嵌入分析以发展为企业级产品随着世界努力应对气候危机和不断变化的天气模式，像 Vitality 这样的平台对于建立各种团队只会变得更加重要和有用。 Vitality 知道第一手资料：其最初的计划是让产品团队在平台中构建基本分析，但客户对分析的需求随着增长而扩大，迅速超过了 Vitality 内部构建的能力。

手动和零星报告Profusion 希望将其客户的业务从对缓慢、被动和手动的基于 Excel 的报告的依赖转变为快速、主动和实时的优化。一家伦敦创意机构要求 Profusion 开发一种解决方案，以改进其向其客户（一家国际现场表演制作公司）报告票务数据的方式。它有两个要求：在易于使用、直观的界面中，通过不同的营销渠道清晰地传达金融投资和回报。能够查询其客户的单个客户视图并提取此数据以在其其他业务工具中使用。

TLC 组织高质量数据以在整个公司和客户中进行分析和分发的方法既费力又费时。客户数据没有以任何真正有意义的方式保留，数据分析和报告大多是手动完成的，并在极其复杂的 Excel 电子表格中进行管理。最重要的是，每月通过电子邮件发送报告，而电子表格的庞大规模正成为问题。他们的数据是任何基于云的系统可以处理的数据量的三倍，因此需要及时进行出色的报告和分析以最大限度地筹集资金。

随着内部分析用例和第 3 方数据源的快速扩展，主要挑战是选择能够处理多个数据源、模型复杂性并支持直接创建仪表板的 BI 平台和云数据平台。在企业方面，该公司渴望简化季度业务审查的报告数据。随着 Quikbooks、Salesforce 和 Pendo 等第三方应用程序列表的增加，客户经理需要一种更简单的方法来统一数据并向他们管理的机构客户展示投资回报率。

分析和 ETL——完全自动化为了向潜在客户展示价值，MindMax 必须衡量和跟踪从漏斗顶部到注册的流程。其遗留的分析和 BI 解决方案需要手动从不同来源提取数据，包括 Salesforce、Google Analytics、Facebook Ads、LinkedIn 和营销自动化系统，以及 Excel 中的手动数据混搭。

As the company grew, Online Commerce Group’s data become so large that, without a new generation Big Data Analytics solution, management couldn’t receive quality reports. The fast growth of the company meant that Data Manager Paul Auen was unable to run reports with large enough data sets to provide any real value to the business end. “Originally, we could do reports manually, but as the data sets got larger, the reports took longer and impacted the transactional database,” Auen recalled. “We were constrained between keeping the servers running or making it available for reports.” Online Commerce Group chose a popular business analytics tool, but it was too complex; Auen’s group was unable to run reports at the speed and frequency needed for the business. Quickly, the team abandoned and turned to Sisense. The company needed a solution that could: 1. Quickly analyze large data sets coming from multiple sources. 2. Provide drag-and-drop modeling and dashboard creation for business users. 3. Automatically generate reports and publish them to the company intranet. 4. Require a minimal capital investment and learning curve for use.

Paylogic considers itself a next-generation technological company, and its organizational structure and operations are highly automated. Instead of employing a standard monolithic ERP system (e.g., SAP), Paylogic has built its IT infrastructure using corporate-environment open-source standards. Automation is approached from a decentralized perspective based on this open-source IT backbone. The company uses different software packages for different functions and ties them together to form an enterprise collaboration platform. The result is impressive: Paylogic enjoys no-compromise IT functionality and flexibility with extremely low software costs. It was into this environment that the company sought to integrate a reporting and analytics tool. The company’s operational and historical data was consolidated into a homegrown data warehouse, based on a MySQL database. Paylogic was looking for a powerful and flexible business intelligence solution that would easily integrate with its existing system, which would not require rebuilding the data warehouse, which would allow non-technical business users to quickly create new reports and analytic queries and which would be relatively inexpensive.

Plastic Jungle’s business success is predicated by its ability to move faster than the competition. When the company brought in a new CFO in 2012, he aimed to deploy a data solution that would match the company culture. Speed, accuracy and agility were the key tenants for the company data approach. “We wanted to make sure we didn’t paint ourselves into corner by taking a traditional approach to data warehousing.” That meant a solution that could grow to massive amounts of data, and allow regular business users to work with data quickly without requiring a huge investment that would leave them beholden to the product.\n\nPlastic Jungle's BI requirements included the ability to:\n• Manage and sustain the entire operations aspect of Plastic Jungle’s data warehouse with little or no operating support from the engineering and IT staff\n• Allow business users to create any ad-hoc reports they required\n• Provide the abstraction layer between the schema and the metrics that the business sought\n• Refresh in close to real-time – a minimum of once per day, and ideally every couple of hours\n• Not incur a significant capital outlay

ProMarket required timely and accurate reporting and analysis of key metrics, such as sales, inventory, profit by product category, spoilage, and optimal order quantities for each store. The company was struggling to process very large amounts of data (over 40 million rows) from its centralized database. The data processing, transformation, and analytics were extremely time- and labor-consuming, making it impossible to generate some of the analytics required by management. Business partners of two leading BI vendors demonstrated their solutions to ProMarket and provided implementation proposals. However, ProMarket selected Sisense due to its ability to fully meet their requirements, faster customization of reports and dashboards, impressive data processing speed, and lower total cost of ownership.

Before implementing Sisense, alpharooms.com faced significant challenges in providing detailed and interactive reports and dashboards to various business functions. The existing tools, such as Excel and OLAP cubes, were not effective in building user-friendly dashboards or offering easy drill-down interfaces. Additionally, IT involvement was required for any modifications, making the process cumbersome and time-consuming. Lee Eckersley, Head of Business Analysis, had experience with other BI tools like Cognos, Hyperion, and Business Objects, but found them too expensive and resource-intensive for the company's needs. The team needed a more efficient and cost-effective solution to handle their data analytics and reporting requirements.

Fiverr needed quick insights on growing data. The company wanted to connect data from MySQL with data from Google Docs, Spreadsheets, and Analytics to better track user actions on their website and mobile app. As users increased, so did Fiverr’s data needs, making it larger and more complex with millions of rows a day from various sources. Despite using an internal big data system based on Hadoop, the data complexity made it difficult for the team to build reports and dashboards quickly. Fiverr’s senior BI director, Slava Borodovsky, emphasized the need for real-time results due to the dynamic nature of their data. The product department relied heavily on BI to determine their product roadmap, making the need for data more urgent as departments began understanding its impact on their success.

Trupanion faced challenges in managing and analyzing large volumes of data across multiple departments. The company needed a solution to track real-time performance, optimize marketing opportunities, and build accurate financial reports. Existing in-house solutions were time-consuming and prone to inaccuracies, leading to a need for a robust BI tool that could be easily used by non-technical users and deployed quickly.

GOAL Academy, an online charter school, faced challenges in effectively utilizing data to keep students on track and align staff with key performance indicators (KPIs). The school relied on various data resources like Google Docs, Excel spreadsheets, and different servers, which made it difficult to create a consistent format accessible online. The institution needed a modern tool to better visualize and understand the data, and to merge multiple platforms into one place. The goal was to find a web-based tool capable of processing data from different sources to help teachers make quick decisions in the classrooms and improve student retention and graduation rates.

Sonny explains that before Sisense, company resources would be invested in reigning in data and wrestling with the complicated process of aggregating, processing and delivering it to the client. “We would build a bunch of pivot tables in Excel on numerous tabs, and then we would give people an import function that would import the raw data so that they could see the dynamic reports in Excel. But there were a number of problems, for example Excel would limit the amount of rows in a report, or the report was slow, or people just didn’t know how to use it.” Sonny also mentions that another significant problem was the task of distributing the data reports to thousands of different clients. To keep the reports current and updated, clients had to manually re-import the data, and eventually customization requests demanded even more time and resources, per client. “Overall, there just wasn’t control over what was happening. On top of that, if I had to update the report configuration, I had to send out thousands of new Excel files that had all the pivot tables defined in them. And ultimately every customer would need us to modify their pivot tables. It was just a nightmare.”

In addition to the custom software that Businessolver had developed internally, they were also using several off-the-shelf technologies such as inContact for their service center; Salesforce.com for sales team and Sage for the accounting department. There was no unified method to tie those disparate data sources together and get an overall picture of the customer interaction. Users would get various exported files from various other users and load them up into MS Excel and analyze in a single location. Due to this lack of uniformity to data access, team members were constantly providing different results to the same question. The problem was not the volume of data but rather the disparate sources of it. It was critical to Businessolver to find and acquire a tool that would allow them to connect all their data sources across the enterprise. That way they would be able to have full confidence that their results were accurate no matter who was collecting the information, and that data could drive decision making based on facts and not based on feelings. Sony Sung-Chu is the Director of Applied Data Science at Businessolver and it was his task to find and test potential solutions. With his years working as a Business and IT Analyst, he came at the question from a very technical perspective. Sony also brought in Sara Johnson, the lead BI analyst for Businessolver, whose background is in economics, math and Business Intelligence. Johnson brought in the first data warehouse at Businessolver and was to be tasked as the internal BI expert, responsible for training and maintenance.

Casting Network’s data was being stored but not really seen, leaving the Sales and Business Development departments tracking KPIs manually. As the company started to expand internationally, the need to aggregate different data sets and evaluate the business on a global scale became even greater. Needing to access Google Drive documents, multiple Quickbook files and 35 SQL databases comprising over a billion rows of data, Nitika had her work cut out for her in pulling together a solution that met Glen’s directive.

Ignite Media processes massive amounts of data, maintaining approximately 3 TB of transaction, demographic, and media performance data. They had been building all their reporting internally using PHP and .Net, but it was becoming increasingly difficult to scale. Writing new reports from scratch to follow a 'hunch' could take weeks, making it impractical to test new ideas. The company had valuable data but lacked the resources to fully leverage it. Mazda Ebrahimi, the VP of Application Development, sought a solution that would allow them to produce results faster and more easily without sacrificing their intellectual property.

Integrous Marketing faced significant challenges in integrating data from multiple disconnected sources such as Adwords, Analytics, email marketing tools, and Salesforce. The process of manually gathering, cleaning, and joining datasets was labor-intensive and prone to errors. Additionally, generating visualizations from the data in Excel was time-consuming and often resulted in outdated and inaccurate information. The company needed a method to automate these steps and provide a better user experience, especially given the limitations in the types of reports and views that some tools provided.

Many companies today often have data that is so complex or distributed, it needs a dedicated team of specialists to clean and maintain it, and only then can a BI tool be useful. There’s a lot of work preparing complicated and dirty data, and many businesses need a simple way to prepare data that most BI tools on the market cannot provide. Mike saw that even after implementing a BI solution for customers. There was often such a steep learning curve, the customer would still be unable to build reports and dashboards independently. Mike’s customers typically come to him after they've failed with another BI solution. His company then takes on managing the client's internal data, some of which as been around for 10+ years, and is big, disparate, and dirty. His team would then wrangle the data together into a BI solution for the client. Mike saw that even after implementing a BI solution for customers, it often had such a steep learning curve, the customer would still be unable to build reports and dashboards independently. It was important that his customers could be self-sufficient, so he wanted to standardize on a tool that he knew would be a win for them.

Quentin Villon, a non-technical data analyst at Qualifa, faced significant challenges in managing and utilizing the vast amount of data collected by the company. Internally, he relied on Excel to manually build activity reports, a process that took two hours daily and resulted in a cumbersome 20MB spreadsheet. This method was prone to errors, with broken formulas and other issues, making it a nightmare to manage. The reports were also delayed, going out a day after the activity took place, which meant they could only acknowledge performance rather than influence it. Generating commission reports was even more time-consuming, taking two days each month. Externally, creating campaign insight reports for potential clients was a labor-intensive process that required several days of manual effort. These reports were crucial for client acquisition but were not reusable, adding to the inefficiency.

foodpanda faced significant challenges with their existing data warehouse, which was unable to efficiently handle terabytes of complex data from multiple sources. The limitations included a lack of data mining functions and the inability to affordably process large volumes of data. Additionally, foodpanda aimed to centralize data to promote transparency and democratization, reducing employee reliance on the BI department. They needed an intuitive, self-service BI solution to shift the focus from data collection to insights and strategy.

The decision to evaluate BI solutions came in the wake of what could be defined as a “cultural revolution” taking place in JDC: from siloed data, with each stakeholder holding on to his or her own data and only sharing it with others on a “need to know” basis, to a corporate culture of complete transparency of information that aims to make data available to everyone, and challenge them to make good use of it. The data itself was being stored in a Cobalt character-based CRM system running on the IBM RS/6000 platform, as well as Excel spreadsheets. This data consisted of sales, operational, utilization, financial and other data. Prior to Sisense, reporting was done with Excel and Access. JDC has been looking for some sort of dashboard software for roughly two years but did not find any product that could suit their needs, until they ran into Sisense in a user group in which two Sisense customers had presented it. The VP of Parts and Business Systems downloaded the trial version right there at the convention, and within two days was already publishing dashboards.

Pantechnik International faced significant challenges in managing and utilizing their vast and scattered data sources. The company struggled with a 'black hole' of largely untapped client data, making it difficult to gain a holistic view of overall performance, cost-effectiveness, and service level agreements (SLAs) of their carriers. Senior management lacked visibility and control over these metrics, which was further compounded by the data preparation nightmare of the ETL process. The company evaluated various solutions, including Qlikview, but found the 'buy before you try' business model unappealing. They needed an embedded BI software with a powerful backend capable of handling thousands of large, scattered datasets.

Blueline’s biggest problem was figuring out which budget allocations were connected to which sources of revenue. A single marketing campaign could bundle together various lines of business, making it hard to get a clear picture of the bottom line of promotional activities or to break down the revenue of product bundles by product and sub-product. The company lacked a way to combine data sources from different departments, leading to standalone reports that didn’t match up. This confusion resulted in inappropriate investments, reactive problem-solving, and missed business opportunities. As Blueline prepared to launch their mobile initiative, Nacer decided it was time to address the problem once and for all, anticipating high volumes of transactions that could cause enormous problems if not managed properly.

EEG had to structure the data to have the functionality and flexibility at an interface level so that they could easily deploy it to any event type and use consistent tools to report on the data. That is where the problem arose: though the data was being collected, the data created data dumps that were cumbersome to understand and not structured to produce easy or accurate data analysis. After Adam sifted through the data dumps of a number of their clients, he realized that some data just couldn’t be connected in a traditional reporting environment - creating gaps of what they could report. After talking to a longtime client, he found that many of their clients were forced to manually pull data and reports together - and were spending a significant amount of time doing it. Adam attempted to solve the problem by doing data analysis for his clients internally and sending out the reports. But, for Adam to do that internally, he had to pull the data out of Salesforce, cobble it together, create reports and charts in Excel and then email them out. Every report would have taken weeks at least, and the clients wanted to see this data frequently and quickly, so it wasn’t feasible.

When CEO Andrew Vignuzzi joined Magellan Vacations, his priority was to enable everyone at the company to access, manipulate, and draw insights from their data quickly. The company needed a BI solution that could provide real-time feedback for agents on sales closings, destination performance, and other metrics. Additionally, the solution had to be user-friendly enough for non-technical users to create their own reports and drill down into the data. The dashboards needed to be agile and scalable without requiring major infrastructure upgrades. The personalized, phone-based service of Magellan Vacations made it difficult to track standard sales metrics like closing rates, commissions, and bookings by destination automatically. The company had already trialed a leading in-memory technology, but its performance was sub-par, requiring specialist IT resources and consultants to work with the tool's proprietary scripts. This tool did not meet their needs, prompting the search for a better solution.

As an organization with many different projects, departments, and moving parts, Amadip-Esment collects vast amounts of information from various sources, including human resources, financial ERP, operational software at restaurants and printing units, and specific software systems containing sensitive data related to persons with disabilities. The team had been manually collecting data and arranging it in Excel pivot tables, which was labor-intensive and limited in analysis. They needed to increase the efficiency of organizing and managing these data pieces and required a core BI platform for managerial reporting and customized data queries. Above all, they needed a system that could bring all their disparate data together for analysis.

WeFi’s database team had been manually running SQL queries, but they struggled to generate the reports that gave the management team crucial feedback. WeFi needed to perform advanced analysis on large amounts of data in three categories: the behavior of millions of WeFi users, including retention activity and data acquisition activity; the performance and activity of wireless networks to which its users are connected; and the activity records of active clients. The average table sizes for these categories were more than 5 million rows, 70 million rows, and 500 million rows respectively.

The operational data at OHT consists of over 20-million records in a 100GB MySQL database. Lior knew that they were collecting all the information he needed to get insights, but he simply couldn’t get to it. Transaction data was coming in pretty fast and, in order to continue to be an industry leader, he needed a way to get a 360 degree view of his business as fast as possible. Lior had various ad-hoc and separate solutions running to try and achieve the reporting and analytics the company needed, including manual analysis and home-grown software. He would often rely on someone from R&D to extract reports or would end up manually doing reporting in Excel, which would take weeks. These efforts were taking significant resources, both human and computer, to try and get the reports that were needed. Many of their analytics requirements were not being met at all, which was leading to a lot of frustration within the company.

Act-On, a software company, faced a significant challenge in managing data from over a dozen web tools used for various business activities. Each tool provided unique BI analytics, making it difficult to identify a single source of truth. The company needed a solution that could integrate these tools seamlessly and provide real-time, actionable insights to improve customer experience and operational efficiency. The complexity of integration, unknown costs, and the need for a tool that could prompt team action were major concerns.