Data visualization can help users better understand and use data, and excellent data visualization depends on excellent design. Starting from his own experience, the 14 chart styles that designers need to understand in the process of enterprise data visualization in combination with specific cases, and conducted a brief analysis for everyone to learn together.
1. What are the big data application scenarios for enterprises?
In recent years, by focusing on the collection of big data, big data has taken up an increasing proportion of enterprise operations. Many companies have begun to use big data analysis to optimize their business, promote corporate advertising campaign measures and the effectiveness of advertising distribution, and accelerate the decision-making speed of corporate management and improve sites. In addition to accelerating the pace of our business in various scenarios where enterprises operate, data visualization also plays a great role in predicting future profits of enterprises.
Next, let me briefly introduce what are the application scenarios of enterprise data visualization? So for different application scenarios, designers need to understand which diagrams can solve problems in a convenient way.
1. Perform product recommendation scenarios based on data analysis results of customer behavior
The reasons why products can be accurately recommended are mostly based on customer information, transaction history, purchase process, and behavior after visiting or purchasing the same product. After data analysis of customer transaction behavior, it is possible to predict customer preferences and propose products suitable for them to establish strong connections with customers. When customers are searching for the target product, they will continue to recommend other related products.
Product recommendation is community marketing based on customer social behavior analysis. By collecting data on people who have a common interest in the same brand and product and analyzing their characteristics, hobbies and preferences, it is possible to market to those potential product users. Companies can customize product recommendations more accurately by analyzing customer behavior data.
Data visualization can be used to display the data together, and all departments can find the data analysis results they need from the large data screen to optimize their business decisions. For example, in the case in the figure below, the large screen shows the overall sales behavior of the company.
2. Consumer portrait scene based on customer reviews
Customer word-of-mouth data has huge potential value. It is the value data for the company to improve product design, pricing, operational efficiency, customer service, etc. This is also the focus of product innovation.
Customer reviews include valuable customer feedback regarding product satisfaction, logistics efficiency, customer service quality, and customer experience and expectations regarding product appearance, function, and performance. It enables the company to improve its products, operations, and services, and establish customer-centric product innovation efforts.
3. DSP advertising optimization scenario based on data analysis
DSP is a service for advertisers who want to increase the cost-effectiveness of advertising. For example, data visualization has a function that can rely on the analysis of the data platform to locate customer portraits with similar behaviors to past purchasers.
In this way, real-time changes and optimizations can be made to the advertisement to obtain effective clicks. At the same time, effective clicks also depend on the time point when the advertisement is clicked, the number of clicks, etc. The advertising plan will be able to determine goals and optimize the quality of advertising based on the analysis of delivery effect data and the duration of advertising clicks.
4. Product pricing scenarios based on large-screen data analysis
The rationality of product pricing requires data testing and analysis. First, we understand the sensitivity of customers to product prices, classify them, and measure the direct response and tolerance of product groups with different price sensitivity to price changes. Through the display and analysis of large data screens, we can provide a basis for product pricing decisions.
5. Prediction scenarios for canceling orders based on customer behavior
The analysis of customer data through the data visualization large screen shows that we can find the negative effects of concentration from user behaviors. For example, there are many customer complaints in business operations, negative customer ratings, and customer purchases are significantly reduced. Through the discovery of negative effects, decision-making improvements can be made to the follow-up behavior of the enterprise’s business.
6. Analyze external situation scenarios based on market trend data
Predict the evolution of external conditions, such as data from product and market competitors’ promotional activities, people’s emotions on social media, and monitor online public opinion to help companies respond to changes in the environment and keep up with the changes in the market. For example, one in the following figure Large screen for public opinion monitoring of tourist attractions.
7. Product lifecycle management scenarios based on IoT data analysis
Barcodes, QR codes, etc. can uniquely identify products, sensors, wearable devices, smart sensors, video capture, and augmented reality. Or use other technologies to collect and analyze product life cycle information in real time. Each product link tracks and collects product usage information for product lifecycle management.
In addition to the above situations, big data is also used in many situations. With the further development of big data, data analysis in business situations becomes more and more necessary for enterprises.
Two, 14 convenient ways to visualize charts
Next I will introduce 14 data visualization charts , from simple to complex, each example has its own unique function and how and when to use them to get the best results .
1. Indicator Chart
Metrics are very useful when you want to immediately understand how your business is performing on specific KPIs. By incorporating a simple visualization tool “gauge indicator”, you can quickly see if you are above or below the target, or on the right track. By coloring in red or green and using the up and down arrows, the indicator is a more effective way to visualize data.
2. Line chart
The reason why line charts are very popular is that they have a wide range of uses in business, because they can quickly and concisely show the overall trend , and are not easily misunderstood. It is especially suitable for displaying trends of different categories in the same time period for easy comparison.
3. Bar graph
Bar charts are great for comparing different values, especially when classifying them into different colored categories. In order to distinguish the difference between a bar chart and a line chart, let us use the same information used in the line chart above for a new visualization in the bar chart
4. Column Chart
When comparing different values side by side, a bar chart is usually used. Column charts can also show changes over a period of time, but when you want to focus on the overall numbers rather than trends, it makes sense to use a column chart (line charts are more effective when viewing trends).
For example, a graph is needed to show the total daily visits and sessions of the website. Since the daily numbers do not change much, the column chart does not give any clear insights, but the information that the column chart can provide is the number of website visitors each day.
5. Pie Chart
Pie charts can help you tell immediately how each value constitutes a whole. It’s more intuitive than just listing the percentage of the total 100%. For example, a pie chart can show which product series have the highest share of total potential customers.
If you want to use pie charts effectively, you should have 6 or fewer categories. When the number of categories is 6 or more, the pie chart becomes too crowded and the values are too difficult to understand. Please refer to the following bizarre pie chart comparing the populations of various states in the United States as evidence that the pie chart provides little information:
6. Area chart
Area charts are useful because you can see the total amount and proportion of each category.
Pivot tables are not an intuitive method of data visualization, especially when you want to quickly extract key values while viewing the exact numbers (rather than trying to find trends), if you cannot use self-service BI that automatically performs this operation Tool, then this is a good way to visualize data.
8. Scatter chart
The scatter chart is divided into several categories according to the color of the circle. The size of the circle indicates the size of the data and is used to visualize the distribution and relationship of two values.
For example, in the scatter chart below, each product line is visualized by the number of units sold and the revenue generated, and its value is displayed as the size of a circle. It also sorts by gender (hover over the circle to see the original product name).
In this example, it can be determined that the most frequent (and most profitable) customer is currently male, which may (for example) lead to more sales for male customers based on business priority.
9. Bubble chart
A bubble chart is similar to a scatter chart, where the size of the circle indicates the size of the value. But the difference between bubble charts and scatter charts is that many different values are combined in a small place, and only one value is displayed for each category. This is a useful way to prove that a small number of categories are very important, while a large number of other categories are not.
The tree diagram helps to show the hierarchical structure and comparison values between categories and subcategories. You can instantly predict the most important areas in general while viewing detailed information.
To do this, color-coded rectangles are nested together and weighted to reflect the overall share. The tree diagram below shows the value of different marketing channels, separated by country/region. At a glance, you will find that AdWords is the most effective channel, but of all channels, the United States is also the most valuable place.
11. Radar Chart
A radar chart (or rose chart) is a type of pie chart. However, instead of displaying the proportion of each value in the whole by the size of the angle, all sectors have the same angle, and the value is displayed by the distance from the center of the circle. In this year’s new crown epidemic, the rose chart was used extensively to show the visualization of world epidemic data.
12. Area chart/scatter chart
This method of data visualization makes it possible to immediately see which geographic location is most important to the business. The data is displayed as colored dots on the map, and the size of the circle indicates the value.
For example, the map below shows website visitors by location, and white shows customer conversion rates (the brighter, the higher the customer conversion rate).
This type of data visualization is very useful because it provides you with two important pieces of information at a glance: the most attractive places in the world for tourists, and the most valuable places for tourists in the world. Such insights can reveal weaknesses in your marketing strategy in a matter of seconds.
13. Funnel chart
The funnel chart is a very special visualization method that shows that as customers browse the sales funnel, the value decreases. The advantage of this graph is that it can increase the customer conversion rate at each stage, so you can quickly see where customers are lost during the sales process. The following channel diagram shows the number of people at various stages of demand, from the first visit to the website to the end of the final sale, the number of people who went through each touch point
14. Fisheye/Cartesian distortion
Finally, this is not a data visualization style, but a useful additional function, you can zoom in on the details of more complex visualization data (such as dynamic model diagrams and bubble charts). Move the cursor over the graph to zoom in on the fisheye-like area, so you can dig and get more details as needed.
No matter which data visualization you choose, it needs to be accurate and effective, and the software used must be able to effectively access the data, and you can use powerful external visualization tools to better improve the results.
Without a powerful and flexible platform, even if you end up with beautiful results, your data will eventually be built on a very unstable foundation.