What data do users need?

The first thing to note here is that the "user" in the title refers to the user of the data, or the data demander, who are often not the external users for the website or the enterprise. The consumers of the data are usually all departments and fields within the company personnel.

Why do we ask this question, in fact, we often encounter such a situation: the company's top complain that the report can not see useful things, is not able to organize the statements (so busy below), but how? Finishing or what they need in the end the data (well, the high-level demand is generally not say, we should try to figure out); at the same time various departments are constantly mention all kinds of data needs, often their needs More specific, sometimes scary, requiring each user's critical operation (to test the server's time). The data department is in an environment where there is such a need for data that there is a lot of diversity, so the time to test the data staff is up. Can we meet all our needs?

Well, first of all, to answer the question after we try to figure out the result of "holiness", what data do the boss or senior need? In fact, very simple, they just want to know how the company's overall situation, so we only need to provide aggregated goals and KPI data, do not need too much, 2-3 statements, about 10 indicators enough to show the company's overall, But in fact, the first thing to do is to have a clear understanding and definition of the company's goals and KPIs.

The main focus of staff: decision-making level

Although the key focus group of KPIs and targets are locked in the company's decision-making level, each employee in the company should really care about the achievement of the company's goals and performance of the KPIs, as the goals and KPIs are the most effective for objectively evaluating the status and benefits of the company way. But often there is a difference in how each department cares about the KPIs it is targeting, so the data is broken down by the way that the data needs to be addressed to different KPIs of various departments.

The company's policymakers may want to see the combined data for these goals and KPIs, but if we provide this type of aggregated data to all of our users, maybe other users will only be able to look at the data far away and do nothing . So we need to give different users different types of different levels of data, because we have to do is to allow each data consumer based on the data Take Actions, and one of the important piece is the data segmentation.

We can subdivide website analysis reports and indicators from multiple perspectives. Each company chooses its own subdivision module according to its own operation type. Of course, the most common subdivisions mentioned here are: content, user And the source, which is the way Google Analytics block.

The main focus of staff: product operations

In spite of the ever-changing forms of the Internet, however, the Internet is still primarily an information service provider's role and ultimately the content, so content is its core competency for websites, as well as for web analytics, So bear the brunt of the breakdown of the content. Previous article - site page metrics and segmentation, some of the site's content metrics and several subdivisions introduced, no matter what indicators or subdivisions to evaluate the content, and ultimately our purpose Is to distinguish between good and bad content, control the operation of the product, so as to maintain or improve the website content.

Content analysis of the results of subdivision will undoubtedly provide a valuable reference for product operation or website operations, clear what is the core content need to grasp, what content needs to be improved. At the same time with some special indicators can also guide the details of the improvements, such as a Pageviews high but Avg. Time on Page short, high Exit Rate page obviously not enough content in the content, but the title or profile information is sufficient Eye-catching, then the direction of improvement can be identified as a way to improve the content; if your site offers e-commerce services, sales (goals) and conversion rates (KPIs) for each or each product segment will get you Better targeted selection of products and modes of operation.

The main focus of staff: user experience, sales

We generally use the user's environment (network, devices, systems and clients, etc.), demographic information (gender, age, geographical etc.), types of user behavior (trends in use, loyalty, value created, etc.) Class data and metrics subdivide users. In the current trend of "user-centered" prevalence, the user is not behind the content of a bit wrong? All of the site is to meet the needs of users, including the provision of all content, but in fact, the user analysis of data analysis and analysis of the content is not common, especially users are also subdivided, the reason is very simple, the content or the product is Can grasp by themselves, and the user can not, so especially based on the data of user behavior analysis, to put it a lot, but it is actually done or applied to practice is not much.

But one must have the user to analyze the data support, that is, user experience design and optimization. For the user experience design, the goal is to meet all the user's habits, so comparing and optimizing all kinds of users in different environments and habits of the data can play a significant role in improving the user experience; and If your website product needs to be sold, user behavior analysis works well for personalized product sales and referrals, just as opposed to the goal of user experience, which is primarily intended to meet the needs of each and every category Preference for each user's needs.

The main focus of staff: marketing

In fact, for website analysts, the source of data analysis is certainly not unfamiliar, many sites will focus on the analysis of this piece of effect, including SEO and SEM, have all developed into a very professional field. Website analysis tools generally distinguish between direct sources, search engines, external websites and promotion sources. In fact, we can use some effective ways to narrow down these channels, including social media, cooperation websites, advertising direct mail Etc. Through the breakdown of these sources to observe the quality of the traffic brought by each channel (performance on the target and KPI indicators), we can see the pros and cons of each marketing channel so as to provide reference for effective promotion.

In fact, there is another piece - offline channels. We often think that the effects of promotions or advertisements on offline television, newspapers and the like are very difficult to monitor with data. However, as long as we seek some solutions, these are also achievable. For example, Offline is a great tool for monitoring offline phone marketing channels. Through the monitoring and analysis of offline channels, we know more about the effect of offline promotion and its impact on online promotion, and ultimately guide and promote the staff to more effectively arrange and implement a complete package of promotion plans.

The main focus of staff: technology, user experience

If your website is not just a few simple pages, but a huge complex system, which provides a wealth of features and applications, then we need to do a type of analysis is the analysis of the various functional points. Before analyzing the common features of several websites in the topic of "Making users find the information they need more easily": site search, navigation design and content recommendation, we can all use these special functions to get data and set special Indicators to analyze their results.

Both the technology and UX teams need to pay attention to the realization and optimization of these functions. Data is the most effective way to evaluate these functions because these functions affect the user experience and satisfaction. A really good website needs to grasp each The realization of every detail of the function.

The people mentioned above cover almost every area of ​​each company, but in fact there is still a missing part of our own - data analysts. In fact, for data analysts, they need to grasp all the data, from the global goals and KPIs to various types of breakdown indicators, as well as various types of functional point data. But these are far from enough. Data analysts must give full play to their strengths and design and build various analysis models that can not only objectively evaluate the company's key business and operating conditions, but also have the summing-up effect , But also can find some potential commercial demand points for the development of the company to provide possible direction and decision-making basis, play a predictive role.

The analysis models are mainly divided into two categories, one is the quantitative analysis model, this has been introduced in my blog, including the funnel model of critical path analysis, user evaluation model based on user behavior analysis, of course, including data mining User interest discovery, content pattern matching, and personalized recommendation model based on them all achieve the prediction effect to a certain extent.

The other is the qualitative analysis model, including the target market research, user-centered research and analysis of competitive advantage. Of course, user research and user experience may now be relatively more done, through the online questionnaire, usability experiments, real-time access to research, combined with some visualization of the heat map, mouse movement monitoring and other tools to assess the user is using the site is the overall feeling And satisfaction, this more user-friendly approach to analytics will result in many new ideas for website and product optimization.

In fact, most of the website analysis tools and BI reporting tools will provide custom Dashboard features so that users can focus on their own indicators, reports and charts integrated on the same Dashboard for easy daily observation and analysis. Originally this was a cool feature, since it was nice to have my dashboard slightly dimmed, but in reality the functionality did not work as well as expected, or the user did not get used to customizing the Dashboard (Of course, there are data organization and associated restrictions and reporting tools ease of use issues), but as a data provider, we customize the public Dashboard at the same time, if necessary, but also to help some specific needs of the community Customize your custom Dashboard.

A great custom dashboard not only organizes data reasonably well, but also visualizes the data so that the analysis of the data does not need to be so tedious. Yes, perhaps users will fall in love with the data. Custom Dashboard at the same time can effectively control the data permissions, the Dashboard which will open the appropriate indicators and statements to the user, thus shielding some sensitive data, the confidentiality of data for the data sector is also an important task.

I do not know reading the entire article will not feel a bit empty, there is no substantive content or practical analysis, but in fact, this article took me a long time to summarize and think, sort out the entire data to provide a viable program of ideas , Hoping to provide a systematic and comprehensive data organization and solutions. Using data as a clue to run through every corner of the enterprise, we can truly establish a Data Driven culture so that the data can not only be displayed simply To meet the different needs of all types of personnel, and ultimately rely on data to improve the efficiency of enterprises in all areas and the effectiveness of the implementation.

»This article uses the BY-NC-SA protocol, please indicate the source: Site Data Analysis» "What data users need? "