USAA’s digital group product managers were tasked with measuring metrics such as performance, member acquisition, and service transactions on an hourly basis, and distributing their analysis across the group utilizing D3.js, an open-source JavaScript library for creating visualizations from data. However, USAA’s product managers needed a way to increase the speed of analysis so that they could quickly and easily identify and share key insights with executives.

Tasrail, a leading logistics company in Australia, faced a significant challenge in managing its monthly operational performance reviews with customers. The company was responsible for transporting over three million tons of goods annually across the Tasmanian region. To ensure timely delivery and adherence to key performance indicators (KPIs), Tasrail conducted monthly reviews with its customers. However, the process of writing static reports for these reviews was extremely time-consuming, taking up to 20 hours each month. This not only made the task difficult to scale but also hindered the company's ability to provide timely insights and data to its customers. Tasrail needed a solution that could reduce the time spent on reporting while making data and insights easily accessible, all without increasing headcount.

The consulting firm was engaged in expensive and time-consuming manual analysis to provide insights to their clients. They needed an approach to deliver insights instantly and more efficiently. The manual process was not only costly but also delayed the delivery of valuable research insights, impacting their ability to help clients make timely decisions.

Rhapsody, after acquiring Napster, expanded its services under the Napster brand across Europe, Latin America, and Canada. The company faced significant challenges in translating and localizing its content into 14 languages for 34 countries. The manual process of sending and receiving string files for translation via email was slow and confusing. Additionally, Rhapsody incurred high project management fees for each translation project, regardless of size. The company needed a streamlined process to control rising translation and localization costs and improve workflow efficiency.

B2B brands face the challenge of localizing digital content to deliver native brand experiences across all markets. They need a translation solution that can manage the volume of content requiring localization, including web-based content, mobile applications, and technical documentation. Additionally, the solution must integrate easily within their existing technology infrastructure. One of the world's largest enterprise business software companies was using outdated software for translation, which required a lot of manual work and was extremely inefficient. The connector to its content management system was unreliable, leading to untranslated pages on the company's websites. With the end of its existing translation software licensing agreement approaching, the company knew it was time to upgrade to a new translation technology platform that was more efficient and could scale with the business' ongoing translation and localization needs.

Kionix’s corporate website provided only English language content that was highly technical in nature. To better service its Asian-speaking customers, Kionix needed websites that were localized for these regions. There were several major areas of concern: Ease of Translation. Kionix wanted to streamline the translation process and limit the ongoing maintenance and support needed from their in-house team. Time to market. The company wanted to rapidly deploy four localized websites – Japanese, Korean, Chinese, and Simplified Chinese – within two months, as part of a larger website redesign project. Quality. High-quality translations for Kionix’s technical business-to-business content were a must. Scalable. Kionix wanted cost-effective, proven technology that would easily scale as its business grew.

Hotel Tonight, a popular mobile travel app, aimed to expand its services internationally to become the world's most-loved hotel booking app. The company faced several challenges: achieving rapid global expansion, integrating advanced technical requirements, managing limited resources, and ensuring high-quality translations. Traditional translation methods were too slow, and the app needed to support multiple currencies and localization file formats. Additionally, the app's design did not accommodate long text strings, complicating the translation process.

Early in 2014, Vivino reached the point where localization was becoming critical to the company’s success and growth. It boasts an active community of wine-lovers from all over the world and the company knew that localization would be key to further engaging them. Vivino did not have anything in place for translation, so this was an entirely new undertaking. After hearing about Smartling through a partner in the wine industry, Vivino product manager, Christos Iosifidis, took a look at the company’s offerings and agreed that Smartling had the best solution for Vivino’s needs. Christos had experience with translation companies in previous roles and was impressed with the client services Smartling provides.

How does a dynamic social network that was built over many years with a range of technologies localize the user experience to attract a global audience? The challenges included: Complex infrastructure. MeetMe knew it needed a solution that supported its file types and infrastructure: heavy use of JavaScript, secure HTTP headers, content delivery networks, IP address white and black lists, and multiple development environments. Painful process. MeetMe would have to extract all of its website content and provide it to an agency for translation into multiple languages. The translated content would then need to be accurately imported back into the source code for testing and deployment across MeetMe’s web properties and applications. Release delays. MeetMe wanted to avoid code freezes and delays as a result of the translation and QA process.

The recent launch of Le Monde in English, by the leading French news group of the same name, epitomized these challenges. This new service is an integral part of Le Monde’s goal to reach one million subscribers, while providing a French and European perspective on global events for the English-speaking world. After testing in-house translation, it was clear that Le Monde needed an external solution that would ease the pressure on their reporters and allow their new site to cope with the pressures of the 24-hour news cycle. The publication approached Lionbridge for an automated solution to their speed issue, with stringent requirements to preserve its distinctive approach to journalism and protect its reputation as France’s official newspaper against fake news. The stakes were high. The release of Le Monde in English coincided with the climax of the 2022 French presidential election: a gripping, fast-moving news story that was the archetype of Le Monde’s future English coverage. Lionbridge’s solution needed the capacity to immediately handle a high volume of sophisticated content, translated to a quality standard that was befitting of the authority on French politics.

Top global brands are dramatically expanding their video usage for marketing, support, training, and corporate communications. Unfortunately, the traditional approach to subtitling and voice-overs is slow and costly. The requirements are changing, pointing to a new approach: cycle time requirements for translation have been reduced to days or, in some cases, mere hours. Cost efficiency is crucial as budgets cannot expand to treat every video with in-studio, professional voice talent. Quality is now dependent on the type of content translated. There is no longer just one standard.

The company experienced global growth for both its enterprise and personal technology products and support services. That same growth brought new pressures to the QA/testing systems, products groups, and their supporting vendors. Volatility in workloads, workflows, and resources led to increased product defect rates, inconsistent quality, and a talent gap. Based on history, quality, and capability, Lionbridge was selected to lead a process transformation that would help the company increase quality and reduce cost on a global scale.

The insurance company needed to ensure that its customer-focused mobile app was accurate in real-world conditions and easily updateable based on customer feedback. The existing simulation software was not delivering the necessary data, prompting the company to issue an RFP. Lionbridge responded by rewriting the RFP, highlighting critical missing elements, which led to a new partnership and significant project modifications. The goal was to test the app's performance in various real-world scenarios to ensure it met customer expectations and provided accurate data for determining insurance discounts.

The translation of technical content has become increasingly complex and challenging due to the need for process efficiency and the requirement to deliver multi-lingual user assistance information across all channels. Companies must ensure that their technical communicators can handle structured creation and dynamic publication of user information globally. This necessitates the ability to deliver consistent, high-quality content in multiple languages to maximize customer experience. The traditional approach of working on complete documents is no longer sufficient, as technical communicators and translators now often work on new and modified chunks of text.

In the early stages of a clinical study, a strong start is critical. Global clinical trial planning requires early consideration of the costs, complexities, and risks surrounding communications. Language is a mission-critical element of culturally and regulatory appropriate content containing all forms of communication. Consider language translation right from the beginning, and you’ll realize a stronger trial execution—from study inception to completion.

Canon faced multiple challenges: 55 markets, reformatting the oldest legacy pages to mobile-optimized new templates, thousands of pages of content, and integration with a new digital asset management system. The migration project involved transitioning content from Tridion 11 CMS to SDL Web 8.5, updating older legacy templates to new mobile-optimized templates. The goal was to migrate content in 55 markets, containing over 300,000 pages of content, while implementing a new Digital Asset Management System. Canon needed to ensure consistency and improvement of the customer journey, site responsiveness, and effective long-term content management, all while reducing costs and post-publishing time to market.

The Thule Group faced several challenges in managing their multilingual content. With a presence in 140 countries and product and online content translated into over 30 languages, the volume of customer-facing content was growing rapidly. This led to huge backlogs of untranslated content and content gaps in critical markets. Additionally, the increasing complexity of content, including multimedia and interactive elements, added to the challenge. Thule needed to balance speed, quality, and cost while ensuring that their multilingual websites remained current and consistent across all markets.

The client wanted to make sense of their 20+ years’ worth of accumulated GTD content, which accounted for more than 100,000 documents and media files translating to 9 Terabytes of data. They needed a solution that could transcribe, summarize, conduct sentiment analysis, enable semantic search, and organize their data. This would help them deliver relevant knowledge to both current GTD practitioners and others looking for specific solutions to challenges in their lives. The client explored all major cloud vendors from the Gartner Magic Quadrant, including Amazon, Google, and IBM. However, they found these solutions to be expensive, geared towards large enterprises, and requiring third-party IT vendors for custom solutions. The client felt there was a gap in the market for companies like them with limited users.

Being in the business of educational technology catered to the end-user, it was essential for Appenza to listen to the voice of the customer to create an engaging and efficient learning app. However, for the development of the app itself, the large reservoir of the Ministry’s academic content needed to be accessible to users. The challenge could be categorized in two ways: Search & Discovery of Content: The Ministry of Education has an immense source of education material across a wide variety of subjects ranging from Literature to Maths. All of this data is maintained in a knowledge bank (KB) for students from Kindergarten to Higher Secondary (Grade 12) in PDFs and videos. While the content was quite rich and informative, it was nearly impossible to discover since none of it was structured by any theme or logic. There was no search facility to dive into the individual content as well. Added to this, the content was a mix of Arabic and English, which made organizing the content more difficult. Understanding User Feedback: It was imperative that Appenza understood the feedback from students and parents for the new and improved mobile app being created for the Ministry of Education, and to measure its performance on an ongoing basis. They, however, did not have the technical know-how to measure public sentiment across the country expressed in thousands of comments on different social media platforms and review websites. They needed an accurate and automated sentiment analysis especially made for Egyptian Arabic, along with an English model for gathering insights from the mined social media data.

For Tapad, the ability to predict the best placements and correct pricing for advertisers’ campaigns, as well as offer RTB to marketers across multiple channels, was critical to the success of its technology. To support this functionality, the company required a database that could reliably process and manage the data linked to billions of mobile display impressions and also respond to inquiries within milliseconds. The key-value store approach of NoSQL databases was ideally suited for Tapad’s needs, and the company began to investigate several well-known systems. However, Tapad found that despite offering high throughput, many products lacked the predictable response times needed to support the platform’s users. Only Aerospike stood up to the test of Tapad’s proof of concept, demonstrating the performance, reliability, and predictability required for their demanding production environment requirements.

Komli Media aims to close the significant gap in click-through rates (CTR) between search ads and display ads. Google achieves approximately a 2% CTR on search ads, while the average CTR for display ads is only 0.1%. To bridge this gap, Komli Media needs to innovate on cutting-edge technologies to process tens of thousands of requests per second, deliver analytics on terabytes of transactional data, and match users with the best possible ads for optimal return on ad spend. The challenge involves implementing a robust infrastructure capable of handling real-time bidding (RTB), ad price prediction, and big data analytics at web-scale, with redundancy to ensure 99.999% availability.

For So-net Media Networks, offering RTB functionality was critical to the success of its new ad distribution service for the Japanese online display market. A distinctive algorithm with the ability to forecast expected click-through rates (CTR) and conversion rates (CVR) was already in place. However, the company required a database that was capable of reliably processing and storing the vast sets of data at the core of the Logicad platform. Additionally, it needed to handle responses in less than 120 milliseconds (ms)—the time within which a user clicked to view a page, the request was received across the network, the profile was retrieved, Logicad’s algorithms were executed, bids were received, a winner was selected, and the ad was served. Network traversal times of 50ms left only 70 ms for server-processing and 2 to 5 database transactions per request. This meant that each database transaction had to complete as fast as possible and within 10 ms at most. With 10,000 bid requests arriving per second, the database also had to handle throughput of 50,000 transactions per second (TPS). Another requirement was the ability of the database to automatically expire millions of records per day. So-net Media Networks also wanted a database with no single point of failure, 24x7 reliability, and availability. Background daily backup capabilities were important, as was the ability for the system to scale out easily as the business grew.

MediaV faced the challenge of scaling its ad bidding and serving platform to meet the rapidly growing demands of China's online advertising market. The company needed a reliable, high-performance database solution capable of handling billions of user profiles and ensuring real-time responses within milliseconds. Additionally, MediaV required a system that could support high query per second (QPS) capacity, fast and consistent reads and writes, massive data storage, scalability, and robustness. Cross data center replication was also crucial due to MediaV's multiple data centers. After evaluating several NoSQL database management systems, MediaV found that none met all their requirements, leading them to develop their own data storage and management system based on an open-source NoSQL solution. However, they eventually discovered Aerospike and decided to evaluate it.

The SiteScout Platform, a demand-side platform (DSP) for real-time bidding and reporting, works with massive volumes of logged cookie matches and user data profiles to distribute targeted display ads with RTB functionality. To further maximize the penetration and reach of ads, SiteScout also offers a number of advanced DSP features, including auto-optimization, retargeting, and mobile traffic support. The demands of the DSP are heavy, requiring low-latency replies, high availability, and the ability to scale, as well as the ability to replicate data across multiple data centers. Early on, SiteScout recognized that a NoSQL database would be best suited for handling its large scale amounts of data. However, the first NoSQL database the company implemented failed to meet SiteScout’s performance demands. The initial NoSQL database was not a true, multi-threaded database, limiting the ability to effectively utilize the machine specification, which was a hugely limiting factor.

madvertise needed a highly scalable, fault-tolerant database solution to support real-time targeting and mobile identifier fusion functionality. The database had to handle ad hoc data with a built-in decay rate, support frequent data updates, and maintain low latency while processing tens of thousands to hundreds of thousands of queries per second. Existing database solutions failed to provide sustained or predictable throughput and lacked a thorough recovery process, leading to load balancing issues and memory restrictions.

The Federated Media Publisher Network platform powers ad delivery through a distributed architecture consisting of a main datacenter that communicates with cloud-based advertising “pods” to work with demand-side platforms, ad exchanges and ad networks for real-time bidding (RTB). Federated Media Publishing also guarantees brand safety for advertisers by looking at page-level context to ensure that ads are served on brand-safe pages, as well as providing targeting through page-level site categorization. Delivering these functions in real-time is an integral part of the company’s new focus on programmatic buying, which automates the placement process by instantaneously selecting who to serve impressions to based on data an advertiser thinks is pertinent to the campaign. Federated Media Publishing quickly determined that relational databases could not provide the ultra-fast response times required to deliver real-time services across its platform. Instead, the company began evaluating NoSQL databases for their ability to handle high volumes of data and respond in milliseconds.

Contextin’s unique algorithmic approach to campaign performance optimization works by analyzing hundreds of granular variables—including page characteristics, user engagement data, and semantics—on an impression-by-impression basis and then extrapolating its learning for each campaign within the context of the specific performance and budget parameters. This enables Contextin to assess bid price and identify the type of impression most likely to get results. With massive sets of proprietary data at the core of Contextin’s platform, the company recognized the need for a powerful NoSQL database that could manage and process vast sets of information without slowing RTB response times. To support its early production platform, Contextin integrated an open source distributed database to which the company was contributing code. However, as its business grew, Contextin began to evaluate NoSQL database options that would offer greater performance. “We need to be able to hit a throughput of about 200,000 to 300,000 queries per second with response times of under 50 milliseconds for all the processing related to each query,” Mr. Naveh explains. “This is a very high load requirement, and naturally we can’t afford to have queries take a lot of time.” Query time and availability became the stumbling blocks for many of the NoSQL databases evaluated. While many of these solutions were capable of working with significant amounts of data, few were equipped to consistently provide the millisecond response times required in the online advertising industry.

The move from supporting one major ecommerce property to serving an extended network of PPC publishers and advertisers placed new demands for managing huge volumes of data and processing it reliably and quickly within milliseconds. The adMarketplace database would work in concert with an existing HP Vertica data warehouse. However, the self-funded company needed an affordable alternative to traditional database management systems (DBMS) that carried $1 million-plus price tags for what adMarketplace needed to do. The first attempt was a popular open source, NoSQL database designed to handle Web-scale volumes of data. adMarketplace had it in production for three months. However, it presented two problems. The first was the complexity of set up. Second was an inability to scale reliably; adMarketplace could not get it to provide the 100% uptime required for the business even working with the database firm’s consultants. At the same time, adMarketplace was expanding the amount of data it managed, and the need for another database solution quickly became clear.

Adfonic’s mobile ad buying platform enables customers to run performance, rich media, and video ad campaigns across a wide range of inventory sources to drive direct response, increase consumer engagement, and build brand awareness. To support the many functions of its platform, Adfonic has placed a priority on applying the right data management solution to each requirement. Some parts of the platform have been well served by traditional SQL database technology. However, when Adfonic rolled out its Madison mobile demand-side platform (DSP) utilizing real-time bidding (RTB), the company quickly realized the need for a different approach. The ad-server in Madison, designed to serve as a real-time ad traffic handling system, demanded responses within 5 milliseconds. Adfonic evaluated SQL databases but found that they failed to meet the critical access times. The company then reviewed several commercial and open-source NoSQL and key-value store (KVS) solutions.

Early on, InMobi’s executive team began evolving the Miip platform to meet changing market demands. Platform innovation brought new requirements for the underlying database technology in terms of performance, scalability, reliability, ease and efficiency. These requirements were beyond the scope of Hbase, which they had originally deployed as their key value store. A decision was made to evaluate other database options, including Cassandra and Aerospike.