“We’ve accumulated a large number of records and duplicates have always been an issue. This is something we needed to commit to in order to ensure the future success of the business.”
JW Player hits play for a billion video users with a view to customer success
Russ Passig, Senior Principal Engineer, Business Systems
Free 14-Day Trial
No Setup or Customization Required.
As the world’s largest network-independent platform for video delivery and intelligence, JW Player is plugged in to the power of video viewing for consumers and businesses. The company, headquartered in New York, markets key video innovations to leading organizations like USA Today, The Weather Channel, CNBC, Fast Company, FOX, Viacom, Univision, AccuWeather, Eurosport, and Fandom. For more information, visit: www.jwplayer.com.
JW Player has a global footprint of over 2.7 billion unique monthly devices and provides media professionals with powerful and flexible technology to deliver video, grow their audience, and monetize with ads. The cataclysmic shift to remote work and quarantine in 2020 has only intensified interest in video and generated increased business for JW Player. According to Deloitte, 80 percent of US consumers are now paying for streaming video services, up from 73 percent before the Covid-19 pandemic. Watch parties and streaming releases became the new weekend “out”, leading to 78 percent of people saying they watch videos online each week, and 55 percent watch them daily. And 72 percent of customers prefer learning about a product or service through a video, says Social Media Week in its Video Marketing Statistics 2020.
JW Player is focused on meeting the increasing demands of customers looking for new ways to leverage video. “Video is about shared experiences,” said Dave Otten, co-founder and CEO, in JW Player’s “The State of Video” presentation. He noted that live streaming increased by 400 percent in 2020 over 2019, with JW seeing an explosion in growth in the categories of news, fitness, community, and sports. The good news for JW Player is that subscriber growth is surging outside of the Big Five of Hulu, Amazon, Netflix, Disney Plus, and Apple TV, growing by 51 percent in one year. One of the takeaways that JW Player recognizes is that the data they manage and access to communicate with customers has to be clean and accurate.
”72 percent of customers prefer learning about a product or service through a video.
Video Marketing Statistics 2020
Making Customer Connections through Clean Data
“Clean data is critical,” said Russ Passig, Senior Principal Engineer, Business Systems at JW Player. He and Alex Jean-Louis, Revenue Operations Coordinator, are focused on ensuring that the contacts and leads in their Salesforce database are exactly aligned to the actual customers and prospects that they are communicating with.
It was JW Player that developed the first embeddable web player way back in 2004. Since then, the company has set a standard for innovation that keeps its customers hitting play on accelerating their business. “A key differentiator for us has been our customer support and a 360-degree approach,” said Russ.
One example of that attention is their customer Notts County FC, the oldest professional football club in the world. The multimedia editor at Notts County acknowledged not only that “the whole setup was very straightforward” but that they had “great support from JW Player and InPlayer. Nothing ever seemed too much of a problem.” That facilitation speaks to the ability of JW Player to understand who their customers are and what their needs are, which is based in part on the intelligence contained in their databases.
Increasingly, however, Russ and Alex were finding that it was an uphill battle to achieve appropriate data governance. “We had to cobble together our own approach to maintaining data integrity,” commented Russ. Every day, Russ and Alex were working in their Salesforce platform and the tools they had to clean data were not adequately addressing their needs.
As often happens with successful businesses, the needs of the business must take priority. As a result, the complexity of the data and the amount of traffic began to overwhelm the business’ ability to create an easily manageable CRM system.
”We had to cobble together our own approach to maintaining data integrity.
Russ Passig
Getting Data Quality to Align to the Business
When you consider that JW Player has over a billion unique users, it’s not hard to imagine the extent of the data in their Salesforce database. “We’ve accumulated a large number of records and duplicates have always been an issue,” said Russ. He realized that they had to get management of data in alignment with the goals of the business. “This is something we needed to commit to in order to ensure the future success of the business,” Russ added.
The team had used RingLead in the past, as well as Cloudingo, which they were not happy with. “The main objective was to strip out all the junk data,” said Russ. “We came into the project with the intent to be successful in cleaning our contact data, and that extends to all of our leads.” As they pursued providers that could facilitate cleaning up contact and lead data, their records – and their duplicates – continued to grow.
In order to get the most out of the Salesforce platform, users need to ensure clean and accurate data across contacts, leads, accounts, and any custom objects. It’s a time-consuming task that was weighing on Russ’ and Alex’s ability to be as efficient as they wanted. Russ relates that the additional tools they looked at just were not robust in the ability to deduplicate their records.
The Data Team Hits Pause
Even though Salesforce can notify users that they are about to create a duplicate record and even block them from doing so, there are many other ways that duplicates can enter the system. Importing contacts from a spreadsheet and migrating contacts to Salesforce from a previous CRM are just two examples. And Salesforce isn’t the only one. Many third-party duplicate checking apps are struggling with the inability to catch all dupes. Both Salesforce and most third-party apps require the user to import the contacts first, and then it would run the duplicate check. However, such an approach puts the user – in this case, Russ and Alex – on the defensive … always on the lookout for new duplicates.
“When we started this project, we were north of one million with records,” commented Alex. They were dealing with contacts and leads mostly with accounts falling under hierarchies. They wanted to target the best way to communicate with their customers, not waste time making sure they were reaching the right customer.
”When we started this project, we were north of one million with records.
Alex Jean-Louis
Pressing Play on Data Cleaning
“We were aware we needed a cost-effective substitute to handle what we do,” explained Russ. But so far, duplicates in the data were continuing to stream through. At this point, JW Player needed a rapid result. “DataGroomr was vetted against a series of other tools,” said Russ. He had identified DataGroomr from the AppExchange and reached out to co-founders Steve Pogrebivsky and Ben Novoselsky.
“I talked to Steve and Ben. I really liked the direct channels of communication with the folks at DataGroomr. Together, they explored data governance validity on JW Player’s database under different scenarios. Russ said he was able to retrain the deduping model according to the needs of their organization. The reason that configurability was possible was due to DataGroomr’s machine learning foundation – a feature that Russ had not come across in his search. That was the defining differentiator for him.
Machine Learning Takes Center Stage for JW Player’s Database
The interest in video is going to grow exponentially, according to Cisco Annual Internet Report, 2018-2023, with today’s bandwidth needs predicted to explode over the next couple of years. For JW Player, that meant that they couldn’t be distracted by the specter of duplicates in data that was their only way to reach that exploding audience. DataGroomr is the only Salesforce deduplication solution to use machine learning to analyze data and identify duplicates. Russ and Alex were finding that the algorithms were much more effective and accurate than using standard filters.
Every company’s dataset is unique and has its own challenges when it comes to deduplication. That was especially true of JW Player, which was dealing with 2.5 million records. With the other rules-based tools on the AppExchange, every time a duplicate is identified, the Salesforce administrator needs to create an additional rule to keep it from recurring. But that constant rule manipulation isn’t necessary with DataGroomr, and it’s the feature that Russ found most valuable.
DataGroomr comes with a pre-trained machine learning model that automatically refines data as you use the system. In addition, a model for the matching rule can be trained from scratch on the data residing in your Salesforce organization. That distinguishing feature would help JW Player to move quickly to remove the duplicates in its database.
”We could leverage the training we were given on the tool and adjust it to work best for us.
Alex Jean-Louis
Results of Clean Data & Clear Views to Customer Data
“We could leverage the training that we were given on the tool and adjust it to work best for us. Plus, access to the tutorials in the DataGroomr Knowledge Base made the learning curve quick,” added Alex. He noted that the merge capability was also a snap. Something that was important for a database that has gone from doing 6.7k merges to 10k merges in much less time than previously.
Before DataGroomr, one person had to be deployed to clean up the database. It was a task that cut into the efficiency the company had built over their more than 10 years in business. Alex added that using DataGroomr has been “smooth, clean, and straightforward. We have achieved a significant dent in the dataset.” Whereas they started with 2.5 million records to assess, they have deduplicated 600,000 records.
With DataGroomr in place on its Salesforce database, JW Player can identify records natively and rely on the tool to implement the rules they set and then to apply it consistently across all records.
Russ noted that they can retrain the matching model periodically, and that’s part of the beauty of machine learning.
“What I liked about it was that the UI is easy to use and get used to,” said Alex. “The dashboard was really important.” Facility with the dashboard was crucial to the operations team because of the sheer volume of records they were dealing with. The access to Customer Service at DataGroomr was something JW Player, with its own focus on direct customer communication, could really appreciate.
Russ said the most important thing for him has been the machine learning aspect and the cost effectiveness. Alex likes the UI and ease of use. Russ said a big differentiator for him has also been that he can communicate directly with Steve and Ben, who he said are ‘disruptors’ in the tech field because of their open collaboration. “They’re knocking it out of the park,” Russ said of DataGroomr.
”They’re knocking it out of the park!
Russ Passig
