How to use website performance optimization to drive product experience improvement

（1） Performance and experience issues caused by the gap between product and user experience

As the Internet dividend fades, product functions and user experience design become more and more involved. There is a gap between the design of product functional logic and the understanding of users when using it. A large number of flash sales, promotional activities, and UGC content make the product logic more complex. Even if various guidance and explanation documents are provided, users still need time to understand and cultivate usage habits. At the same time, in order to further enrich the functional modules, a large number of rich media, third-party components, and customer advertisements are constantly being added, resulting in excessive and unreasonable external cooperation content, increasing system load, and dragging down product performance. The ultimate cost is to sacrifice a certain level of website performance and user experience.

（2） Performance and experience issues brought about by the complex network environment

As is well known, there are various primary and secondary operators in various parts of the country, which greatly increases the complexity of the national network environment. Due to slow infrastructure updates and numerous sudden human problems of operators, frequent IDC failures occur. Enterprises can only appease users and lie flat waiting for repairs, and the investigation time for these problems can only be resigned to fate. At the same time, the vast geographical distribution, scattered user distribution, and personalized network access methods make the access network complex, and enterprises cannot effectively estimate the user usage environment. Even with the help of widely distributed data centers and multi line BGP access, it is still difficult to solve network environment problems, which further exacerbates the difficulty of optimizing the network environment and makes the actual user experience more difficult to predict.

（3） Performance and experience issues caused by significant differences in PC environments

As the country with the largest internet user base in the world, behind the massive user base in China is a huge difference in user hardware configuration. Some may use i9-11900K+RTX3080 Ti to watch 4K high-definition live videos on Bilibili, while others may use the Millennium released Pentium 4 and integrated graphics card to browse text news on portal websites. This results in different groups with different browser versions, their own rendering mechanisms, and differences in local host performance, resulting in user experience differences such as abnormal access, slow speed, and local resource consumption. Faced with this situation, how to understand the actual experience of the vast number of users, balance or evaluate the differences in user experience, and make trade-offs among them has become a challenge that every website operation and development must face.

（4） The issue of ensuring system availability brought about by the sequelae of pursuing iteration speed

Due to the rampant competition on the internet, products have to selectively ignore product architecture and stability when it comes to the choice between feature window period and fine tuning. The lack of rigorous architecture, business development exceeding the support capabilities of the architecture, resulting in system overload, system crash, response timeout, and other issues, are caused by many factors:

Firstly, the business iteration speed is very fast, and intrusive monitoring methods cannot be implemented in a short period of time. However, when a business system malfunctions, it needs to be quickly sensed;

Secondly, development resources are scarce or uncooperative, and infrastructure related monitoring cannot directly reflect business issues, resulting in high implementation costs for application monitoring.

Finally, if the application calls a third-party API interface, the availability of the third-party API interface cannot be guaranteed, and if a malfunction occurs, it cannot be responded to and handled in a timely manner.

Looking at the disassembly, we may feel that these are all single point issues, but if there is a chain reaction after the business volume increases, these problems will be superimposed and amplified, directly affecting the user experience.

（5） Lack of monitoring methods from a user perspective leads to a relatively passive response to customer complaints

Although the product features undergo various tests when launched, the operations team continues to monitor user usage. But for the operation and maintenance team, it is only after the customer complains that they know that the system has encountered a problem, which is very passive in response, and even abnormal recurrence and positioning of the problem may take a day, seriously affecting the NPS; Most common monitoring methods also start from their own perspective and cannot directly reflect user issues.

So, in the face of so many influencing factors, how should we test our website from the perspective of real users, quantify website user experience, and identify website performance bottlenecks? Here, we take marketing activities in the e-commerce industry as an example. With increasingly fierce competition, promotional activities such as Double Eleven and 618 have become important annual marketing activities in e-commerce and other pan trading industries. However, the influx of a large number of users in a short period of time can cause website loading delays, or business service lag and other issues that affect the user experience.

Specific issues include:

Before going online, it is impossible to simulate real users and test the actual product experience when peak users have high concurrent access.

There is no accurate evaluation of the user's actual browsing path, unable to locate the conversion bottleneck, and unsure of how to optimize it.

During the promotion phase, product information updates are frequent, and after updates, complaints from users from various regions such as "product images cannot be loaded" and "page opening is slow" are often received.

The performance of peer competition activities cannot be obtained, and it is impossible to understand the changes in the marketing trend of competitors.

In the past, the above issues have been difficult to solve, and specific reasons for their difficulty include:

Although there are methods such as task walls, the operation and maintenance team cannot find enough real traffic that meets actual needs for product user experience testing, and purchasing related traffic is time-consuming and expensive.

The marketing promotion is very urgent, and the delivery time left for the R&D team is relatively limited. Wanting to add relevant invasive probes for monitoring can slow down product delivery and potentially affect product stability.

The operation and maintenance team is unable to proactively test related issues, resulting in problems that can only be discovered during actual user experience and can only be passively resolved. But the recurrence of problems and fault location may drag down the entire operation and maintenance team, leading to an indefinite extension of repair time.

Therefore, the operation team and maintenance team need a product or solution that can solve the above problems. Cloud dial testing, as a non-invasive cloud native monitoring product for business, has become the best choice. Through Alibaba Cloud's global service network, simulate real user behavior and continuously monitor the availability and performance of websites, their networks, services, and API ports 24/7. Implement fine-grained problem localization at the page element level, network request level, and network link level. Rich monitoring correlation items and analysis models help enterprises timely identify and locate performance bottlenecks and experience dark spots, reduce operational risks, and improve service experience and efficiency.

（1） Global monitoring node coverage

More than 200000 LMs worldwide, over 500 IDC terminal monitoring nodes, over 400 domestic and foreign operators, and hundreds of thousands of registered members ensure that the monitoring scale meets the increasingly large business scale.

（2） No need for embedding, ready to use out of the box

Zero intrusion monitoring, simply enter the URL and perform simple configuration, without the need for research and development support. A complete website performance data analysis report can be obtained in just a few minutes. Resource pack&pay as you go multiple purchase modes to meet the needs of operation and maintenance testing.

（3） Business oriented, preset multiple analysis models

The monitoring cycle is refined to the minute level, with more than 20 monitoring related parameter settings in 7 categories, supporting multiple mainstream protocols, and providing 7 × 24-hour real-time monitoring, alarm, and performance analysis services for fine-grained faults. From the perspective of the end customer, through multi-dimensional combination analysis of regions, operators, and other factors, drill down to analyze the details of a single sample, utilize a rich indicator system and chart types, intuitively locate the problem, affected area, and root cause, reduce analysis time, and improve operation and maintenance efficiency. Truly achieve refined monitoring.

（4） Intelligent alarm, precise positioning

Real time alarms are implemented for the first screen usage, overall performance, and availability, with rich alarm policy settings and deep integration with Alibaba Cloud's alarm center, effectively shortening MTTR. Support the discovery of page element level errors, accurately pinpoint problem attribution to a single network request process, and improve problem localization efficiency.

Taking the marketing promotion of a certain e-commerce enterprise as an example, the website has over one million monthly active users, mainly distributed in third, fourth, and fifth tier cities nationwide. The annual expenditure on website operation and maintenance exceeds 2 million yuan. However, due to the frequent updates of product information during the promotion phase, users from various regions often complain that "product images cannot be loaded" and "page opening is slow", resulting in low user conversion and complaints against the operation and maintenance team.

Faced with this dilemma, we have solved this problem through cloud testing products and further optimized website performance to support business growth.

（1） Stress testing

Before the marketing activities or new system launch of the enterprise, use cloud testing to select monitoring points from different city operators across the country, set browsing and network tasks, instantly obtain real user experience data from the frontline, accurately locate the page elements that have problems, and help the technical team repair problems in a timely manner. Simulate peak users with high concurrent access, increase peak pressure, observe changes in major performance indicators, and identify performance bottlenecks.

（2） User experience optimization

Through the first screen monitoring and real-time monitoring functions, problem verification and fault recurrence can be immediately carried out, and website performance can be evaluated and optimized. And through transaction flow analysis, understand the real user experience process, optimize browsing paths, identify conversion bottlenecks, and improve conversion rates.

（3） Competitive analysis iteration

By utilizing the zero intrusion feature, collect and analyze the performance of competitive marketing activities in the same industry, understand the changes in competitive marketing trends and response plans, and conduct targeted IT investment and optimization iterations to make up for marketing shortcomings and stabilize the leading position.

After the above measures, website performance has significantly improved, and quantitative indicators related to user experience have increased by more than 30%, effectively driving business growth. In addition to the above scenarios, cloud dial testing can also be widely applied in many scenarios such as network interfaces, service availability monitoring, CDN service monitoring and selection, DNS resolution status, hijacking analysis, etc.