X

CarOnSale, a disruptive pan-European platform for car dealers, identified data as a key differentiator in their market. As an online platform, they aimed to cut through the complexity of traditional car trading by harvesting and analyzing data around car auctions. This would provide them with unique market intelligence. The company recognized the need for a centralized architecture, hosted in the cloud, to collect and analyze data at speed and scale. They explored different options to support ELT (Extract, Load and Transform) as opposed to the traditional ETL approach. After selecting Snowflake as their cloud-based data warehouse, they needed to find the ideal data integration solution. Aynaz Bagherynezhad, Data Team Lead, had used Fivetran in a previous role and when Snowflake recommended it as the best way to connect to data sources, it confirmed her own experience.

Code2College, a nonprofit organization aimed at helping minority and low-income students achieve tech/STEM careers, was facing challenges in managing and analyzing student data. The organization's data, including student attendance, grades, and teacher input, was kept in spreadsheets or gathered by word of mouth. This approach was inefficient and time-consuming, especially when specific data on a student's performance or an overall view of the student population was required. The organization used Salesforce for operations and Canvas as a learning management tool. However, extracting information from these platforms to answer a single question would require a day's work, which was untenable given the small size of the data team. The team wanted to centralize their data using Google's BigQuery data warehouse tool to streamline retrieval and expedite responses to student needs. However, the challenge was how to transfer data from platforms like Salesforce and Canvas into BigQuery.

Condé Nast, a global media leader with 37 brands reaching millions of consumers, was faced with the challenge of managing and monetizing trillions of data points generated from its digital assets. The company lacked a central mechanism for managing and maintaining data integration sources, making data not readily available to consumers downstream. The demand to integrate more sources globally continued to grow, and pulling data into the data lake with custom scripts was cost prohibitive. Each marketing technology platform had its own API, data structure and other properties that required its own custom script. Creating the connectors on the fly and managing them on an ongoing basis wasn’t scalable, posing a significant challenge to the company.

Coupa, a Business Spend Management (BSM) company, provides a cloud platform that digitizes and consolidates spending information across various sectors, creating actionable insights into spending behavior. However, Coupa faced challenges with its own data about its platform and customer usage. The data was siloed, impeding better insights and decision-making. The process of collecting this data and making it accessible to the relevant personnel was complex, costly, and resource-heavy. Coupa had invested in a data team to manage its data, with the goal of pulling data from various sources into a single place for creating actionable insights. However, the analytics strategy was immature and largely consisted of ad hoc procedures. If a UX designer wanted to know how customers were interacting with a particular feature, they’d have to request the engineering team to build a script from scratch, a process that could take weeks.

Chris Klaczynski, a Marketing Analytics Manager at Databricks, was tasked with supporting the primary marketing objectives of driving pipeline generation, growing the database, and improving ROI. However, as Databricks rapidly expanded, the need for centralized and documented data became more and more apparent. Data silos were appearing around the business, including on Chris’ marketing team, where data was stored in its own data warehouse. It was critical that Chris’ newly-built dashboards were supplied with trustworthy, timely data for marketing operations to keep running smoothly. However, without dedicated engineering resources, and in the face of a rapidly expanding marketing team, scaling with demand became next to impossible. Databricks faced a number of challenges with their traditional data warehouse, including issues with their Salesforce and Marketo pipelines, issues appending data natively to existing tables, and schema changes that were always breaking pipelines, resulting in outages and stale, untrustworthy data.

Daydream, an early-stage startup, provides financial insights to stakeholders across modern businesses. The company's business modeling and planning tool democratizes access to financial information by bringing together processes and data sources that are typically siloed. However, the Head of Engineering for Daydream, Shubham Sinha, faced a critical decision. The success of the startup hinged on its ability to move massive amounts of data from its customers’ cloud-based business systems, each with their own login credentials and access challenges, into the Daydream platform for analysis. The two options were to either ask its customers for login credentials to their business systems, posing a potential security risk, and use custom-built data pipelines to onboard data or to rely on Fivetran to broker the credential sharing exchange and onboard data using its pre-built data pipelines. Maintenance also posed a challenge as each cloud platform has its own APIs, processes, and data structures, many of them requiring custom integrations through scripting.

The Denver Broncos, a successful pro football team, faced a significant challenge in maintaining their data pipelines. The team's Senior Director of Ticket Strategy and Analytics, Clark Wray, and his lean team were spending excessive time on home-brewed data integrations. Whenever an original data source or API changed, it would disrupt the data connections they had built, often halting the flow of information for hours. If these issues were not addressed promptly, the business risked operating on inaccurate data. Additionally, the team was constantly adding new data sources to communicate and reach the next generation of fans. It was crucial for them to connect and centralize their email data in Dynamics 365, marketing automation data in Eloqua, and fan feedback in Qualtrics.

DOUGLAS, a leading premium beauty platform in Europe, was facing a significant challenge in its journey to become a 'Digital First' business. The company's existing infrastructure and processes, particularly around Business Intelligence (BI) and data analytics, were not up to the mark. The systems for collecting data were scattered, and there was an overreliance on spreadsheets and manual input, which were not scalable. This lack of a centralized, automated data collection and analysis system was hindering the company's growth and its ability to gain valuable insights from its data.

DPD Polska, a leader in the Polish courier market, was facing challenges with its existing data management system. The company was using a series of on-premises PostgreSQL and Microsoft SQL Server databases to track its trucks, parcels, and people. However, the array of custom SQL databases was preventing DPD from producing timely reports, meeting disaster recovery time objectives, testing new data and analytic products, scaling up its revenues, and increasing its customer base. For instance, one of DPD’s databases had three different usage contexts. The company was in need of a log-based replication solution that would not impact its source systems. The main pain points were the replication time lags, the risk of errors in manual data distribution, and the need for more flexibility, greater reliability, and higher operational scalability.

Engel & Völkers, a prestigious broker of premium residential property and commercial real estate, was facing a significant challenge in integrating various data sources. The company's data engineering team was inundated with requests from different departments for data integration. The process of creating custom solutions to respond to these requests was resource-intensive, leading to prioritization of tasks and inability to cater to the growing number of requests. The company was in dire need of a tool that could reduce the effort required to integrate new data sources and enable faster data integration, thereby promoting wider adoption of self-service analytics within the organization.

Frontify, a platform that helps companies grow their brands, faced a significant challenge in building a single source of truth for their data. The company needed to understand how people interacted with their platform to optimize user experience and resource allocation. However, their data analytics team was small, and their data infrastructure was unstable. They relied on custom Python scripts to pull data from business applications into a MySQL database, which often resulted in slow, incomplete data. Their BI tool was user-unfriendly and slow, causing reluctance among employees to use it. The data team was burdened with the task of updating reports and dashboards. To address these issues and become truly data-driven, Frontify needed a scalable and powerful data stack that could be accessible to everyone.

GroupM, a global media agency based in Oslo, was facing challenges in collecting and analyzing data for their clients. The agency, which serves over 200 clients and provides shared services for other agencies in the group, was using Supermetrics to pull marketing data directly into Google Sheets. However, this method was proving to be inefficient and problematic. Pipelines would occasionally fail due to hard-to-detect issues, and there were formatting problems with the spreadsheets as well as manual errors. Preparing data for analysis in Google BigQuery, GroupM’s data warehouse, was labor-intensive, and clients were demanding faster access to more insights. One client, with a broad business portfolio spanning retail and hotels, was looking for dashboards that could handle historical data analysis as well as day-to-day reports. GroupM was determined to find a more robust solution.

Hashtag You, a brand builder in the direct-to-consumer e-commerce sector, faced a significant challenge in leveraging and structuring data within its organization. As a data-driven company, the use of analytics in marketing, product and customer analytics, and operational analytics was crucial to its business model. Initially, Hashtag You implemented several piecemeal solutions through Google Sheets with self-created data pipelines. However, the company soon realized the need for a centralized and scalable approach to data ingestion. The challenge was not only to establish robust data pipelines but also to connect new data sources quickly and easily. The company needed a solution that would allow non-data specialists to make these connections. Furthermore, the company had to manage advertising across various platforms, combine marketing and webshop data, link with other data pipelines, and analyze campaign performance.

HOMER, an early learning company, was facing significant challenges with its data management. Despite the company's data-driven roots, the data and analytics architecture was considered to be at a foundational stage when Joe Nowicki joined as Vice President of Data and Insights in February 2021. The company's data team was spending a significant amount of time building ETL pipelines, a laborious and time-consuming process. Flattening and maintaining Stripe data was costing the team dozens of hours each month, preventing them from adding value to the wider organization. The legacy data practices had led to a feeling of distrust, with leaders unable to depend on dashboard views and unclear business intelligence. Access to timely data was critical but lacking. There had been multiple visions and revisions of the entire data infrastructure, leading to the selection of Databricks’ Delta Lake, to set HOMER up for future Machine Learning applications.

Houseware, a software development company with less than 20 employees, was facing significant challenges in providing a platform and toolkit for its customers to build internal data products. The company's goal was to go beyond the scope of general analytics and data visualization tools, delivering metrics such as ARR, NRR, customer churn, conversion rate, and other KPIs. However, they were struggling with a lack of data insight, reliability, and availability. Their marketing campaigns were inefficient, and they were unable to turn data into customer retention optimizations. The trust in data was decreasing due to errors. Users had to learn data analytics tools and database methods, such as table joins, and develop custom metrics from scratch. Data dashboards and analytics often broke down, took too long to produce results, or required too much custom programming. Poor APIs and data pipelines limited the types of analytics that developers could construct to meet customer needs. The tools produced insights without any actionable recommendations, and building data connectors required lots of programming time and effort.

Hunt, Gather, an Austin-based creative agency, was struggling with limited reporting tools that hindered their ability to share deep performance data with clients. The agency was in dire need of a holistic approach to the reporting of its digital marketing efforts. They required a suite of tools that would enable the collection and analysis of data, and ultimately the generation of key insights, all in a single location. The development team had previously built a few pipelines on their own, but these were time-consuming and costly. It could take up to six months to build pipelines in-house, and the team was also spending significant amounts on ELT platforms that were proving to be inefficient.

IJsvogel Retail, a Dutch pet and garden products chain with a history of nearly 130 years, was grappling with the challenge of managing and leveraging its vast and disparate data. With over 180 stores, more than 1,600 employees, and over 800 wholesale customers, the company was generating a significant amount of data. However, this data was not being effectively utilized to inform business decisions. Instead, old data and log files were often discarded rather than compiled and analyzed. The company's small IT department found it difficult to promote the adoption of new applications across the company. The lack of a unified, reliable, and stable data source was hindering the company's ability to make informed business decisions.

Imperfect Foods, an online grocer dedicated to eliminating food waste, faced a significant challenge in managing and utilizing its vast customer data. With hundreds of thousands of customers and an increasing number of data sources, the company struggled to act on this information effectively. The lack of a centralized view of the customer data made it difficult to understand what traits led to high-value customers or what factors influenced customers to order. Imperfect Foods needed a way to consolidate all of its customer data across its entire data stack to leverage it for marketing activation to increase signups and product usage. The company was also limited on engineering resources, which further complicated the situation.

Involve.ai, a customer intelligence platform, was facing challenges in providing its customers with a holistic view of their customers due to the inability to pull data from multiple data sources efficiently. The process of data integration was time-consuming and resource-intensive, with unreliable and difficult-to-modify data schemas. The company's clients required different approaches and specific apps tailored to their sales and delivery processes, which the previous data integration solution could not scale to meet. Without access to data from source systems, Involve.ai was unable to produce comprehensive insights, leading to a more reactive approach to data analysis. The challenges included an inability to produce comprehensive and accurate insights, inflexible automation for scheduled data replications, no way to perform data transformations prior to importing into Snowflake, and slower time to market, which limited the company’s growth rate.

The global Covid-19 pandemic forced restaurants worldwide to quickly pivot to delivery and take out services. ItsaCheckmate stepped up to help these restaurants consolidate orders from various ordering apps directly into their existing Point of Sale (POS) systems, eliminating the need to manually transfer the orders to the POS and manage their menus on multiple platforms. With business booming, ItsaCheckmate decided it needed to use data to maintain quality experiences for its customers and enable the support staff to handle an increase in orders. The data was available, but it was cost prohibitive for the company to organize and manage it in any meaningful way. The ItsaCheckmate platform is powered by dozens of integrations with online ordering apps such as Uber Eats, Grubhub, and DoorDash, as well as with all the POS systems that large chains or small mom-and-pop restaurants may use. When an order cannot be processed properly, ItsaCheckmate can resolve each individual error in real-time, but analysts need to conduct a thorough and rapid post-event analysis to resolve the underlying issues that cause these errors to arise to begin with. Systematic analysis of this siloed data was a manual process, requiring analysts to pull a list of order errors into an Excel spreadsheet – a process that could take up to a day.

JetBlue, a major airline operating over 900 flights daily to more than 110 cities, was grappling with the challenge of managing and analyzing the vast amount of data generated by its operations. Every person, plane, and journey generated data points that could provide insights into customer sentiments, revenue forecasting, fuel consumption, aircraft maintenance, and operational readiness. However, the sheer volume of data, sourced from 130 different systems, was overwhelming and difficult to organize. The airline needed a solution that could centralize this data, making it readily accessible for analysis and decision-making. The challenge was to bring all this data into a single platform quickly and accurately for analysis.

Kilo Health, a global leader in digital health and wellness, faced a significant challenge as it rapidly expanded. The company, which started with just seven people in 2013, has grown to over 700 employees managing more than 30 products with over 5 million customers worldwide. This rapid growth led to an exponential increase in data points, which the company needed to manage effectively to become a fully data-driven organization. The challenge was to find a solution that could support this rapid growth and provide intelligent and unbiased insights to stakeholders.

Kuda, a digital bank launched in Nigeria in 2019, experienced a fourfold increase in customers within six months. As a mobile-first digital company, Kuda recognized the need to be data-driven and identified the modern data stack as essential for achieving its goal. Initially, a five-person data team was manually building data pipelines and relied on SQL Server Reporting Services (SSRS) to extract insights from transactional databases. Their data was split into 12 disparate Azure SQL databases with no way to successfully join the data across their internal sources. The team was looking to move from running OLAP queries on an OLTP database, which was proving to be a challenging task. They needed a more scalable solution that would relieve them of having to build and manage data pipelines.

Learnerbly, a learning and development marketplace, was facing significant challenges in managing and utilizing its data. The company had no dedicated data organization, leading to data being siloed across different departments. This lack of a consistent source of truth made it difficult to compare and corroborate records across different sources. It was also impractical to join records across data sources, which impaired their ability to service clients effectively across their entire life cycle. Furthermore, engineers were often diverted from product development to data operations. The company needed better access and control over its data to scale and attract enterprise clients, who have higher employee headcounts and more stringent demands regarding visibility into ROI of adopting a new platform.

Lendi, an Australian mortgage broker with over $12 billion AUD in home loan settlements, was facing a significant challenge in its data management. The company's proprietary technology allows borrowers to search over 2,000 loan products from more than 40 lenders, making it a competitive player in the market. However, the industry's competitiveness and the need to deliver the right experience to the right person at the right time on the right digital platform required reliable, accurate insights into borrowers' needs and preferences. The problem was that building an accurate profile of each customer required tapping into behavioral data on third-party engagement platforms such as Facebook, Google, and Bing. This data was readily available to Lendi, but the insights from each platform were siloed and didn't integrate easily. Even when the data could be brought into the same repository, the data structure was often inconsistent, creating the need to clean the data before it could be put to use.

Lufthansa Systems, a division of Lufthansa Airlines, is a leading provider of IT services in the airline industry, serving around 300 airlines worldwide. One of its offerings is Lido/FPLS (flight planning services), which optimizes flight routes in terms of cost, fuel, and time, generating millions of dollars in extra profits for its customers each year. The challenge was that creating these optimized flight plans required massive amounts of data, including up-to-date weather reports, air traffic data, and airline-specific data such as flight schedules, payload, operational conditions, and contracted petrol prices. Lufthansa Systems needed a solution that would allow its central data repository to receive continuous updates from these data sources and distribute optimized flight plans and other data to each customer's site.

Malt, an online freelancer marketplace, was facing challenges with its data ingestion process which was slowing down analysis. The existing custom-built ingestion framework lacked performance and reliability, and the company was spending too much time debugging data pipelines. The company's Head of Data, Olivier Girardot, was tasked with enabling a small analytics team to access insights and deliver maximum value to the business with fewer resources. The company needed a consistent, automated approach that could be run by two engineers. One of the main objectives was to analyze data from digital advertising to maximize advertising spend. Another challenge was regulatory compliance. Malt needed a solution that protected the personal information of freelancers when their data was moved to the data warehouse, ensuring it was kept within the European Union.

Memrise, a language learning app used by over 50 million people worldwide, faced a significant challenge in identifying gaps in customer engagement. Despite having a robust cloud-based platform and a commitment to data analytics, the small data team was overwhelmed with coding, manually building data pipelines, and fixing broken APIs. As the customer base grew, the need for a more focused approach to analytics became increasingly critical. The team needed a solution that would allow them to spend less time on technical issues and more time on analyzing data to improve the user experience and drive business growth.

Mighty Digital, a growth, analytics, and strategy consulting firm based in Ukraine, was faced with the challenge of helping a transportation startup optimize its ad budget efficiency, user activation rates, and campaign engagement. The startup had no clear understanding of the cost-effectiveness of various ad campaigns due to an inefficient ETL pipeline built using Airflow and Python transformations. The data architecture was prone to errors, missing data points, and overall inefficiency, leading to inaccurate advertising results. The existing solution was convoluted, involving multiple softwares stitched together to create a complex architecture that provided no insights. This led to a lack of data insight, reliability, and availability, inefficient marketing campaigns and ad spend, inability to turn data into customer retention optimizations, and decreased trust in data due to errors.

MyCamper, a Swiss start-up likened to the Airbnb of campervans, faced a significant challenge in managing and analyzing the data collected on its web platform. The company recognized the importance of this data, but initial attempts at analytics were laborious and time-consuming, involving manual extraction of data from Excel spreadsheets. Using Google Analytics proved to be easier but was limited in scope. As an early-stage start-up, MyCamper had more pressing priorities and lacked the in-house skills to manually build out data pipelines. This resulted in a gap in data analytics that needed to be filled. The company also struggled with historicizing data in a way that could be retrieved for analysis. They were unable to track specific data sets, compare historical data to present, or even have a comparable baseline.