LinkedIn

Twitter

GitHub

Blog Landing

<p>With all the growth in machine learning models within organizations, there’s a challenge that we hear repeatedly: We’ve built the data pipelines, we’ve trained and tested the models, but it’s hard to get the results into the hands of business users, so they can navigate it and make sense of it all. </p>
<p>Sound familiar?</p>
<p>At Sisense, we have a phrase for this challenge, “bridging the last mile of machine learning.” It’s a bridge that every data science, data engineering, or analytics team must put in place to turn the ML models they are building and deploying into easy-to-understand and interpret insights. Failure to do so often leads to models aging out before they are effectively utilized, or worse, never being put to work at all to drive decision-making. </p>
<p>The thing is, machine learning models are now pervasive, so it’s more important to derive as much value as possible from them than ever. In healthcare, machine learning can predict patient readmissions or identify patterns in patient data to improve care management and resource allocation. For eCommerce, machine learning models can suggest products or services tailored to individual preferences. In financial services, they can be used for fraud detection and risk assessment. Manufacturers leverage ML to forecast equipment failures and recommend maintenance actions and supply chain optimization. </p>
<p>But you have to enable every corner of the organization to be able to act on those models so that they can be turned into actionable insight.</p>
<p><strong>Turn your AWS ML models into intelligence that anyone can act on</strong></p>
<p>If you’re running Amazon SageMaker, then you’re probably well acquainted with plugging in popular machine learning algorithms and frameworks, including TensorFlow, PyTorch, and XGBoost, or using pre-built or custom algorithms to train your models. Or you may be running Amazon Forecast, using advanced learning techniques, such as neural networks (and all their variants) or autoregressive integrated moving averages (ARIMA), to generate highly accurate forecasts and create probabilistic predictions. Perhaps, you are using other ML services, like Amazon Comprehend, to identify key phrases, language, sentiment, and other common elements in text. Or, like many companies, you’re using Amazon Redshift to provide high-performance analytic queries on large data sets—perhaps more focused on descriptive versus predictive analytics.</p>
<p>But the secret is to surface and visualize that data and securely democratize it, so your business teams in marketing, sales, finance, service, or operations can easily put it to work―they can operationalize it. Or if you’re in a product team, and you’re providing an application to your customers, why not embed analytics imbued with AWS-based AI/ML models based on your customer’s data in your app so your customers themselves can make better decisions in your app? </p>
<p>For example, Skullcandy uses AWS Comprehend, a natural language processing (NLP) service, to analyze and extract insights from large volumes of customer feedback. This analysis helped them identify common themes and trends in customer sentiment, which they used to guide their product design process. And they use ML to drive return rate predictions that are crucial for managing inventory, production, and overall business strategy. But they leverage Sisense to provide interactive dashboards and visualizations so their teams can use the data to make better decisions.</p>
<p>As Brent Allen, Director of Infrastructure and WebOps, summed it up, “Sisense allows data to be accessible in a way that everyone can see it, understand it, and filter it independently. That was probably the biggest win for us.”  </p>
<p>What it means is to use an analytics platform that your users—whomever they are—can use to easily engage with using dashboards, widgets, and a semantic layer that makes the complex underlying data simple to access, understand, and fully self-service. Or allow developers to add data and insights using that same business layer using an SDK to take a code-first approach to composing analytics. </p>
<p>An analytics platform that is built around an API-driven architecture, webhooks, and connectors can establish a secure and efficient communication channel between it and your AWS tooling, whether SageMaker, Forecast, Comprehend, or data services like RDS, Redshift, Athena for data mashups from sources like Parquet, ORC, or just plain old CSV. </p>
<p><strong>Enable your business teams to get more out of Redshift</strong></p>
<p>Redshift is often part of every AWS-based data engineering initiative. In fact, like many organizations, you might be starting your AWS data journey with services like Amazon RedShift or Athena, especially if you’re migrating from an older on-premises data warehouse.</p>
<p>What’s important is to establish a dynamic and seamless integration between your analytics tooling that your business teams will engage with daily and Amazon Redshift. That way, you can empower users to access Redshift-powered data in their daily queries and dashboards.</p>
<p>But the key is to choose the right analytics tooling that hides the Redshift query complexity from end users. Optimizing queries and workload management to ensure efficient data retrieval and processing, maximizing the use of sort keys, or taking advantage of RedShift's automatic query rewriting feature, is all critical to driving query performance. But ultimately, your analytics tooling needs to tap into it automatically so your end-users don’t have to think about it. </p>
<p>At Sisense, we take full advantage of the advanced features and optimizations offered by Redshift, such as the MPP-aware query optimizer and columnar-oriented data storage. By making use of these enhancements, we’re able to efficiently process complex analytic queries that involve multi-table joins, subqueries, and aggregations automatically. And the connection we provide is a live connection so that data is automatically and immediately refreshed whenever fresh data becomes available in Redshift.</p>
<p>Another innovation is Sisense Notebooks, which can enable data scientists, data engineers, and analysts to access key features in Amazon Redshift programmatically and incorporate them into their Sisense analytics. For example, you can tap into Lambda UDFs, to execute existing procedural assets for performing additional processing that may otherwise be difficult to conduct in a typical query. You can even tap into Redshift ML from Sisense Notebooks to incorporate predictions directly into your dashboards. </p>
<p>There are so many ways to use Sisense with AWS, turning the data in your AWS-powered cloud warehouse or data mashups into powerful visualizations or bringing AI/ML into your dashboards and reports.</p>
<p>In fact, that’s just the tip of the iceberg, to learn more about the ways Sisense works with AWS, check out the eBook “<span style="text-decoration: underline;">Three essential analytics use cases on AWS that you've got to know.</span>” You’ll get practical strategies to power up your AWS ML and analytics initiatives and find out how ZoomInfo and Gameday are boosting their success.</p>

Secrets to boosting your machine learning analytics success on AWS

<p>Every product manager today is exploring what it means to build a data product, because <a href="https://medium.com/kyligence/every-product-will-be-a-data-product-19e648f0333">every product is (or will be) a data product</a>. The margins on data products are fantastic, and they drive real value for their end customers. However, every data product isn’t equally valuable - products that quickly surface the most relevant, actionable insights, provide context for a decision, or propose prescriptive next steps are far more valuable than those that simply share aggregate metrics or assist with diagnostic insights.</p>
<p>Determining what specific insights are most valuable is often an iterative process, involving rapid development and re-development of what gets delivered to end users, and that can be a real challenge, since building useful analytics capabilities is often labor-intensive and highly manual.</p>
<p>At Sisense, we help product teams build data products using embedded analytics. Sisense Fusion includes all the building blocks that advanced data products need to deliver predictive and prescriptive analytics directly into product workflows, and we work closely with the developers and product managers who build their solutions on our platform. Because of that, we know exactly how much work goes into getting to product/market fit, and we’re developing new AI-powered capabilities to dramatically reduce the effort to get a great data product to market.</p>
<p>In our last post about the Analytics Assistant, we described how its context-aware NLQ and automated insight generation makes discovering relevant information much simpler for end users. In this post, we’ll talk about how the Analytics Assistant uses the same capabilities to instantly generate data product capabilities, supporting users who prefer no-code options as well as pro-code developers.</p>
<p>The Sisense Analytics Assistant is an intelligent tool designed to help users design and format visualizations, customize UX behaviors, tell data stories, and troubleshoot issues in their data products. The assistant is powered by advanced machine learning algorithms that can process large amounts of data and generate insights that can inform decision-making processes.</p>
<p>One of the key capabilities of the Sisense Analytics Assistant is its ability to help non-developers design and format custom data visualizations in a no-code environment. Users can simply describe the type of widget they want, the title, color, and other formatting options, and the assistant can interpret the user's requests and generate the JSON and javascript to implement the widget accordingly. This means that product teams save time and effort by delegating the widget design process to the assistant, and focus on analyzing the data.</p>
<p>The Sisense Analytics Assistant also supports dashboard generation and customization. Users can leverage the assistant's capabilities to edit and customize their dashboards, including layout, formatting, adding dashboard filters, and adding a summary to the dashboard description. The assistant will be able to provide recommendations on the best practices for designing effective dashboards, and suggest improvements based on the user's needs.</p>
<p>Data Storytelling is another key capability of the Sisense Analytics Assistant. The assistant can suggest the best way to tell a data story by automatically creating a dashboard on a topic, choosing the right queries and visualizations, and supplying textual insights that summarize the key findings on the topic. The assistant will be able to identify patterns and trends in the data and suggest visualizations that effectively communicate the insights to the user.</p>
<p>These capabilities can go far beyond enabling non-developers to experiment with new data visualizations in their data products; developers can also leverage this approach to allow the end users of a data product to specify what insights they need.  By integrating the Analytics Assistant into the data product UX logic, analytics and data visualizations can be completely dynamic and highly personalized, with minimal effort from the development team.</p>
<p>For example, a user with a decision to make might ask for relevant insights in a specific step of their operational workflow. The data product, using the context of the workflow stage, the user’s query, and the Analytics Assistant can dynamically generate analytic queries and visualizations.</p>
<p>Finally, the Sisense Analytics Assistant has troubleshooting capabilities. Users can ask for guidance from the assistant on the best way to do things in the Sisense Fusion platform. The assistant has knowledge of the Sisense documentation site, community posts and blogs, as well as the GPT knowledge base, and can recommend the best practices to get things done. The assistant will be able to identify common issues and suggest solutions to the user, helping them to save time and effort troubleshooting problems in the system. This is tremendously helpful for developers and non-developers alike.</p>
<p>With the Analytics Assistant, we’re making it much easier for product teams to develop compelling and valuable data products that deliver real value for end users. Experimentation becomes far more affordable for teams, and as a result, their data products move quickly to reach product market fit.</p>

Using contextual AI to build new analytics tools

<p>Whether your background is in data and you’re new to product management, or you’re a seasoned product manager now responsible for developing a data product, you’re likely developing a plan to add analytic insights to increase the stickiness of your product and improve expansion revenue. You - or your CPO - have probably read the research from Eckerson Group that reports data products drive an average increase in ACV of 25%.</p>
<p>Time-to-market is likely your primary consideration; getting something in the hands of users that can be used to iterate, improve, and outpace your competitors is putting your data product as the top priority on your roadmap.</p>
<p>However, there are pitfalls ahead! Early missteps can eat up valuable time and create false signals to your leadership that a data product might not be viable in your market. At Sisense, we have worked with thousands of customers building data products. Here are the key tips from looking at the wins (and failures) of those initiatives.</p>
<p><strong>One: Setting your sights too low - only considering descriptive analytics</strong></p>
<p>Basic descriptive analytics often don’t provide enough insight to move the value dial. It’s no longer a differentiator, and your customers expect more. While they provide a foundation for understanding historical data―and that’s crucial―they can’t make the impact on your customers’ businesses that predictive and prescriptive analytics can because they are less actionable. </p>
<p>Here are five examples across logistics, customer success, sales, marketing, and finance and accounting applications that make the value difference clear:</p>
<ol>
	<li><strong>Logistics and Inventory Apps:</strong> Descriptive analytics can show which products sold well in the past, but they cannot predict future demand. Predictive analytics, however, can forecast demand patterns, enabling businesses to optimize inventory levels and reduce carrying costs. Prescriptive analytics might suggest optimal ordering quantities and reorder points to minimize stock issues and improve supply chain efficiency.</li>
	<li><strong>Customer Success Apps:</strong> Descriptive analytics can identify the customer attrition rate but offer no insights into future churn or strategies to reduce it. Predictive analytics can identify customers most likely to churn, allowing businesses to proactively engage and retain them. Prescriptive analytics could recommend targeted retention strategies, such as opportunities to personalize promotions.</li>
	<li><strong>Sales Apps:</strong> Descriptive analytics can provide an overview of historical sales but cannot anticipate future revenue. Predictive analytics can estimate future sales based on past data, customer behavior, and market trends, allowing businesses to set realistic targets and allocate resources efficiently. Prescriptive analytics could suggest targeting specific customer segments or adjusting pricing to maximize revenue and achieve sales goals.</li>
	<li><strong>Marketing Apps:</strong> Descriptive analytics can reveal the performance of past marketing campaigns but cannot predict the success of future campaigns or suggest improvements. Predictive analytics might forecast the potential impact of different marketing strategies, helping businesses allocate resources effectively. Prescriptive analytics can recommend the ideal marketing mix, channels, and messaging to optimize campaign performance and ROI.</li>
	<li><strong>Finance and Accounting Apps:</strong> Descriptive analytics can provide a historical view of cash inflows and outflows but cannot anticipate future cash flow patterns or provide guidance for improvement. Predictive analytics can forecast future cash flows based on historical data, market trends, and other factors. This allows finance to identify potential cash shortages and take proactive measures to improve cash flow, such as reducing expenses or increasing sales. Prescriptive analytics might recommend optimal cash flow management strategies, such as recommending better payment terms with vendors.</li>
</ol>
<p><strong>Two: Failing to blend data products directly into your user experience</strong></p>
<p>Poorly integrating analytics into your product will hurt productivity, and what’s worse is your users will lose context. For example, HBR found that workers spent just under four hours a week reorienting themselves after toggling into a new application. Over the course of a year, that adds up to five working weeks or 9% of their annual time at work.</p>
<p>Analytics that are poorly integrated and don’t precisely mirror your product’s look and feel can create a disjointed and less effective user experience, making your data products less actionable and valuable, hurting your product in three ways:</p>
<ul>
	<li><strong>Disconnected workflows:</strong> Avoid requiring your users to switch between multiple user interfaces to see your analytics, disrupting their workflow and reducing efficiency. Constant context-switching can lead to frustration and lower productivity, as users will spend extra time and effort navigating between tools.</li>
	<li><strong>Inconsistent user experience:</strong> Some analytics tooling often has distinct interfaces and features that create an inconsistent user experience. This inconsistency can result in confusion and hinder users from fully leveraging the powerful analytics in your app. A simple acid test—if the analytics tooling you’re integrating can’t precisely match your Figma designs, you’re probably falling short.</li>
	<li><strong>Hindered user adoption:</strong> Users may be reluctant to adopt poorly integrated analytics, especially if they are complex or cumbersome. This resistance can lead to decreased utilization, hurting adoption, satisfaction, and, ultimately, retention. </li>
</ul>
<p><strong>Three: Building the analytics infrastructure versus building the analytics</strong></p>
<p>At the beginning of the journey to add data products to their products, every product manager often faces the question of how much analytics componentry to build in-house. Should the team spend cycles building the visualization components, ad hoc query capabilities, predictive algorithms, and the underlying data engine to ensure strong query performance as customer data grows?</p>
<p>The truth is that product teams should spend their time on the data, the metrics, and understanding the kinds of questions that their customers would benefit from answering in-app―i.e., the parts that deliver the *real* differentiated value. In fact, product teams that start with building the analytical infrastructure typically face several challenges:</p>
<ul>
	<li><strong>Long time-to-market:</strong> Developing a full analytics solution from the ground up, or one based on basic components like Chart.js from scratch, can be a lengthy process, delaying time-to-market and benefits realization. In the meantime, competitors leveraging off-the-shelf solutions gain a competitive advantage.</li>
	<li><strong>Development costs:</strong> Building analytic infrastructure to deliver data products always requires a significant investment in time, resources, and budget, and allocating resources for continuous maintenance and updates can lead to substantial unplanned costs. It’s too easy for development teams to get buried in ongoing performance optimization.</li>
	<li><strong>Limited features and scalability:</strong> Data products built on home-grown analytics tooling often lack extensive features and scalability. As customer requirements evolve or data volume increases, in-house solutions are often hard to adapt.</li>
	<li><strong>Maintenance and support challenges:</strong> Home-grown analytics built into your product will require ongoing maintenance, updates, and support, which can strain a product and services team’s resources and focus.</li>
	<li><strong>Opportunity cost:</strong> The time, effort, and resources spent on building analytics components, predictive algorithms, and data infrastructure, are often better utilized by focusing on delivering the right metrics, insights, recommendations, and actions.</li>
</ul>
<p><strong>Four: Lack of commitment to your differentiators </strong></p>
<p>Another important pitfall to avoid in a data product is trying to be all things to all customers. While it may seem like a good idea to offer a wide range of features and capabilities to appeal to a diverse audience, it can actually hurt your product in the long run.</p>
<p>Firstly, trying to cater to every customer's needs can result in a dense product with a poor signal-to-noise ratio for the user, making it difficult to use and navigate. This can be overwhelming for customers and ultimately drive them away from your product.</p>
<p>Secondly, trying to appeal to everyone can dilute your product's unique selling proposition and make it less differentiated in the market. Focusing on a specific target audience and delivering a product that meets their needs and solves their pain points can help establish a strong USP and position your product as a market leader. And a solid commitment to the core value proposition of your product can often make more of an impression on reviewers than trying to hedge your bets by offering minimal implementations of a wide range of features.</p>
<p>Thirdly, trying to be everything to everyone can also increase your development and maintenance costs, resulting in a drain on resources and a less efficient use of your team's time.</p>
<p><strong>Reaching Product-Market Fit</strong></p>
<p>Instead, focus on understanding your target audience's needs and pain points and develop features and capabilities that address them specifically. This will help you deliver a more focused and differentiated product that resonates with your target audience and enables you to stand out in the market. Once you provide a few insights, your customers will tell you what else they want to see.</p>
<p>With an awareness of these four classic pitfalls - setting your sights too low by delivering data, not insights; letting a disjointed user experience limit the utility of your insights; moving too slowly by overplanning or overinvesting too early in analytics infrastructure; and not identifying your core value proposition - you are now equipped to deliver a valuable data product quickly, and well-positioned to collect the feedback that is essential to finding product-market fit.</p>

Pitfalls that every product manager should avoid

Editor’s picks

<p>Since 1770 when Britain's James Hargreaves patented his spinning jenny that allowed a single spinster to run eight spindles and produce eight times as much raw thread and yarn as before - cutting both time to market and labor expense involved with producing textiles in Blackburn, Lancashire - doing more with less has been the driving force behind growing a business.</p>


<p>This productivity remains an elemental economic force - with a decisive effect on profit.</p>


<p>In our modern economy, software applications measure linear-feet equivalents of today's "thread and yarn." Such raw, furnished data, unlike cotton or wool fibers, begs translation, comparison, and analysis. Consequently, every team lead needs an agent by which to see, interpret and act on that data.</p>


<h2>The Dashboard - 3 Types for Business</h2>


<p>And that's what dashboards – imperative to business intelligence software – do. Of course, dicing and splicing that data constitutes a need for tailored dashboards, of which three types are recognized:</p>


<ul>
	<li><strong>Strategic</strong> - aggregates critical, overarching metrics, presenting a 10,000-foot view of a business.</li>
	<li><strong>Analytic</strong> - gathers and compares particular metrics across time and many variables, drilling down to actionable data per team.</li>
	<li><strong>Operational</strong> - monitors data in real time, alerting a team to any issues that need to be addressed.</li>
</ul>


<p>Regardless of a dashboard's purpose, it should reflect a company's particular needs and culture, displaying <a href="https://www.sisense.com/kpi-library/">Key Performance Indicators (KPIs)</a> based on a firm's high-level (and/or low-level) objectives. These KPIs will stand as quantifiable measurements of each goal; metrics, by any other definition.</p>


<p>That's important, because according to Sruthi Varanasi of <a href="https://reportgarden.com/google-analytics-metrics/" target="_blank" rel="noreferrer noopener" aria-label=" (opens in a new tab)">ReportGarden</a>, “A metric is a quantifiable measure that is used to track and assess the status of a specific business process.”</p>


<h2>Metrics – Lifeblood for a business; Heart of a Dashboard</h2>


<p>Suffice to say, metrics are the truest barometer of how your online business is doing.</p>


<p>Consider this: A 15 percent increase in conversions is just that, a successful trend. Subsequently, metrics serve as buoys that can keep your business sailing in deep water or warn you when shoals are near. A 21 percent dip in visibility over a month is just that, a falling trend, indicating you might need to revisit strategy and adjust – on the fly.</p>


<p>It stands to reason, then, that constructing a clean, uncluttered, incisive dashboard that represents key business intelligence metrics is equal parts science and art.</p>


<p>You want a dashboard whose widgets illustrate – at a glance – essential data from which sound business decisions can be made – whether those decisions concern the content of a webpage or the features of an actual product.</p>


<h2>Dashboard enABLEd! How to Determine Which Metrics to Track</h2>


<p>So, how do you decide which specific metrics should populate that dashboard from which you will extract actionable data? How do you identify those <a href="https://www.sisense.com/glossary/kpi-dashboard/">KPIs</a> for each business goal? Following these four steps will enABLE you (apology for the acronym within the acronym) to populate your dashboard with meaningful data:</p>


<ol>
	<li><em>A</em>pply S.M.A.R.T. methodology.</li>
	<li><em>B</em>ring the selection to the team.</li>
	<li><em>L</em>imit KPI assignment to three primary, overriding goals.</li>
	<li><em>E</em>liminate the urge to add more metrics to the dashboard.</li>
</ol>


<h4>1. Apply S.M.A.R.T. to each KPI</h4>


<p>For a basic example, if an overriding goal is to increase monthly recurring revenue (MRR), the questions to ask – and answer (more than yes/no) - to assess the validity of a KPI begin with:</p>


<ul>
	<li>Is a metric <strong>Specific</strong> to a goal? What needs to be accomplished and why?<br />
<i>We want to increase MRR to increase margins and subsidize a new product launch next year.</i></li>
	<li>Is it <strong>Measurable</strong>? What kind of historical change has been evident? How will we know the goal was reached? <br />
<i>According to historical analytics, we can feel confident that an MRR increase of between 3 – 5 % would be achievable.</i></li>
	<li>Is it <strong>Attainable</strong>? Are the resources readily available to achieve success? Is the goal reasonable? Is it likely to bring success?<br />
<i>We can ramp up social media promotion, launch a campaign, or otherwise put effort behind ramping up sales to drive revenue.</i></li>
	<li>Is it <strong>Relevant</strong>? How meaningful and worthwhile is the goal? In the current situation can we commit to its achievement? <br />
<i>Our competition has lost revenue, so more of the market is available to us. The more revenue generated, the more reward for us.</i></li>
	<li>Is it <strong>Timely</strong>? Is the goal ahead of the curve, or behind? What's the deadline for achieving it? What's the overall timeline set for adopting the goal?<br />
<i>After strategic planning, we can achieve a substantive bump in MRR over the subsequent quarter.</i></li>
</ul>


<p>So, your team devises this KPI: Increase MRR by 3% during Q2. What metric goes on the Dashboard? A monthly monitor of incoming revenue.</p>


<h4>2. Bring KPI selection process to the team</h4>


<p>Gain consensus on those metrics paramount to the team's and the company's success. Asking for a collective viewpoint not only helps distill the essence of paramount KPIs but also builds morale. Each team member gets some skin in the game.</p>


<h4>3. Limit KPI assignation to no more than three primary goals</h4>


<p><a href="https://segment.com/academy/intro/choosing-metrics-that-matter/" target="_blank" rel="noopener noreferrer">Segment's Analytics Academy</a> declares the purpose behind each solitary metric populating your dashboard should focus attention on a specific business process (goal!) that needs to be optimized. Using the sample KPI above, it could be one of three under an overarching goal to drive an increase of MRR.</p>


<h4>4. Eliminate unnecessary metrics</h4>


<p>The rule of thumb is to have no more than seven metrics displayed on any single dashboard because, after all, it functions as a quick-glance representation of a goal's status. Thus, its <a href="https://www.sisense.com/blog/4-design-principles-creating-better-dashboards/">design</a> should advance easy comprehension, simple updating, and clean navigation without secondary data distractions.</p>


<p>Your team should make hard decisions on which metrics to include. Consider: secondary data get in the way, conflating interpretation, overwhelming the reviewer. Fewer metrics are better metrics.</p>


<p>Each time you visit the dashboard, you should remember that KPIs keep your business strategy agile, fleet, responsive. Positive data dictates stability and steadiness. Negative data compels your team to adjust, adapt and provide alternatives.</p>


<p>An <a href="https://www.betterbuys.com/bi/dashboard-best-practices/" target="_blank" rel="noopener noreferrer">effective dashboard</a> illustrates this crucial data and discloses a course of action to take.</p>


<h2>Metrics on Dashboard: What Are My Choices?</h2>


<p>Once you've followed the ABLE steps to determine your KPIs, you're ready to populate your dashboard. At this point, you may ask, "What are metrics that achieve near-universal adoption by businesses?"</p>


<p>That depends on the purpose behind your team, the audience (your team? An executive?) that will be reviewing the dashboard, the "actionability" of the selected KPIs, and the type of visuals preferred.</p>


<p>Metrics for a marketing team might include tracking web traffic sources, incremental sales, social sentiment, conversion rate, and SEO keyword ranking. A sales team might want to monitor sales growth, product performance, average purchase value, and average profit margin.</p>


<p>A financial team can follow working capital, <a href="https://www.sisense.com/kpi-library/financial-kpis/debt-equity-ratio/">debt-to-equity ratio</a>, and current ratio. An e-commerce team might monitor <a href="https://www.sisense.com/solutions/executive/ltv/">customer lifetime value</a> (CLV), customer retention rate, <a href="https://www.sisense.com/dashboard-examples/customer-service/customer-churn-analysis/">customer churn analysis</a>, and monthly recurring revenue.</p>


<p>Other salient KPIs can address net profit, revenue growth rate, project schedule variance (PSV), and average revenue per customer. Because your KPI choices are ultimately subjective, the A.B.L.E. methodology can help your team judiciously arrive at which data would be most constructive to track and display.</p>


<h2>Vital Metrics on (Dash)Board: The Skinny</h2>


<p>As long as any <a href="https://www.sisense.com/blog/how-to-translate-business-goals-into-kpis/">KPI on your dashboard is based in company goals</a>, is relevant to the team behind achieving that goal, is attainable, measurable and remains timely, the dashboard itself should render keen data from which you can take incisive action to engineer successes -- as well as avert disasters.</p>


<p>Taking the time to apply the SMART methodology, bring in the team, limit primary goals and amount of KPIs assigned to each, and eliminate the urge to overpopulate a dashboard with secondary data, will help you select the most meaningful metrics for your business onto your dashboard.</p>


<p>Perform these steps. Pick your metrics. Build your dashboard. Mine your data.</p>


<p>Grow your business.</p>


<p><span style="font-size: x-large;"><strong><i>About the Author</i></strong></span></p>


<p>&nbsp;</p>


<p>&nbsp;</p>
<figure><img class="alignleft" style="float: left; margin: 4px 10px 0px 0px; border: 1px solid #000000;" src="https://cdn.sisense.com/wp-content/uploads/Keith-Craig-bio-pic-150x150.jpg" alt="Dashboard Design" width="100" height="88" /></figure>
<p>&nbsp;</p>
<p>Keith Craig is Content Marketing Manager for <a href="https://www.betterbuys.com/" target="_blank" rel="noopener noreferrer">Betterbuys</a>. He has more than a decade of experience using, researching and writing about business software and hardware. He can be found on <a href="https://twitter.com/keithalancraig" target="_blank" rel="noopener noreferrer">Twitter</a> and <a href="https://www.linkedin.com/in/keithalancraig" target="_blank" rel="noopener noreferrer">LinkedIn</a>.</p>
<p>&nbsp;</p>

Building a Dashboard? 4 Keys to Find The Most Telling Metrics

<p>Once upon a time business intelligence was the domain of IT teams and data analysts. IT was like the gatekeeper of analytics, with departments having to rely on them to prepare the data and churn out reports for their BI projects.</p>


<p>Not anymore. Over the past few years, as technology evolved, businesses have had a choice of solutions offering analytics and visualization capabilities that open up business intelligence to everyone.</p>


<p>It’s a good thing, right? All of a sudden, businesses can get insights from previously unexplored data sources and forgotten branches of the business are discovering new insights that could change the strategic direction of the company – helping you gain competitive advantage.</p>


<p>Now an organization will have different types of business intelligence users – which brings both benefits and challenges. One of the biggest challenges is getting a handle on who’s accessing and using the data that’s available.</p>


<p>Before data spirals out of control, let’s take a look at some of those business intelligence users who’ll be working with your precious business data.</p>


<h2>Benefits of embedding BI across the business</h2>


<p>The whole point of bringing business intelligence to the masses is to embed BI across entire organizations, making data a part of how they conduct their business and streamline operations, and to bring agility to the whole process.</p>


<p>One of the biggest bonuses of handing responsibility for data reporting over to individual business intelligence users is that it enables on-demand reporting, with ad-hoc reports allowing users to get more specific insights that answer a specific query.</p>


<h2>What are the challenges?</h2>


<p>With more power given to business users, there’s the risk of IT teams taking their eye off the ball. While business intelligence users no longer need the technical skills that they used to, IT are needed more than ever, and there’s even greater need for data governance to ensure the success of BI.</p>


<p>Gartner predicted that, in 2017, <a href="https://www.dataversity.net/self-service-business-intelligence-big-may-not-everyone/">less than 40% of Self-Service Business Intelligence initiatives would be governed sufficiently to prevent inconsistencies</a> that adversely affect the business. And BARC said something similar in the <a href="https://bi-survey.com/bi-deployment">recent BI-Survey</a> when it recommended a Data and Tool Governance framework, plus closer alignment between IT, the business, and business intelligence users to prevent the total loss of control over data.</p>


<h2>And the department most likely to benefit from business intelligence is…</h2>


<p>Actually, <a href="/blog/top-benefits-business-intelligence-marketing/">benefits of business intelligence</a> are being felt across departments, in all industries. Data has become central to all roles, although product and market-driven results are more noticeable when you’re looking at the bottom line.</p>


<p>For instance, analytics can guide the sales cycle, and help marketing teams target leads. In HR, business intelligence comes into its own, helping businesses manage one of their important assets – people. HR holds a gold mine of information that can help with compensation planning, or link people to financial performance, for example.</p>


<p>We’ve talked about business intelligence use cases in a previous post – with some <a href="https://www.sisense.com/blog/3-ingenious-use-cases-for-business-intelligence-tools/">ingenious ways organizations and individuals are leveraging the power of data</a>.</p>


<h2>Different types of business intelligence users</h2>


<p>We’ve drilled down into our very own data to find out who are the key players in business data reporting.</p>


<p><strong>1. The Data Analyst</strong></p>


<p>The data analyst, as the name suggests, lives and breathes data. A statistician by nature, they will drill deep down into data, looking for fresh insights that will be used to underscore business strategy. The role includes documenting all your business data, identifying patterns and creating reports and dashboards that will support the decision-making process.</p>


<p><strong>2. The Executive</strong></p>


<p>The CEO, of course, is at the helm of the business, driving the company’s success by improving <a href="https://www.sisense.com/solutions/operations-and-logistics/">operational analytics</a> efficiency and constantly looking for ways to reduce costs.</p>


<p>Meanwhile, Gartner once described analytics as the <a href="http://www.enterpriseappstoday.com/business-intelligence/gartner-taps-predictive-analytics-as-next-big-business-intelligence-trend.html">“combustion engine of business.”</a> So business intelligence and the CEO seems like a match made in heaven.</p>


<p>Business intelligence gives an organizational overview that allows the CEO to spot trends across the entire structure of the business – insights that can support:</p>


<ul>
	<li>Business growth</li>
	<li>Innovation</li>
	<li>Operational efficiency</li>
</ul>


<p><strong>3. The Business User</strong></p>


<p>Business intelligence users can come from across the organization. We often talk about two types of business users, the casual business intelligence user and the power user. The difference is that a power user has the capability of working with complex data sets, while the casual user will make use of dashboards to analyze predefined sets of data.</p>


<p>Generally the business user is often a manager, who is looking for ways to help a department operate more efficiently and more effectively. This type of business user is often untrained in using BI tools but is capable of picking up the basics of reporting and can use business intelligence successfully to report on business activities. Often they’ll undergo further training and become fully equipped to perform more in-depth analysis.</p>


<p><strong>4. The IT Team</strong></p>


<p>IT is another key player in the BI process. The IT user still plays a central role – maintaining the infrastructure and giving departments the tools that allow them to fulfill their own data requests – however, the role has become more rounded.</p>


<p>These days, IT will cooperate much more closely with departments and the business as a whole. It’s their job to ensure that business users are getting the most from data analytics, and they’re also a key part of data governance and <a href="https://www.sisense.com/product/govern/">BI security</a>. They’ll also bridge the gap between IT and business, helping increase BI adoption across the board.</p>


<h2>Take BI by the Reins</h2>


<p>It’s all too easy to lose control of business intelligence projects and to become bogged down in data. One way to take back control is to log all the BI users. This will become essential in governing your data but it’s also a useful way to identify gap areas where business intelligence is weak or isn’t producing the results you want.</p>


<p>A great suggestion is to classify business intelligence users based on how much training they’ve received or their analytical capabilities. This approach will help take your analytical maturity to the next level.</p>

4 Business Intelligence Users You’ll Encounter in Any Company

<p>The proliferation of <a href="https://www.sisense.com/product/business-teams/">business intelligence platforms</a>, tools, and software stacks is dizzying. To help people who are currently in the market for business intelligence tools or anyone who wants to gain a clearer understanding of the field, we've created this comprehensive list of online resources you can use to compare business intelligence products and vendors. These include user reviews and ratings, analyst commentary and professional guides that can help you self-evaluate. As we are a vendor in the space ourselves we've included mostly third-party reviews and reports for an unbiased perspective on BI tools. However, if you do want our own view of the marketplace and best practices for vendor selection, you might want to check out some of these white papers:</p>


<ul class="sisense">
	<li><a href="/whitepapers/business-analytics-evaluation-guidelines/">4 Steps To Successfully Evaluating BI Software</a></li>
	<li><a href="/whitepapers/embedded-analytics-guide/">The Ultimate Guide to Comparing Embedded Analytics Solutions</a></li>
	<li><a href="/whitepapers/sisense-vs-alternatives/">Sisense Compared to the Alternatives</a></li>
</ul>


<p>[line]</p>


<h2>BI Software Comparison Websites</h2>


<p>The following business intelligence software comparison sites enable you to compare between business analytics tools based on expert reviews and - more importantly - ratings and reviews from actual users.</p>


<h3>Gartner Peer Insights</h3>


<div class="wp-block-image">
<figure class="alignright"><img src="https://cdn.sisense.com/wp-content/uploads/Gartner1-770x388.png" alt="Gartner" /></figure>
</div>


<p>This website is operated by revered analyst firm Gartner. While it also focuses on reviews by actual users, these are carefully guided by the review format to provide a highly detailed view of each product and its strengths and shortcomings within the scope of each deployment.</p>


<p><a href="https://www.gartner.com/reviews/market/analytics-business-intelligence-platforms">Visit website</a></p>


<h3>G2 Crowd</h3>


<div class="wp-block-image">
<figure class="alignright"><img src="https://cdn.sisense.com/wp-content/uploads/g2crowd-770x265.png" alt="g2 crowd" /></figure>
</div>


<p>G2 Crowd is a leading business software review platform and contains over 375,000 user reviews for various B2B applications. The site also publishes grid reports that allow users to easily compare between the leading tools and evaluate their advantages, <a href="https://www.g2.com/products/sisense/pricing">pricing</a>, etc.</p>


<p><a href="https://www.g2.com/categories/analytics-platforms" rel="nofollow">Visit website</a> or download the <a href="https://www.sisense.com/whitepapers/compare-top-bi-vendors-using-g2-crowd/">G2 Crowd BI Implementation Index</a>.</p>


<h3>Capterra</h3>


<div class="wp-block-image">
<figure class="alignright"><img src="https://cdn.sisense.com/wp-content/uploads/capterra-770x270.png" alt="Capterra" /></figure>
</div>


<p>Capterra is a free service that helps businesses find the right software, and business intelligence is one of the 300+ categories the website covers. You can use Capterra to browse and compare business intelligence software, read buying guides, blogs, user reviews &amp; <a href="https://www.capterra.com/p/86955/Sisense/" target="_blank" rel="noopener noreferrer">pricing</a> and request a personalized match from Capterra's experts according to specific business needs.</p>


<p><a href="https://www.capterra.com/business-intelligence-software/" target="_blank" rel="noopener noreferrer">Visit website</a></p>


<h3>GetApp</h3>


<div class="wp-block-image">
<figure class="alignright"><img src="https://cdn.sisense.com/wp-content/uploads/GetApp-770x255.png" alt="GetApp" /></figure>
</div>


<p>GetApp is an extremely comprehensive directory of reviews for business software, SaaS and cloud applications. In terms of BI applications, it contains listings and reviews for dashboard software as well as big data, <a href="https://www.sisense.com/glossary/data-mining-basics/">data mining</a>, <a href="https://www.sisense.com/glossary/predictive-data-analytics/">predictive analytics</a> and other types of analytical tools. Within each category, you can view product reviews and <a href="https://www.getapp.com/business-intelligence-analytics-software/a/sisense-prism/pricing/" target="_blank" rel="noopener noreferrer">pricing </a>as well as compare two or more products based on their ratings in various criteria. Additionally, the site's "Lab" feature offers research, reports, and insights into industry trends.</p>


<p><a href="https://www.getapp.com/business-intelligence-analytics-software/business-intelligence/" target="_blank" rel="noopener noreferrer">Visit website</a></p>


<h3>Finances Online</h3>


<div class="wp-block-image">
<figure class="alignright"><img src="https://cdn.sisense.com/wp-content/uploads/Finances_Online-770x327.png" alt="Finances Online" /></figure>
</div>


<p>Despite its name, this website actually offers a comprehensive directory of various business software (including non-finance related). You can use it to "compare products, read customer reviews and reports from experts to easily find a solution that will let your company grow". When it comes to business intelligence the site features reviews, articles and product comparisons.</p>


<p><a href="https://business-intelligence.financesonline.com/" target="_blank" rel="noopener noreferrer">Visit website</a></p>


<h3>TrustRadius</h3>


<div class="wp-block-image">
<figure class="alignright"><img src="https://cdn.sisense.com/wp-content/uploads/trust_radius-770x324.png" alt="Trust Radius" /></figure>
</div>


<p>TrustRadius defines itself as a community of professionals sharing software reviews, software discussions, and best practices, with a strong focus on user reviews - which are often long and detailed, enabling you to get a good picture of the functionality of each product and its pros and cons.</p>


<p><a href="https://www.trustradius.com/business-intelligence-bi">Visit website</a></p>


<h3>SelectHub</h3>


<figure class="wp-block-image"><img src="https://cdn.sisense.com/wp-content/uploads/selecthub-screenshot.png" alt="Screenshot SelectHub website" /></figure>


<p>According to SelectHub's mission statement, its one goal is to make technology investments more impactful by reducing the pains of evaluating and onboarding while increasing ROI. The site provides leaderboards, reviews, <a href="https://www.selecthub.com/business-intelligence-tools/sisense/" target="_blank" rel="noreferrer noopener" aria-label=" (opens in a new tab)">pricing</a> and ratings for various types of enterprise software tools, including business intelligence.</p>


<p><a href="https://www.selecthub.com/business-intelligence-tools/">Visit website</a></p>


<h3>IT Central Station</h3>


<figure class="wp-block-image"><img src="https://cdn.sisense.com/wp-content/uploads/ITCentralStation-Screenshot.png" alt="IT Central Station" /></figure>


<p>While at first glance this may look like yet another directory of reviews for business software, the place where IT Central Station really shines is in its active user base and communities - with thousands of users and new reviews being constantly posted, commented on and updated. Reviewers undergo multi-tiered authentication before their posts are approved to appear on the site. In addition to reviews, the site provides comparisons, product descriptions and FAQs.</p>


<p><a href="https://www.itcentralstation.com/categories/business-intelligence-bi-tools">Visit website</a></p>


<h3>Software Advice</h3>


<figure class="wp-block-image"><img src="https://cdn.sisense.com/wp-content/uploads/software-advice-screenshot.png" alt="Screenshot of Software Advice website" /></figure>


<p>Software Advice aims to simplify the software selection process with the promise of matching buyers with the right software in 15 minutes. In addition to the usual reviews, ratings, and leaderboards, the site also offers to provide a free telephone consultation to interested prospects.</p>


<p><a href="https://www.softwareadvice.com/bi/" target="_blank" rel="noopener noreferrer">Visit website</a></p>


<h3>DiscoverCloud</h3>


<figure class="wp-block-image"><img src="https://cdn.sisense.com/wp-content/uploads/discovercloud-screenshot.png" alt="Screenshot of DiscoverCloud website" /></figure>


<p>DiscoverCloud was founded in 2012 in order to help business owners find the tools and information required to grow their business. Its Marketplace features an ever-growing number of SaaS solutions and outsourced services, currently listing over 3,000 applications. The site provides reviews, ratings, and market landscape analysis matrices.</p>


<p><a href="https://www.discovercloud.com/categories/business-intelligence" target="_blank" rel="noopener noreferrer">Visit website</a></p>


<h3>Technology Advice</h3>


<div class="wp-block-image">
<figure class="alignright"><img src="https://cdn.sisense.com/wp-content/uploads/Technology_Advice-770x252.png" alt="TechnologyAdvice" /></figure>
</div>


<p>Technology Advice focuses on product comparisons and has a rich library of original research, white papers, and studies. In addition, you will find the expected reviews from software users, as well as an automated recommendation system.</p>


<p><a href="https://technologyadvice.com/business-intelligence/">Visit website</a></p>


<h3>Solutions Review</h3>


<div class="wp-block-image">
<figure class="alignright"><img src="https://cdn.sisense.com/wp-content/uploads/solutions_review-770x229.png" alt="solutions review" /></figure>
</div>


<p>Unlike its counterparts, this website does not focus on user reviews but does serve as a useful hub for industry related news, commentary and best practices. The site also published periodical buyer's guide which can help comparing BI vendors according to their feature sets.</p>


<p><a href="https://solutionsreview.com/business-intelligence/" target="_blank" rel="noopener noreferrer">Visit website</a></p>


<h3>Software Insider</h3>


<div class="wp-block-image">
<figure class="alignright"><img src="https://cdn.sisense.com/wp-content/uploads/software_insider-770x300.png" alt="Software Insider" /></figure>
</div>


<p>This site uses semantic technology to deliver insights from data driven articles, visualizations, and resarch tools. They allow ou to create a shortlist and filder by everyting from platform type and analysis features to business size and supported languages.</p>


<h3>BetterBuys</h3>


<div class="wp-block-image">
<figure class="alignright"><img src="https://cdn.sisense.com/wp-content/uploads/BetterBuys-770x262.png" alt="BetterBuys" /></figure>
</div>


<p>BetterBuys lets you browse vendor and compare top solutions. They also offer a free comparison guide that breaks down the top BI vendors in an easy to use report.</p>


<p><a href="https://www.betterbuys.com/bi/reviews/">Visit website</a></p>


<div class="wp-block-image">
<figure class="alignright"><a href="https://www.sisense.com/whitepapers/sisense-vs-alternatives/"><img src="https://cdn.sisense.com/wp-content/uploads/Sisense_Compared_To_The_Alternatives.jpg" alt="Sisense vs Alternatives " /></a></figure>
</div>


<p><br />
<br />
[line]</p>


<h2>Analyst Reports</h2>


<p>If you've had your fill of user reviews, you might want to check out some of these analyst reports and resources. All of them offer both a comparison between different products, as well as serving as valuable resources for understanding general concepts and considerations within the BI and analytics space.</p>


<h3>Gartner Magic Quadrant for Analytics and Business Intelligence Platforms, 2018</h3>


<div class="wp-block-image">
<figure class="alignright"><img src="https://cdn.sisense.com/wp-content/uploads/Gartner1-770x388.png" alt="Gartner" /></figure>
</div>


<p>The quintessential report for all things business intelligence, Gartner's Magic Quadrant needs no introduction and is usually considered the most well-researched and highly-regarded point of reference to understand the layout of the BI landscape. You can get a copy of it <a href="https://www.sisense.com/gartner-magic-quadrant-business-intelligence/">right here</a>.</p>


<h3>The BARC BI Survey</h3>


<div class="wp-block-image">
<figure class="alignright"><img src="https://cdn.sisense.com/wp-content/uploads/BARC_001-770x365.png" alt="" /></figure>
</div>


<p>The Business Application Research Center (BARC) annually surveys thousands of business intelligence software users, gathering feedback on the tools that they use across a wide variety of KPIs meant to measure each tool's business value, user experience, customer satisfaction and competitiveness. The report is a very useful aggregation of the opinions of actual users who work with the product in their day-to-day.</p>


<p><a href="https://bi-survey.com/" target="_blank" rel="noopener noreferrer">Visit website</a></p>


<h3>Dresner Wisdom of Crowds Business Intelligence Market Study</h3>


<div class="wp-block-image">
<figure class="alignright"><img src="https://cdn.sisense.com/wp-content/uploads/dresner-770x297.png" alt="Dresner" /></figure>
</div>


<p>The Wisdom of Crowds Report, in addition to providing a set of detailed criteria for evaluating BI vendors, is also based off of the feedback of thousands of actual users who participated in the study and ranked the leading BI tools based on their own experience. You can <a href="https://www.sisense.com/whitepapers/wisdom-crowds-business-intelligence-market-study/">download the report here.</a></p>


<h3>Forrester Wave Reports</h3>


<div class="wp-block-image">
<figure class="alignright"><img src="https://cdn.sisense.com/wp-content/uploads/Forrester-770x372.png" alt="Forrester" /></figure>
</div>


<p>Another highly esteemed analyst firm, Forrester periodically releases its Wave Reports focusing on specific segments in the market, e.g. cloud business intelligence or <a href="https://www.sisense.com/why-sisense/">agile BI</a>. These reports are available to download from the <a href="https://go.forrester.com/">Forrester website</a>.</p>


<h3>Butler Analytics</h3>


<figure class="wp-block-image"><img src="https://cdn.sisense.com/wp-content/uploads/butler-analytics-screenshot-small.png" alt="Screenshot of Butler Analytics Website" /></figure>


<p>While analyst Martin Butler does not release annual reports, his website is definitely worth visiting. In addition to specific product comparisons and lists, the blog is worth checking out for a refreshing and often critical voice of the BI industry and the oft-exaggerated claims made by vendors. You can <a href="https://www.butleranalytics.com/">visit Butler Analytics here</a>.</p>


<div class="wp-block-image">
<figure class="alignright"><a href="https://www.sisense.com/whitepapers/sisense-vs-alternatives/"><img src="https://cdn.sisense.com/wp-content/uploads/Sisense_Compared_To_The_Alternatives.jpg" alt="Sisense vs Alternatives " /></a></figure>
</div>