Real-World Case Studies and Customer Success Stories with Appen platform

Appen Data Annotation Platform helped Trust Lab analyze user sentiment on millions of pieces of user generated content. Trust Lab used... this data to identify high risk and unsafe content, accounts, and transactions at scale. The platform enabled Trust Lab to provide detailed trust and safety recommendations to internet platforms. This improved user safety and helped platforms enforce better content moderation. Trust Lab praised Appen for its expertise and scalability in delivering sentiment metrics.
Show More..

Appen's computer vision automation helped a leading global security and aerospace company fight wildfires. The company used machine learning to... predict wildfire paths and improve resource deployment. Appen's platform enabled accurate data labeling with multi-sensor integration and advanced annotation tools. The solution saved time by allowing data annotation at the storage point, keeping data secure. The platform supported many data types, helping data scientists use new sources and improve model accuracy.
Show More..

Appen helped Infobip speed up AI chatbot deployment by providing high-quality, crowdsourced datasets in multiple languages. Infobip needed fast, accurate... data to train its Answers chatbot platform for global clients. Appen delivered hundreds of unique utterances per intent, meeting a 95% relevance threshold. This partnership reduced Infobip's time to deployment and improved chatbot training quality. Infobip valued Appen's managed services and quick support for tight project deadlines.
Show More..

Appen Data Annotation Platform helped Realeyes speed up video annotation. Realeyes cut annotation time from three months to two weeks.... Appen's platform let them scale data labeling for thousands of hours of video. Realeyes used Appen's quality controls and custom tools for accurate, secure results. The solution supported global, culturally diverse data needs for emotion and attention analysis.
Show More..

Appen's data annotation platform helped Brandwatch analyze over 500 million documents daily. Brandwatch replaced their freelance network with Appen's global... crowd, improving speed and reliability in data annotation. The platform enabled fast experimentation, real-time metrics, and integrated QA tools. Brandwatch now delivers more accurate digital intelligence insights to clients, increasing agility and customer confidence.
Show More..

FlamingoAI used Appen's training data to launch its virtual assistant fully automated from day one. The challenge was to solve... the cold-start problem for financial and insurance clients who wanted instant automation. Appen provided high-quality, real-world data to train the assistant before deployment. After seven weeks, the assistant answered over 80% of customer questions and achieved a 30% conversion rate for life insurance quotes. FlamingoAI improved customer experience and sales conversion with Appen's data solutions.
Show More..

Appen helped Microsoft Bing improve search quality in the U. S. and global markets. Appen assembled and trained linguistic teams...quickly, surpassing quality standards. They processed millions of search data pieces monthly in over a dozen markets. Appen developed proprietary data analysis tools and delivered hundreds of training hours each month. Their flexible staffing and project management enabled Bing to enter new markets fast and maintain high-quality search results.
Show More..

Appen helped Nearmap scale its AI for aerial imagery with high-volume data annotation. Nearmap needed a flexible, high-quality workforce to... manage complex 3D image annotation. Appen provided a dedicated team, growing from 5 to over 180 analysts, and delivered over 180,000 hours of annotation annually. This partnership accelerated Nearmap’s AI product launches and ensured operational resilience. Nearmap maintained high quality and throughput while meeting growing demand.
Show More..

Appen helped Dialpad boost labeler accuracy from 70% to 88% in just weeks. Dialpad uses Appen to create transcription and... NLP training data for their speech recognition models. The platform enables accurate audio transcription, data classification, and geolocation labeling. Dialpad now maintains high accuracy across diverse models. This partnership supports scalable, high-quality machine learning for human conversation analysis.
Show More..

Appen's text-to-speech engine helped GuildLink deliver audio Consumer Medicines Information to patients. GuildLink needed to make medicine information accessible for... people with vision impairment, language barriers, or learning disabilities. Appen built a custom lexicon and processed text files to create accurate audio output. GuildLink became the first to offer audio CMIs, leading to partnerships with Australian government websites. This improved access to medical information for many consumers.
Show More..

Appen's Machine Learning assisted Video Object Tracking helps HERE Technologies speed up map data annotation. HERE needed to label thousands... of kilometers of road sign video for accurate mapping. Manual annotation was slow and costly. The solution combines human labeling with machine learning to track objects across video frames. HERE now creates actionable data for map fine-tuning much faster than before.
Show More..

A leading multinational technology company needed an automatic speech recognition (ASR) system for children’s applications. Their existing ASR, trained on... adult speech, failed to process children’s speech accurately. They partnered with Appen to collect and transcribe 105 hours of audio from 400 child speakers across diverse demographics. Appen’s linguists developed scripts and managed data collection, helping the company deliver a new ASR for North American English. The result enabled the client to launch edutainment platforms for children, meeting project goals on time and on budget.
Show More..

Appen supported Cohere by providing expert annotators and real-time data generation for PANDA Plus, enhancing the Command LLM. Appen delivered over... 2,400 expert hours in 12 weeks, enabling high-quality, real-time annotation and feedback. This partnership improved model alignment, safety, and performance for enterprise AI deployments. Cohere used Appen’s AI Data Platform for structured, scalable preference data and fine-tuning.
Show More..

Appen helped a leading AI platform improve its AI music generation feature. The client needed annotated music data to train... their model and meet genre standards. Appen provided expert music annotation and real-time collaboration. This led to better quality AI-generated music, more advanced features, and a faster market launch. The platform now delivers a stronger user experience and higher engagement.
Show More..

Appen used its AI data platform (ADAP) to help a leading model builder evaluate 3-6 large language models (LLMs) in complex domains like healthcare, legal, and finance. The team delivered... over 500,000 annotations in 5-day sprints of 50,000+ each. Expert evaluators ran structured A/B tests to check model accuracy, relevance, and Responsible AI standards. The project gave the client actionable insights for model improvement and domain-specific accuracy. Appen's process enabled rapid iteration and better LLM performance across industries.
Show More..

Appen used its AI data platform to help a global technology company improve its large language model (LLM) across 70+ dialects in 30+ languages. The project delivered... over 250,000 rows of dialogue data for supervised fine-tuning. Native speakers ranked model responses for relevance, coherence, and accuracy. The LLM now gives more accurate and culturally aligned answers. User satisfaction and model accuracy improved with better linguistic diversity.
Show More..

HomeTeam by ReflexAI uses AI roleplay simulations to help crisis counselors and loved ones support veterans facing mental health challenges. ReflexAI partnered with... Appen to gather high-quality, ethically sourced training data and fine-tune their AI model for sensitive conversations. The project achieved 93% positive user feedback, with users reporting greater confidence in handling mental health and suicide discussions. Appen's diverse contributor base ensured the model reflected real-world veteran experiences. The collaboration improved efficiency and reduced the risk of harmful AI responses.
Show More..

Appen’s AI Data Platform helped Onfido boost AI fraud detection by 10x. Onfido needed secure, flexible data labeling for biometric... and document checks. Appen delivered a custom on-premise annotation tool, meeting strict privacy and scalability needs. Weekly collaboration ensured precise data for machine learning. Onfido used Appen’s API for continuous model retraining, improving identity verification accuracy.
Show More..

Appen helped GuildLink convert consumer medicines information (CMI) from text to speech. GuildLink needed fast, accurate audio versions for patients... with vision impairment or language barriers. Appen built a custom Australian English lexicon and processed medical terms for clarity. The solution enabled GuildLink to distribute up-to-date audio CMIs within 15 minutes of updates. GuildLink became the first company worldwide to offer audio CMIs, improving medicine information access for many.
Show More..

Nearmap needed to scale its AI data labeling for high-volume aerial and 3D imagery. Appen provided a dedicated, on-site team... in Cavite, growing from 5 to over 180 analysts. The team delivered more than 180,000 annotation hours annually, supporting pilots and full production. Nearmap accelerated time-to-market for AI-powered products with Appen’s flexible workforce. The partnership maintained high quality and operational resilience as project volumes increased.
Show More..

Appen's machine learning assisted video object tracking helped HERE Technologies speed up map data annotation. HERE needed to label thousands... of kilometers of road sign data for their sign-detection models. Manual frame-by-frame annotation was too slow and costly. Appen's tool combined human labeling with AI to track objects across video frames. This made it much faster to create ground truth data for HERE's mapping algorithms.
Show More..

GuildLink used Appen's text-to-speech technology to make medicine information more accessible. Patients with vision impairment, language barriers, or learning disabilities... can now access audio versions of Consumer Medicines Information. Appen built a custom lexicon and automated the conversion of text files into high-quality audio. GuildLink became the first company globally to offer audio CMIs. This innovation led to partnerships with Australian government websites.
Show More..

Dialpad used Appen to improve their speech recognition and NLP models. They needed better training data after their previous provider... only reached 70% accuracy. Appen delivered transcription and audio categorization services. Within weeks, labeler accuracy rose to 88% and stayed high across many models. Dialpad now scales custom training data for hundreds of clients using Appen's platform.
Show More..

Cohere used Appen to scale supervised fine-tuning and LLM evaluation for its Command model. Appen provided 200 expert annotators, delivering... over 2,400 hours of real-time feedback in 12 weeks. The PANDA Plus program enabled high-quality, preference-based data collection and model improvement. Appen's platform integrated directly with Cohere's API for live annotation and editing. This partnership helped Cohere enhance model alignment, safety, and performance for enterprise AI deployments.
Show More..

A leading multinational technology company needed an automatic speech recognition system for children’s applications. Their existing system, trained on adult... speech, failed to process children’s speech accurately. Appen helped by collecting and transcribing 105 hours of audio from 400 child speakers across diverse demographics. The new ASR system now powers North American English edutainment apps for kids. The project finished on time and within budget, meeting all objectives.
Show More..

Appen’s Data Annotation Platform helped Trust Lab analyze millions of user opinions on social media content. Trust Lab used sentiment... analysis to identify high-risk and unsafe content at scale. The platform enabled detailed trust and safety recommendations for internet platforms. With Appen, Trust Lab provided user sentiment metrics to help partners enforce safer online environments. This improved user safety and retention for social media platforms.
Show More..

Appen helped Microsoft Bing improve search quality in the U. S. and global markets. Appen quickly assembled and trained teams...of data annotators and linguistic experts. They processed millions of search data pieces monthly across more than a dozen markets. Appen developed a proprietary data analysis tool and delivered hundreds of training hours each month. Their support enabled Bing to benchmark performance and expand rapidly into new regions. Appen also provided process improvements and flexible staffing for ongoing success.
Show More..

MediaInterface used Appen's pre-labeled datasets to expand its AI-powered speech recognition solution, SpeaKING, into the French market. The company needed... a French lexicon with high-quality phonetic transcriptions, especially for names and places, to meet local healthcare documentation needs. Appen provided datasets with 21,000 French names and 14,000 place names, filling critical data gaps. This enabled MediaInterface to offer better customization and improved speech recognition for French healthcare clients. The result was faster deployments and higher product accuracy for new customers in France.
Show More..

Onfido used Appen’s custom on-premise data labeling tool to boost AI fraud detection by 10x. Onfido needed secure, flexible annotation... for sensitive biometric data to fight advanced fraud. Appen delivered a scalable, customizable solution that handled images, videos, and documents. Weekly collaboration ensured precise data for training Onfido’s machine learning models. The partnership enabled continuous model retraining and improved identity verification accuracy.
Show More..

Brandwatch used the Appen Data Annotation Platform to speed up data analytics. The platform replaced their freelance network, giving access... to a global crowd for faster, more reliable data annotation. Brandwatch now ingests and analyzes over 500 million documents daily. Real-time metrics and automatic quality checks improved their model release process. The team delivers more accurate digital intelligence insights to customers.
Show More..

ReflexAI used Appen to build HomeTeam, an AI platform for veteran mental health support. Appen provided high-quality, ethically sourced training... data to fine-tune the AI for sensitive conversations. The project achieved 93% positive user feedback from crisis counselors and loved ones. HomeTeam now helps users practice mental health and suicide prevention talks. The partnership improved model efficiency and reduced the risk of harmful responses.
Show More..

Microsoft Translator wanted to add more languages, including rare and endangered ones, to its AI translation platform. The challenge was... finding and annotating enough data for these languages, especially where native speakers are few. Appen sourced language data from native speakers and provided expert annotation, including solutions for gender-ambiguous sentences. With Appen’s help, Microsoft Translator now supports 110 languages, making knowledge more accessible and preserving linguistic diversity.
Show More..

Realeyes used the Appen Data Annotation Platform to speed up video data labeling. They needed to process thousands of hours... of footage quickly and accurately for emotion and attention analysis. Appen provided a scalable, culturally-diverse annotation solution. Realeyes reduced annotation time from three months to two weeks. The platform's quality controls ensured high data accuracy. Integration and custom tools helped Realeyes serve global clients better.
Show More..

Appen helped GumGum speed up data annotation for machine learning. GumGum used Appen's platform to classify text and images for... digital ad placement. The platform let non-engineers set up annotation jobs and supported multiple languages. GumGum now annotates 10,000 rows of data in days, not weeks. Model development is 10 times faster, freeing up data scientists for research.
Show More..

FlamingoAI used Appen's training data to launch fully autonomous virtual assistants for financial and insurance clients. The assistants were ready... to work from day one, skipping the usual human training phase. After seven weeks, the virtual assistant answered over 80% of customer questions and achieved a 30% conversion rate for life insurance quotes. Customers found the experience quick and convenient. Appen's data helped FlamingoAI solve the cold-start problem and improve online customer service.
Show More..

Appen helped a leading large language model builder evaluate model accuracy and relevance. The project delivered over 500,000 annotations in... 5-day sprints, covering complex domains like healthcare, legal, and finance. Appen used expert evaluators and its AI data platform for structured A/B testing. The client gained actionable insights to refine models and improve Responsible AI compliance. The process enabled rapid iteration and better domain-specific performance.
Show More..

Infobip used Appen’s custom datasets to train conversational AI chatbots for global clients. They needed high-quality, diverse language data to... improve chatbot understanding and customer satisfaction. Appen delivered validated, unique utterances in multiple languages, meeting strict quality and relevance standards. This partnership helped Infobip speed up chatbot deployment and reduce reliance on in-house data collection. Infobip valued Appen’s managed services and fast, responsive support.
Show More..

Appen helped a leading AI platform improve its AI music generation feature. The client needed annotated music data to train... its model. Appen recruited expert music professionals to classify and review music pieces. This sped up the launch and improved the quality of AI-generated songs. The platform added new generative music features and enhanced user experience.
Show More..

Appen helped a global technology company improve its multilingual LLM using supervised fine-tuning. The project covered over 70 dialects in... 30+ languages. Appen recruited native speakers to provide human preference rankings and structured feedback. More than 250,000 rows of dialogue data were collected. The LLM now delivers more accurate and culturally aligned responses.
Show More..

CallMiner used the Appen data annotation platform to analyze sentiment and emotion in call center audio. Before Appen, CallMiner could... only process about 3,000 data samples in a few months. With Appen, they now analyze tens of thousands of samples much faster and more accurately. The platform's reporting features replaced manual calculations and improved confidence in results. Appen's security-compliant annotators freed up CallMiner's team to focus on higher-value tasks.
Show More..

Appen's annotation platform helped CallMiner scale sentiment and emotion analysis for customer service calls. Before Appen, CallMiner could only process... about 3,000 data samples in a few months. With Appen, they now analyze tens of thousands of samples quickly and accurately. The platform's reporting features replaced manual Excel work, saving time and boosting efficiency. CallMiner also gained more diverse perspectives and improved model explainability.
Show More..

Adobe Stock used Appen's platform to improve search relevance across 200 million assets. The challenge was to identify images with... features like copy space, which metadata alone could not capture. Appen enabled Adobe to annotate images with polygons, helping train models to surface key attributes. This solution made it easier for users to find valuable images quickly. Adobe combined human-in-the-loop machine learning with metadata for better search results.
Show More..

Appen helped a major international software provider update its Unicode Common Locale Data Repository (CLDR) across 66 markets. The provider needed... in-market resources and strong project management for this complex, multi-stage project. Appen sourced and managed nearly 200 participants, ensuring high quality data entry, voting, and forum participation. The client saved time and money, avoided sending employees into the field, and achieved a successful CLDR update. The partnership now continues on a recurring basis for global projects.
Show More..

A leading software and hardware company used Appen to test its confidential mapping software before a major mobile OS release. The company needed... accurate, local feedback on walking and transit directions in multiple cities, but required strict confidentiality. Appen recruited and trained local testers, managed secure device distribution, and ensured all participants signed NDAs. The project improved map accuracy, saved costs, and maintained secrecy. Local expertise sped up testing and provided thorough feedback.
Show More..

Appen helped a top-10 multinational software provider optimize its global SaaS eCommerce platform. The provider faced challenges with transaction funnels... across multiple markets and product types. Appen designed a multilingual usability study to find bottlenecks and improve user experience. The study delivered clean, accurate data for both quantitative and qualitative analysis. The provider improved its eCommerce sales and returned for further optimization studies.
Show More..

Appen helped a leading creative software company improve their AI video description generator. The challenge was inaccurate and unclear AI-generated... video descriptions. Appen used expert human validation and automated checks to refine 40,000 descriptions. The process achieved over 95% accuracy. The client now has more readable, coherent, and accurate video descriptions at scale.
Show More..

Appen helped a leading graphic design software company evaluate and refine their LLM image generator. The project focused on creating... high-quality, culturally relevant images from text prompts in 20+ languages. Appen localized prompts and reviewed AI-generated images for cultural fit and design quality. Hundreds of evaluations per language ensured the model met user expectations. The result was improved user satisfaction and engagement for the global design platform.
Show More..

Microsoft Translator used Appen's AI data solutions to expand its translation platform to 110 languages. Microsoft faced challenges sourcing and... annotating data for rare languages. Appen sourced native speakers and provided high-quality, annotated datasets. The collaboration enabled Microsoft Translator to support 108 new languages, including endangered ones. This helped preserve linguistic diversity and improved global access to knowledge.
Show More..

Appen helped a leading software provider optimize its global SaaS eCommerce transaction funnel. The provider faced challenges with user experience... and sales in multiple markets. Appen designed a multilingual usability study to find issues in the purchase process. The study used feedback from over 50,000 users worldwide. The provider improved its eCommerce platform and increased sales efficiency.
Show More..

Appen helped a leading multinational software provider optimize its global eCommerce transaction funnel. The provider needed to improve sales and... user experience across multiple markets and product types. Appen designed a multilingual usability study to gather user feedback and identify bottlenecks in the purchasing process. The study revealed areas for improvement in the search, selection, and purchase steps. The provider used the data to enhance its eCommerce platform and increase sales in key markets.
Show More..

A major international software provider needed to update its Unicode Common Locale Data Repository (CLDR) for 66 markets. The company partnered... with Appen to source in-market resources and manage the complex project. Appen provided project management, documentation, and local expertise, ensuring high quality data entry and support. The client saved time and money by outsourcing, and achieved a successful, timely CLDR update. The partnership led to ongoing collaboration for future global projects.
Show More..

Adobe Stock used the Appen platform to improve search relevance for over 200 million assets. Adobe needed to help users... find images with features like copy space and object isolation, which were not captured by standard metadata. Appen provided accurate training data by annotating images, helping Adobe build better models. Now, Adobe Stock users can quickly find the most useful images for marketing materials.
Show More..

Appen evaluated an LLM image generator for a leading graphic design software company. The project covered 20+ languages and focused... on cultural relevance and visual quality. Appen localized prompts and reviewed AI-generated images for cultural fit and design standards. Hundreds of evaluations per language ensured the model met user expectations. The result was improved user satisfaction and engagement with the global design software.
Show More..

Appen helped a leading creative software company improve its AI video description generator. The company needed accurate, readable, and scalable... video descriptions for its software features. Appen validated and refined 40,000 AI-generated video descriptions, reaching a 95% accuracy rate. Human experts corrected errors and improved grammar, while automation ensured consistency. The project reduced factual mistakes and made descriptions more precise and clear.
Show More..

Appen provided in-market evaluators to a leading multilingual search engine provider. The team reviewed and labeled ad data, helping the... client measure ad relevance for specific queries. This improved ad targeting and allowed the provider to analyze and fix irrelevant ads. High-quality data from Appen supported machine learning, leading to more relevant ads and higher revenue. The client expanded search advertising into new markets with confidence in ad quality.
Show More..

Appen's platform helped a major global eCommerce company speed up ad-hoc feature testing. The company needed fast, flexible evaluation for... site features without using up internal resources. Appen provided project management, a pool of experienced raters, and a platform with API integration. Projects launched within two days and finished in two to three days on average. The client scaled evaluations, kept quality high, and met tight deadlines while staying on budget.
Show More..

Appen helped a leading social media company collect over one million data samples in less than two months. The company needed... diverse training data to improve its machine learning model for understanding user intent, sentiment, and entities. Appen recruited hundreds of participants quickly and delivered high-quality, varied data. The client released its product on time and improved its platform features. The crowd model also helped control project costs.
Show More..

Appen helped a leading multilingual search engine improve local business listings. The search engine struggled to keep listings accurate as... more businesses appeared online. Appen used in-market evaluators to verify and correct over 750,000 listings in two years. The project grew from 10 to 440 evaluators across 31 markets. This improved user experience and data quality for local search.
Show More..

Appen helped a leading social media platform improve content relevance with a personalized news feed. The company needed a diverse... rater pool and fast global scaling to train its machine learning algorithm. Appen delivered a scalable onboarding module and 24/7 data collection, expanding the pilot into an ongoing program. The platform now delivers more personalized content and refines its algorithm faster, meeting strict quality standards and adapting quickly to changing needs.
Show More..

Appen helped a major social network provider improve its search functionality. The company switched from a previous vendor due to... quality and scaling issues. Appen ran a pilot with 80 US-based raters, quickly exceeding quality expectations. Within a year, the program grew to 1200 raters across four markets and fifteen projects. The client now launches new search projects faster and with higher quality data.
Show More..

Appen helped a leading social media platform improve content relevance using personalized data. The client needed a diverse pool of... raters and fast global scaling to train its algorithm. Appen delivered a scalable onboarding module and 24/7 data collection, enabling better personalization. The pilot expanded into an ongoing program, supporting new markets and experiments. The platform now delivers more relevant content and can address issues like news spam.
Show More..

Appen provided a vendor-neutral team of quality analysts for a leading multilingual search engine provider. The client needed to expand... its search engine across 25 markets but lacked in-house expertise for quality feedback across multiple vendors. Appen selected experienced evaluators and auditors to form an independent team, ensuring consistent application of rating guidelines. The program improved the reliability of search results and data quality. Consistent standards and regular guideline updates led to stable metrics and ongoing program success.
Show More..

Appen helped a leading social media company collect over one million diverse data samples in under two months. The company needed... high-quality training data to improve its machine learning model for understanding user messages. Appen recruited hundreds of participants quickly and delivered data covering many categories, including transportation and sports. This enabled the client to launch its product on time and enhance features like the help center and ads. The project also helped control costs and provided valuable geographic and demographic diversity.
Show More..

Appen helped a major global eCommerce company speed up feature testing with ad-hoc evaluations. The company faced urgent evaluation needs... that strained internal teams. Appen provided project management and a pool of expert raters. Projects launched within two days and finished in two to three days. The client completed ten projects in a few months, keeping quality high and costs controlled.
Show More..

Appen helped a major social network provider improve its search functionality. The company switched from another vendor due to quality... and scaling issues. Appen ran a pilot with 80 US-based raters, quickly exceeding quality expectations. Within a year, the program grew to 1,200 raters across four markets and fifteen projects. The client now launches new search projects faster and with higher quality data.
Show More..

Appen helped a leading multilingual search engine improve its local business listings. The search engine struggled to keep listings accurate... as more businesses appeared online. Appen used in-market evaluators to verify and correct over 750,000 listings in two years. The team checked names, addresses, phone numbers, and more. This work improved user experience and search accuracy.
Show More..

Appen provided in-market evaluators for a leading multilingual search engine provider. The evaluators reviewed and labeled ad data to improve... ad relevance across multiple international markets. This helped the provider measure ad performance, identify root causes of irrelevant ads, and train machine learning algorithms. The project enabled the provider to expand search advertising into new markets with confidence. High-quality data from Appen led to more relevant ads and increased revenue opportunities.
Show More..

Appen's crowdsourcing solution helped Zefr boost video review output from 15,000 to 100,000 videos per month. Zefr needed to scale... contextual advertising data labeling and improve quality. Appen provided a large pool of reviewers, enabling faster, more accurate machine learning model training. Zefr now delivers precise, scalable contextual targeting for brands. The partnership gave Zefr guaranteed throughput and improved data quality.
Show More..

Appen helped GumGum speed up data annotation for machine learning. GumGum used Appen's platform to classify text and images for... digital advertising. Annotation tasks that took weeks now finish in hours or days. GumGum's model development process is now 10 times faster. The platform is easy to use and supports multiple languages. GumGum can now focus more on research and less on manual data labeling.
Show More..

Appen helped Shotzr speed up image labeling for location metadata using machine learning. Shotzr faced a challenge with scaling manual... annotation for nearly 100 million images. With Appen's platform, Shotzr identified over 17,000 images that did not need extra labeling after the first job. They expect to remove over 61 million assets from location data review. This lets Shotzr focus on images that need location tags and build new models faster.
Show More..

Appen's data annotation services helped EmPushy improve their AI model accuracy from 60% to over 85%. EmPushy used Appen’s Open... Crowd and Dedicated Crowd to annotate push notifications for better audience targeting and GDPR compliance. The project captured over 200,000 campaigns and 223,000 judgments across 35 domains and 500 brands. This enabled EmPushy to deliver more empathetic marketing campaigns and acquire new clients. Appen provided speed, quality, and strong customer service.
Show More..

EmPushy used Appen's data annotation services to improve push notification campaigns. They needed multi-label annotation and GDPR-compliant data classification for... better audience targeting. Appen provided both Open Crowd and Dedicated Crowd contributors to categorize call-to-action content in notifications. This helped EmPushy raise their model's F1 score from 60% to over 85%. They processed over 200,000 campaigns and collected 223,000 judgments, enabling more accurate marketing models and client growth.
Show More..

Shotzr used the Appen platform to speed up image labeling for location metadata. Their team struggled to keep up with... annotating a growing image library. Appen helped automate the process using machine learning and high-quality training data. After the first project, Shotzr identified over 17,000 images that did not need extra labeling. They expect to remove over 61 million assets from manual review, letting them focus on images that need location data.
Show More..

Appen helped Zefr boost video data review from 15,000 to 100,000 videos per month. Zefr used Appen’s crowdsourcing platform to... label video content for contextual advertising. This improved both the quality and speed of Zefr’s data insights. Zefr can now promise customers faster, more reliable results. The partnership lets Zefr scale operations and focus on growth.
Show More..

Appen Data Annotation Platform helped Johns Hopkins University track millions of spider leg movements for behavioral neuroscience research. The team used... AI and machine vision to annotate 100,000 frames, selecting 10,000 high-quality images for model training. Both LEAP and DeepLabCut algorithms achieved mean pixel errors of 8.2 and 7.6, with error rates of 4.5% and 3.3%. The platform enabled fast, accurate annotation, saving over 1,500 hours of manual work. Researchers gained new insights into spider web-building patterns using high-quality training data.
Show More..

Appen's platform helped the London School of Economics speed up data labeling for political text analysis. The team replaced slow,... expert-only annotation with Appen's global crowd, cutting weeks of work down to four to five hours. They labeled 20,000 sentences from six political parties, each up to 20 times, and replicated studies in several languages. Appen's reporting features and validation metrics improved data quality and reduced bias. The solution enabled LSE to build machine learning models for political text sophistication and publish results in top journals.
Show More..

Appen helped Dr. Mark Harvey create a sustainable database for the Larrakia language. The project started with unlinked and error-prone... text and audio data. Appen aligned the databases, enriched metadata, and added detailed time-stamping. Their team supervised phonetic annotation and acoustic measurements. The result is a useable, long-lasting resource to help preserve and teach the Larrakia language.
Show More..

Appen Data Annotation Platform helped Johns Hopkins University label 100,000 frames of spider web-building videos. This reduced manual annotation time... from over 1,500 hours to just a few weeks. Researchers used the annotated data to train machine vision models and track spider leg movements. The project enabled accurate behavioral neuroscience studies and revealed predictable web-building patterns. High-quality data annotation was key to advancing their research.
Show More..

Appen's data annotation platform helped the London School of Economics speed up political text labeling. The LSE team replaced slow,... expert-only methods with Appen's global crowd, reducing bias and cost. In one project, 20,000 sentences were labeled in hours instead of weeks. The platform enabled multilingual studies and improved reporting. LSE used the results to build machine learning models for political text analysis.
Show More..

Appen helped a leading provider of vehicle electronics software collect speech and text data for in-car infotainment ASR systems. The challenge was... to train the system to understand many dialects and commands. Appen used native speakers and advanced recording to ensure quality and variety. The company deployed the system in over 20 new languages and met strict audio standards in more than 40 languages. This led to smarter, more natural in-car voice experiences.
Show More..

Appen provided a leading global automotive OEM with speech training data for over 20 languages. The OEM faced challenges with... voice recognition accuracy and localization for connected cars. Appen delivered a full-service solution including data collection, in-car testing, and linguistic consulting. The OEM achieved faster data collection, expanded language coverage, and rapid scaling to new markets. This partnership enabled the OEM to improve its in-car voice recognition systems and enter new regions confidently.
Show More..

Appen helped a leading provider of vehicle electronics software collect natural language and text data for in-car infotainment systems. The challenge was... to train automatic speech recognition to understand many dialects and commands. Appen used native speakers to gather spontaneous and scripted speech, plus text data, in over 40 languages. The company rapidly expanded ASR capabilities to 20+ new languages. The result is smarter, more natural in-car voice experiences with strict audio quality compliance.
Show More..

Appen helped a leading global automotive OEM train its in-car voice recognition system for over 20 languages. The OEM faced... challenges with localizing speech data and collecting audio in diverse driving environments. Appen provided localization, data collection, in-car testing, and linguistic consulting. Their Detroit facility enabled realistic environment testing and high-quality data collection. The OEM achieved faster data collection, expanded language support, and rapid scaling to new markets.
Show More..

Appen's platform helped a large US-based gaming company scale its AI-powered customer support. The company needed more training data for... chatbots to better understand player requests. Appen replaced manual Excel processes with a central data repository and enabled fast onboarding of internal labelers. The team grew, reviewing labeled data faster and improving chatbot response times. Data rows increased nearly 12-fold, leading to more personalized support and better customer interactions.
Show More..

Appen’s Data Labeling Platform helped a leading US-based gaming company improve customer support with AI. The company faced challenges scaling... their machine learning model due to limited resources and manual processes. Appen’s platform enabled quick onboarding of internal labelers and centralized data management. The team increased labeled data nearly 12-fold, improving chatbot response times and customer experience. Team leads could monitor performance and adjust resources for better results.
Show More..

Appen helped a popular children's video platform improve search relevance and content filtering. The platform needed to keep children safe... by filtering out inappropriate videos. Appen set up a quality team to review and rate content for different age groups. The team checks tens of thousands of videos each month. They flag videos with hidden inappropriate content, making the platform safer for kids.
Show More..

Appen helped the Larrakia Nation Aboriginal Corporation of People improve their Larrakia language database. The challenge was unlinked text and... audio data, with errors and gaps. Appen aligned databases, enriched metadata, and added phonetic annotation. Their work made the database more accurate and sustainable. The project aims to preserve and teach the Larrakia language for future generations.
Show More..

Appen partnered with CLEAR Global to support chatbot development for mental health in Sheng, a fast-evolving language in Kenya. The project delivered... five consultation sessions, a Sheng language research document, and a reusable template for future language projects. CLEAR Global can now create their own research documents for new ASR models in African languages. Appen improved its process for documenting lower-resourced languages. CLEAR Global praised the detailed and accurate Sheng language document.
Show More..

Appen partnered with CLEAR Global to help develop chatbots for mental health support in Sheng, a fast-evolving language in Kenya. The project focused... on creating a detailed Sheng Language Specific Peculiarities (LSP) document and a template for future research. Appen delivered five consultation sessions and provided tools for CLEAR Global to research lesser-known languages. This work enables CLEAR Global to build better voice-enabled chatbots for underserved communities. CLEAR Global praised the accuracy and value of the LSP document for supporting new language development.
Show More..