EBOOK
Proven use cases and evaluation criteria
EBOOK
Proven use cases and evaluation criteria
Chapters
INTRO
Over the last decade, analytics has evolved from static reporting and dashboards to real-time, interactive solutions that answer data questions in the business context. Having access to BI tools is only half the battle—today’s analytics consumers are busy SaaS customers and internal stakeholders who require a seamless experience in the apps they already use.
As we enter the AI-fueled Data Renaissance, many organizations are adopting embedded analytics to quickly fill product gaps, enable promising new generative AI technology, and make data more accessible to their own decision-makers. It’s easier than ever to provide cohesive, branded data experiences within your product or popular business applications.
In fact, the embedded analytics market is projected to grow to over $100BN by 2027. As with any period of fast growth and flourishing new ideas, it’s up to you to assess the market, determine the cost-benefit of change, and update your strategy to maintain your competitive edge. Use this guide to understand the landscape of embedded analytics, define your use cases and success metrics, and assess vendors for the best value.
Embedded analytics allows product leaders to build engaging analytics experiences within their product. While early examples were limited to basic dashboards and charts embedded via iFrame, today’s software teams can weave powerful, dynamic dashboards and self-service analysis tools into nearly any application by leveraging an analytics platform built specifically to deliver ongoing, best-in-class data experiences. Even better, you can embed these experiences in days—not months or quarters.
While they aren’t all created equally, many modern analytics platforms offer some version of embedded analytics as a service, providing teams the tools to integrate their analytics and reporting capabilities directly into other applications, products, and services. Product and data teams can use these embedded capabilities as a scalable, easy-to-manage alternative to building a custom analytics solution in-house.
While embedding an existing analytics platform into your products or SaaS services solves many of the obstacles around data access, capacity, development velocity, and cost, it can bring its own set of challenges to the table:
Some solutions are still too tailored to the analyst, alienating a large swath of business use cases.
Others don’t allow developers to fully customize their analytics offerings to meet the needs of enterprise customers.
And many are simply revamped legacy data solutions that can’t perform as quickly and reliably as modern, purpose-built BI platforms.
To select the right
embedded analytics
solution, leaders across
your organization
must first agree on a collective vision.
As you embark on your embedded analytics journey (or look for an alternative to existing solutions like Tableau), you must be able to identify your organization’s most critical use cases. Start by getting in the room with business, product, data, and developer stakeholders. Discuss your goals and define how success will be measured.
Below, you’ll find 10 key embedded analytics questions to help you begin your discovery process. Answering these questions will help you determine your use cases and the best way to implement them.
Who are our ideal analytics end users? Do we have personas for each customer segment?
How often do we want customers to use our analytics?
How important is self-service? Can our users all code or will they need a natural language, visual interface?
What kind of customization will our customers likely require?
How will we measure success and ROI? Which outcomes and metrics should we track?
What are our data monetization goals? Will we monetize directly by selling data as a service, or indirectly through internal analytics?
How will we price and package our data analytics offerings?
Do we have any special security, compliance, or governance concerns?
Which developer and analyst resources can we commit to this initiative?
What would be the quantifiable repercussions of doing nothing, or keeping the status quo?
To aid in your discussion, consider these top four embedded analytics use cases and real-world examples during your initial vision- and goal-planning session. Pay special attention to the potential outcomes you want to deliver for your business.
Use these outcomes as a pillar to ensure organization-wide buy-in and alignment. The more your team stays rooted in the outcomes, the easier it will be to make critical decisions during the implementation, maintain focus and motivation, and inevitably measure the success and impact of the project.
A primary use case for embedded analytics is expanding your product capabilities and optimizing delivery cycles. Incorporating white-labeled visualization and analysis tools into your product offering helps you attract new customers, increase expansion deals, and open up entirely new revenue streams.
Potential outcomes
Increase ARR from new logos and expansion deals by selling analytics as a paid feature or as part of a premium SaaS pricing tier. By embedding self-service, customizable dashboards for different customer segments, you’ll add value for existing users and have more success closing enterprise deals where analytics are non-negotiable.
Add intuitive analytics features that encourage customers to keep coming back. Embedded analytics with self-service capabilities help a wider audience to use your product regularly, including non-analysts who use natural language search vs SQL or want to ask their own questions on the fly without building a report from scratch.
Increase the stickiness of your product with robust analytics, reducing churn by giving users insights they can’t get elsewhere. This type of feature also helps your customers make the business case for your product—your analytics can quantify their ROI and produce persuasive visualizations and reporting to the leadership team.
Accelerate product development by decoupling your core product roadmap from analytics and AI capabilities. Embedded analytics won’t block releases or distract engineers from their main focus, as it requires fewer resources to manage and operates independently from your codebase.
Implementation tip
This product-focused use case for embedded analytics is often part of a larger initiative to modernize your architecture and your data stack. If you’re making this change alongside a move to cloud-based data infrastructure or an investment in AI technology that requires access to real-time data at scale, consider how each component will integrate before investing.
Harri built a three-tiered analytics product for recruiters and talent teams in the hospitality industry. By tailoring personalized experiences for three different personas, defined by their level of data access, they were able to win a new major contract.
Embedded analytics platforms also offer expansive opportunities for self-service partner portals—think third-party portals for suppliers, dealers, agents, and other partners outside of your customer base whose actions affect your bottom line. As part of their relationship with your company, you might provide these third parties with a login to a branded experience where they can explore performance metrics, benchmarks, and other relevant data.
Potential outcomes
Give partners an easy way to track their performance, filling a gap in their own internal reporting. This often increases performance, allowing them to see where they’re missing their targets and make improvements that contribute to both of your bottom lines.
Attract independent agents, marketplace members, and other users for whom analytics is an important factor in continuing to partner with your company. By feeding the user analytics that underlines your company’s value add, you’re investing in a healthier, long-term, and mutually beneficial relationship.
Increase the stickiness of your product with robust analytics, reducing churn by giving users insights they can’t get elsewhere. This type of feature also helps your customers make the business case for your product—your analytics can quantify their ROI and produce persuasive visualizations and reporting to the leadership team.
Answer questions and make decisions faster. Third-party portals provide a shared view of the metrics important to both you and your partners. Live dashboards eliminate the need for producing and emailing reports, while self-service capabilities let partners answer questions for themselves without manual back-and-forth with your partner service team.
Help partners take action. Real-time analytics can make partners more proactive, especially when you can build in automated, custom alerts to flag important trends and events. For example, if a supplier is notified that demand is increasing via website activity or buying intent, they may decide to scale up production and avoid any delays in order fulfillment. You may even consider building automated actions that trigger an alert or action when identified metrics meet a predetermined threshold.
For companies that sell data as their core product, embedding analytics can enhance the value for data consumers. Instead of simply offering thaeir data in rows and columns, you give customers an interface to visually explore, search in natural language, and find answers using AI or other analytics languages—like SQL.
By embedding analytics, you deliver immediate value—empowering your customers to explore data from the moment they subscribe to your data sources without having to invest in additional data visualization software. By increasing the perceived value of your product, users will be willing to pay more for your services.
Potential outcomes
Raise prices for existing data products or offer real-time analytics as an add-on. Embedded analytics gives you additional ways to monetize your data, allowing you to build scalable, flexible pricing models with premium tiers that include user-friendly analytics. You can also open new revenue streams by serving customers exclusive access to specific datasets.
Eliminate usage roadblocks by making it easier for different user personas to adopt and stay engaged with your data products. While direct access to a database might be ideal for developers and analysts, other stakeholders are more likely to find ongoing value from a product that lets them analyze your data from a visual, no-code interface.
Empower your customers to build better products for their own end users. By using embedded, white-labeled analytics in their applications, your customers can justify higher pricing and meet the requirements of larger, more mature buyers.
MDaudit helps healthcare organizations to reduce compliance risk, improve efficiency, and retain more revenue streams. Their flagship RCM platform provides the insights customers need to stay compliant, reduce risk, and pass audits—instantaneous, self-serve data is a crucial part of the equation. The team rolled out embedded analytics and AI-powered natural language search, transforming their data offerings in a fraction of the time it would take to build it themselves.
This internal embedded analytics use case empowers decision-makers and data stakeholders across your organization. Give non-analysts and line-of-business users full access to important metrics, via live dashboards, natural-language search, and exploration features that require zero code.
The beauty of self-service products lies in their scalability and ease of implementation—you can create dashboards or reports once, embed them in whichever SaaS products make sense for each team, and give business users the ability to drill down and answer their next question without ever needing to leave their existing workflow—avoiding the switching cost of navigating to yet another tool.
Potential outcomes
Make more money for your business by improving key metrics around customer acquisition, adoption, and expansion. RevOps, CROs, and product leaders can all use self-service analytics tools to learn what’s working and optimize.
Attract independent agents, marketplace members, and other users for whom analytics is an important factor in continuing to partner with your company. By feeding the user analytics that underlines your company’s value add, you’re investing in a healthier, long-term, and mutually beneficial relationship.
Customer success teams can improve their NPS, drive renewals and upsells with contextual customer data, and use behavioral data to understand where and why users have friction in their products.
Customer success teams reduce time to resolution and other support metrics that affect the bottom line, while other departments have full visibility into how their activity impacts operational efficiency.
Customer success teams reduce time to resolution and other support metrics that affect the bottom line, while other departments have full visibility into how their activity impacts operational efficiency.
While most companies have a BI solution in place and some even self-service, its value is contingent upon whether users are willing to log into a separate analytics tool to get answers. Embedded self-service analytics gives your stakeholders an integrated and completely streamlined experience that increases widespread adoption.
The classic question of build vs buy is still a lively conversation in the world of embedded analytics. Some companies start with a few quick custom dashboards to meet a time-sensitive customer request only to find themselves burdened with their ongoing upkeep as the company grows. Others decide intentionally to build and manage in-house analytics, only to realize that they’ve tasked their engineers with maintaining a full-fledged BI platform over their own core products.
To deliver an analytics experience that delights your users, you have to do more than the bare minimum. And this takes time, especially in a world with rapidly changing customer demands for AI-powered, search-based data analysis.
Many companies opt for the “buy” path to get to market quickly without compromising on features and scalability. Building custom analytics spreads your engineers thin, often taking months to even roll out a solution. Instead, you can get up and running within hours—not months, leaving the meticulous work of building visualizations and analysis tools to a vendor while your data team and developers focus on creating powerful data models and excellent products.
The IOT arm of Cox Communications was stifled by the limitations of its antiquated legacy BI tools. Their team was incurring over $90,000 a year reacting to ad-hoc data requests and spending hours each time a new dashboard had to be manually added. By adopting an embedded analysis solution with natural language querying, they minimized engineering cycles and saved nearly 80% of those costs.
Another advantage to buying an embedded analytics platform is that you’re not solely responsible for incorporating new technology into your product. When building analytics in the age of AI, you must be able to quickly incorporate the latest generative AI and ML capabilities into your analytics solution. Customers are demanding real business value from GenAI, and trying to deliver this on your own is impossible without greatly impacting delivery in other areas.
Once you’ve decided on your use cases, it’s time to dive into vendor research. We’ve identified seven evaluation categories you can use to assess each potential platform—look for each of the features and capabilities listed below to understand how well the vendor is prepared for today’s data reality.
TIP #1
A strong embedded analytics provider makes your end user’s experience intuitive and seamless, from exploring visualizations to querying data using natural language. The platform should enable business users to ask and answer their own questions, customize their views, apply filters, and drill down into the data—all with as much or as little code as they prefer.
Live, interactive dashboards
Dynamic dashboards with real-time data empower users to explore live data on their own. Look for vendors that offer more than predefined drill paths—flexible analysis tools, conversational BI, and AI-automated insights.
Seamless UX
A customizable user experience that fits the look and feel of your product, with options to change colors, fonts, and logos as well as build your own visualizations from your connected data sources.
Clean, intuitive visualizations
Look for a wide range of visualizations with customization options, including charts and graphs like histograms and bar graphs, maps, and heatmaps.
Accessibility
Responsive design, compatibility with assistive technologies, and a multimodal experience that users can access across a broad range of devices.
Powerful, natural language search
A critical feature for self-service analytics, your platform should let users search and analyze their data without using SQL.
TIP #2
With GenAI becoming more interwoven with our online experiences, this should be an area you pay close attention to. Evaluate how the platform incorporates AI and LLMs to empower not only the end user, but also your team. And pay specific attention to the way they prioritize security and accuracy.
NLP-based search
A search interface that allow users to ask questions like they would in ChatGPT to get AI-generated answers from your data. This extends the value of your analytics far beyond those who know how to query in SQL or other programming languages.
Human-in-the-loop feedback
Human oversight to AI-generated answers. Your platform should allow users to confirm, reject, or edit results based on their domain expertise, avoiding 100% reliance on AI while it’s still undergoing rapid evolution.
AI-powered insights
Advanced uses of AI, including anomaly detection, predictive analytics, and conversational BI. If vendors aren’t exploring creative ways to use AI, they’re unlikely to stay at the forefront for long.
Transparency
Full visibility into how NLP and AI-driven insights are translated into SQL with visuals showing how each answer was generated.
While the hype around LLMs is high, finding the real value of GenAI for BI is a larger discussion, especially when you’re bringing that experience to your end users through an embedded analytics solution. Vendors should be prepared to discuss how they approach augmented analytics and explain the specific AI capabilities they use to automatically prepare data, identify patterns and trends, and make predictions.
TIP #3
The most effective, impactful analytics platforms offer robust tools that help developers quickly embed and deploy analytics anywhere they’re needed. Ask the vendor to see examples of how current customers use their platform, including teams with more advanced, customized implementations that relate to your use case.
Low-code embed
Embed search and visualizations into web and mobile applications with a few lines of code, and look for availability on existing front-end cloud ecosystems like Vercel.
REST API
Have the option to pull data from the platform and present the data in your own custom-designed UI or trigger actions in other apps based on query results.
Range of embed options
Choose whether you embed certain components—like single visualizations, entire dashboards, or search and AI functionality—or the full capabilities of the platform.
Resources
Look for accessible resources like a community or a Discord channel, and check to see if there are relevant training materials and certifications to help ensure your team’s success.
Developer playground
Test the platform with interactive developer playgrounds or a free trial.
White labeling
Ask about client-side and server-side customizations for creating an on-brand experience for your customers.
SDK
Access an enterprise-ready SDK that offers a full set of programmatic capabilities to deploy, manage, and maintain an embedded deployment. Ideally, your developers will be able to work in the languages they already know and love.
TIP #4
To get the most out of an embedded analytics platform, it’s essential to be able to integrate and connect with various data sources, third-party applications, and other business tools. Your developers must be able to extend the platform’s value without requiring substantial resources or slowing time-to-market.
Modern APIs
Powerful APIs that minimize unnecessary coding and help developer teams perform programmatic operations like retrieving data, managing dynamic permissions and access for users and groups, performing version control on analytic content, and more.
Configurable utilities
Customizable actions that sync data and perform data writebacks from your analytics platform into other B2B applications. You should be able to rigger workflows with webhooks, customize schedules and alerts, and configure actions across multiple products.
Third-party plugins
Extended capabilities that allow you to add new, custom visualizations.
Integrated playground
An integrated playground rather than a separate IDE for developers to prototype and preview outcomes with existing analytic content.
TIP #5
The analytics platform you choose should be built on enterprise-ready, cloud-based infrastructure that can handle modern data requirements. Ask which cloud platform they run on, how they plan for fluctuations in capacity, and how they help you meet SLAs.
Auto-scaling clusters
The platform should automatically scale horizontally to handle large datasets and high traffic, or decrease capacity when you no longer need it.
Business continuity
Ask about disaster recovery measures, backups, and continuous monitoring capabilities that ensure continuity even during a major event.
Direct query architecture
The platform should support direct and live queries to powerful cloud data platforms like Snowflake, Databricks, Google BigQuery, and AWS Redshift—leveraging your existing investments in data storage rather than forcing you to extract data into in-memory engines.
Performance monitoring
Ask how the platform tracks query response times, loading speeds, and overall system performance to identify bottlenecks or areas for improvement.
TIP #6
Assess how serious the vendor is about security, privacy, and compliance by looking for features like multi-factor authentication, advanced access controls, and the ability to manage governance across the organization.
Authentication
Look for SAML-based SSO compatibility, multi-factor authentication, IP whitelisting, and automated session timeouts to adhere to your standards for access controls.
Permissions
Manage and enforce how your data is used by anyone across your organization with enterprise-grade row-, column-, and object-level security.
Audit logs
Look for full visibility into activity at the calculation, table, column, and query level.
Compliance
Ask whether the platform complies with rigorous standards like SOC 2, GDPR, and HIPAA.
Governance
Automatically adopt the security protocols and permissions from the application you are embedding into, ensuring end users are granted the appropriate levels of access.
TIP #7
Dig into the services and support each vendor provides on top of their products.
Onboarding and implementation
Ask what kind of onboarding services, including dedicated resources and hands-on implementation tasks the vendor is willing to help with.
Ongoing support
Which support tiers are available and how much do they cost? Will you have a dedicated contact, phone and online support, or just access to a helpdesk?
Documentation and training
Look for comprehensive documentation, tutorials, and support resources to assist users in navigating the platform effectively. Check for developer-specific resources that will accelerate any initiatives you have to extend the platform’s capabilities.
Custom consulting/professional services
Additional services to help you get up and running faster. Ask about different levels of involvement, from connecting and configuring your data to embedding to full-scale transformation.
ThoughtSpot Embedded empower you to build self-service, AI-Powered Analytics in your products without the resource drain of developing an in-house analytics solution from scratch.
Now, you can accelerate data-driven decision-making for all your employees, customers, and partners. And with our enhanced ThoughtSpot Embedded Developer SDK, you can ship scalable, tested, and fully customized analytics experience in your applications and portals.
Embed live, self-serve dashboards in minutes
On the next page, you’ll find a worksheet with 10 key embedded analytics questions to help you begin your discovery process. Discuss these questions with your key stakeholders to help you determine your use cases and the best way to implement them. In the subsequent pages of this chapter, you’ll also discover some of the top embedded analytics use cases to consider during this initial vision- and goal-planning session.
Configure, extend, and automate
Create on-brand embedded experiences, configure governance settings, integrate with cloud platforms like Snowflake, Databricks, Google BigQuery and Amazon Redshift, and use ThoughtSpot APIs to extend the capabilities of your embedded analytics products.
Take advantage of AI
With natural-language search, AI-powered insights, and advanced automation capabilities, ThoughtSpot Embedded gives you answers whenever you need them, without predefined drill paths, while human-in-the-loop capabilities ensure your team can override AI-generated responses with their own expertise.
As the analytics market evolves and matures alongside advances in AI, the cost of building your own solution from scratch will continue to be prohibitive to many software companies. ThoughtSpot Embedded opens up new opportunities and revenue streams while protecting and accelerating your core product initiatives. Learn more—schedule your demo with a product specialist today.