## 1. Control risks
"Management" represents a process. Modern management believes that management usually goes through four steps: Planning, Organizing, Leading, and Controlling, and these four steps are constantly cycled. Taiwan Semiconductor Company ASE experienced a 13-hour downtime due to improper management in the past. <a href="https://www.ithome.com.tw/node/33659" style="color:#0088EE;" target="_blank ">"Revelation of ASE ERP downtime for 13 hours"</a>, Sheng Mincheng said, "In IT systems, any small flaws will be magnified, because too many small problems add up to form a big problem. , to avoid problems, in addition to technical issues, "management" is the foundation of the IT system. "
Scholar Robbins believes that "control is the most important part of the management function."
In order to effectively achieve the goals, ensure that the process is consistent with the goals, and correct major deviations, we need "control" to supervise the activity process, which can be divided into pre-event control, in-event control and post-event control according to the time of occurrence.
When faced with huge instantaneous network traffic, how can enterprises prepare to minimize risks?
### (1) Precontrol phase
Before starting corporate activities, whether it is gathering past experience or re-evaluating the current ability to bear risks, it is an act that is responsible for oneself and customers.
NoQ provides "system load testing". Before holding any event, it is necessary to understand your system load. It can help you avoid system crashes due to sudden large traffic during important events. Penetration testing and code scanning can help you understand security vulnerabilities in your system.
Our professional testers will analyze the system bottlenecks through the system design information provided by the customer. Using Apache JMeter, we will simulate tens of thousands of users using the customer's system at the same time to understand the performance of the customer's system under high traffic and identify possible weaknesses in the system.
After the above tests, we will evaluate whether to use RoomQ or NoQ Flash for this event. The point of doing this is always<a href="https://wiki.mbalib.com/zh-tw/%E7%AE%A1%E7%90%86%E6%8E%A7%E5%88%B6" style="color:#0088EE;" target="_blank">Prevent quality deviation between input resources and output</a>.
#### (2)Continuing control phase
Also known as continuous control and real-time control, it emphasizes taking actions while proceeding, focusing on the process of carrying out activities. Its purpose is to take corrective actions immediately when any deviation is found in the activities to avoid causing greater losses, for example <a href ="https://mymkc.com/column/content/779" style="color:#0088EE;" target="_blank">"Walk-through management"</a>, managers will visit the work site to understand Employees identify current problems and immediately help solve them to avoid further damage.
*Image credit: Tom Peters and Bob WatermanIn Search Of EXCELLENCE*
During the online activity process, RoomQ is your patrol manager, helping you go through the system every 15 seconds. You can check the current data through the backend provided by RoomQ, and control the website traffic to ensure that the activity is carried out as planned, and you can freely adjust according to the situation. Update the text information on the waiting page to keep users informed and reduce anxiety or irritability.
Don't forget that this patrol manager also provides emergency "temporary rescue." When the event is running and you are already facing abnormal operation such as system delays, please contact us immediately and let us assist in emergency response.
#### (3) Post-event control
This is the most common post-mortem review and remedial measures. The most common method of control is to make up for it before the sheep has gone. However, statistics show that there are <a href="https://itic-corp.com/blog/2020/06/forty -percent-of-enterprises-say-hourly-downtime-costs-top-1million/" style="color:#0088EE;" target="_blank">40%</a> of enterprises said a one-hour system downtime order They lost over a million dollars,<a href="https://www.yottaa.com/2019-rsr-ecommerce-industry-research/" style="color:#0088EE;" target="_blank">90 %</a> of customers leave because the page loads too slowly, <a href="https://www.yottaa.com/2019-rsr-ecommerce-industry-research/" style="color:# 0088EE;" target="_blank">23%</a> of consumers will never visit a brand website that loads too slowly again, and the consequences are always unexpected.
## 2. Queuing Psychology
The above is a discussion of how to prevent large traffic damage on the enterprise side from a management perspective. Next, I will introduce the queuing psychology of psychology master David Maister, <a href="https://davidmaister.com/articles/the-psychology-of-waiting- lines/" style="color:#0088EE;" target="_blank">〈 The Psychology of Waiting Lines〉</a> The entire article proposes 6 basic principles that affect the end user’s queuing experience. These principles determine whether the entity Queuing or queuing online, the user experience while waiting.
#### 1. The actual waiting time is shorter than the perceived waiting time
Usually customers are affected by the subjective perception of waiting time and cannot always notice the actual passage of time. "Perceived Waiting Time" refers to the time that consumers feel after selecting, organizing and interpreting external stimuli. There have been studies in the past on <a href="http://jeb.cerps.org.tw/files/JEB2017-008.pdf" style="color:#0088EE;" target="_blank">customers’ time values</a a>Discuss and believe that perceived waiting time is the key indicator to explain customers' waiting reactions.
If users are allowed to do other things while queuing, such as leaving the screen to make a bowl of instant noodles, continuing to watch a Korean drama in the middle of the show, or even posting on Facebook to show off that they have won the chips and are waiting, they can be removed. The sense of constraints of queuing.
##### 2. Let users feel the beginning, which is better than waiting in vain
Not only are customers shown the menu while queuing, but also common "appetizers" are included. In Taiwanese restaurants, after guests order their meals, they usually serve plates of pickled vegetables (pickled cucumbers, kimchi), or Beans (fried peanuts, salted edamame) will make your mouth think - it has already begun!
In the same way, you can also put some "side dishes" on your waiting room page, such as other related information about this product, or the introduction and registration time of the next event...
#### 3. A known limited waiting time is better than an uncertain waiting time.
From the perspective of customers waiting in line, time is lost, and the economic and psychological costs caused by waiting cause them psychological pressure.
<a href="https://www.jstor.org/stable/1251932?seq=1" style="color:#0088EE;" target="_blank">Hui and Tse </a> will queue information (Queuing Information) is defined as "information that continuously updates the consumer's position in the queue."
Clear queuing information solves the problem of "unpredictability of actual waiting time" and displays <a href="http://jeb.cerps.org.tw/files/JEB2017-008.pdf" style="color:#0088EE ;" target="_blank">Queue information</a> includes the number of people before and after the customer's location. Through constantly updated queuing information, consumers can know where they are in the queue. This is more effective than just telling them the waiting time. Contribute to the generation of positive emotions.
Image source: <a href="https://uat8.noq.hk/zh/blog/harbour-north" style="color:#0088EE;" target="_blank">RoomQ</a>
#### 4. Waiting time with an explanation is better than waiting time without any explanation.
When it rains heavily, we usually wait more patiently for the pizza delivery guy.
When there is a traffic jam, we will send a message to tell our family, "I am stuck in traffic, so I can't make it to the restaurant."
When customers are only told that the system is delayed due to "technical issues", it sounds ambiguous and unreliable. However, if they are willing to explain that the system traffic is too heavy and are temporarily diverted to wait, and to ensure a safe and fair position, it feels much better. .
#### 5. Fair waiting is better than unfair waiting
The first-come, first-served queuing mechanism is a model of fairness, just like a race. For customers waiting in line, their hearts are particularly sensitive. At this time, the "sense of fairness" can help stabilize their emotions and avoid chaos. RoomQ and NoQ can effectively prevent anti-bots and flood attacks. Anti-flooding, this is definitely one of the important functions for handling large traffic systems.
#### 6. Anxiety makes waiting longer
Anxiety usually stems from the unknown,<a href="https://read01.com/zh-tw/JmR266.html#.X5-EiVMzaCQ" style="color:#0088EE;" target="_blank">"Progress bar ”</a> The function is to make the time transparent, allowing customers to see that their team is moving forward, which can effectively alleviate anxiety.
To give an interesting example, usually our experience is:
A hungry stomach will accumulate anxiety every minute and make you extremely irritable!
The slogan of the Taiwanese restaurant chain "Tingshi 8" is "If you wait for more than 15 minutes, you will get a free meal coupon." Because of this guarantee, every customer is happy to wait, so while waiting for the meal, they keep their eyes on the moving waiter, every time they pass by I thought silently in my mind: Don’t come over, don’t serve the food, just wait another 2 minutes!
Through front, middle and back control management and understanding the psychology of queuers, customers’ waiting experience can be concretely improved.
Only by mastering self-management and understanding the public psychology can waiting be seen as expectation, not harm.