How do you help big brands make decisions based on social data?
By using Elasticsearch to do real-time analysis of 200 million conversations across the social web each day
Case Study Highlights
Use social sentiment to understand your customers
- Real-time search allows you to measure sentiment change immediately
- Analyze metadata for richer, more complete customer insights
- Tap into millions of real-time data points to make actionable decisions
Create a real-time decision making platform
- Understand social data from Twitter, Facebook, blogs, and forums
- Networked Insights quickly scaled from indexing six months of live data to thirteen months
- Use Elasticsearch to easily search Apache Hadoop data
Analyze Social Data to Make the Right Marketing Decisions
When it comes to social data there are many tools out there that help marketers measure the effectiveness of their campaigns, but there is only one solution – that with the help of Elasticsearch – is changing the way marketers make data-driven decisions. That solution is Networked Insights, a 7-year old Chicago-based company that has been leveraging real-time consumer data to transform the way major brands like Samsung, MillerCoors, Revlon, and Procter & Gamble solve their marketing problems. Networked Insights accomplishes this through collecting consumer data from all across the social web – from forums like Reddit to blogs like Wordpress and blog commenting platforms, to other social media networks like Twitter, Facebook, and Tumblr. Using the millions of data points that come in daily from these data sources, Networked Insights is able to segment the data into key target audiences that in turn provide actionable insights for marketers. The result is smarter marketing decisions.
If we had to reindex all of our data it would take us three months in other technologies. It took us a day and a half in Elasticsearch.
Unparalleled Real-Time Analysis with Elasticsearch
Networked Insights discovered Elasticsearch more than two years ago when they were looking for a technology that could jointly search and index data in real-time. Their previous solution required scheduled downtimes that slowed or prevented searches from occurring. More importantly, additional critical resources were always required to re-index all the data and made scaling with their ever-growing data a major business problem. Elasticsearch was chosen to replace their previous solution because of it being the only real-time search engine available. As a result, Networked Insights was able to not only scale with their growing data, but also increase their window of live searchable data from six months to thirteen. Real-time access to that volume of historical data is unparalleled to any other analytics provider.
Use Meta Data to Better Understand Your Customers
In addition to improving the speed and scale of their marketing decisions platform, SocialSense, Networked Insights was able to leverage Elasticsearch to facet a richer real-time dataset using the implicit and explicit data embedded within social data. The richer explicit data allows SocialSense users to search and segment audiences based upon metadata like geo-location, preferred mobile device, or demographics. The access to more implicit data enables Networked Insights to classify all of their data it fetches from across the social web into searchable data like purchase intent, audience interest affinities (through tracking thousands of celebrities, brands, TV shows, etc.), and sentiment.
Deliver Actionable Insights
These enhanced capabilities are leveraged on a daily basis for all of Networked Insights' clients to inform a wide-range of marketing decisions, but below are three examples of common business problems Elasticsearch's search and analytics engine has empowered:
- Brand Positioning & Competitive Intelligence: Elasticsearch platform's ability to facet and search big data in real-time across multiple social media channels has provided rich audience insights to brands when they are re-evaluating their brand positioning or conducting breakthrough competitive analysis across brands. As a result, Networked Insights has de-risked and accelerated many challenging repositioning for clients.
- Product Launch: Elasticsearch's platform has helped enable Networked Insights to classify and segment target audiences in real-time based upon explicit and implicit consumer data. This has helped brands make strategic decisions on the timing and location of specific product releases, as well as what creative would resonate most with their target audiences'. As a result, Networked Insights has helped increase awareness and adoption of client's products and improved sales numbers beyond their original predictions.
- Crisis Management: With Elasticsearch's ability to consume, index, and search massive quantities of data in real-time, Networked Insights has the ability to zero in on specific minute-by-minute conversations to empower brands to evaluate negative consumer reactions while in-flight to a product launch or specific advertisement. As a result, Networked Insights has successfully identified client crises before they have had an opportunity to escalate, and provided actionable insights into how brands should best react.
Networked Insights’ Benefits Using Elasticsearch
Solve Hadoop's “Last Mile" Problem with Elasticsearch
Networked Insights stores 13 rolling months of Tweets, Facebook posts, blogs and other social media in Hadoop, which runs alongside Elasticsearch. They use Storm to flow data into Elasticsearch to do real-time analysis of the data using their SocialSense platform.
Elasticsearch's horizontal scalability and real-time search enables Networked Insights to deliver intelligence for clients to make smart data-informed marketing decisions to protect and enhance their brands.
Simplified administration and management
Elasticsearch makes it easy for a CTO to manage the overall system and predict how much data is growing from one, easy-to-use JSON REST interface.
Analytics provide real-time insights
Networked Insights uses the analytics capabilities in Elasticsearch to monitor and analyze social conversations to learn more about their clients' customers and end-users to fine-tune product messaging and timing.