a Better Machine Learning Pipeline
- Is Python Strangling R to Death?
- How COVID-19 Is Impacting the Market for Data Jobs
- Is Hadoop Officially Dead?
- To Centralize or Not to Centralize Your Data–That Is the Question
- What’s the Difference Between AI, ML, Deep Learning, and Active Learning?
- Understanding Your Options for Stream Processing Frameworks
- Tracking the Spread of Coronavirus with Graph Databases
- Hacking AI: Exposing Vulnerabilities in Machine Learning
- More Features…
Most Read News In Brief
- Researchers Explore Link Between American Individualism and Poor COVID-19 Response
- Left for Dead, R Surges Again
- Data Prep Still Dominates Data Scientists’ Time, Survey Finds
- HPE Acquires MapR
- Why Gartner Dropped Big Data Off the Hype Curve
- Global DataSphere to Hit 175 Zettabytes by 2025, IDC Says
- Bitnine Looks to Scale PostgreSQL
- War Unfolding for Control of Elasticsearch
- Spark 3.0 Brings Big SQL Speed-Up, Better Python Hooks
- IBM Brings Back a Netezza, Attacks Yellowbrick
- More News In Brief…
Most Read This Just In
- UBS Launches Big Data Shareholder Activism Tool
- FortressIQ Launches Adaptive Computer Vision-Based Firewall for Data Privacy
- KNIME Analytics Platform 4.2 is Now Available
- Hazelcast, Sorint Expand Partnership to Address In-Memory Computing Adoption
- Privacera Raises $13.5M in Series A Funding
- Syniti Acquires Virtyx Technologies
- BP Invests $5M in Geospatial Analytics Software Company Satelytics
- Qubole Expands Integration with Informatica’s AI-driven Data Engineering Integration
- TileDB Closes $15M Series A to Expand its First Universal Data Engine
- MariaDB Platform X5 Adds New Distributed SQL
- More This Just In…