DEV Community

Cover image for Resolving user requests in 30 minutes. Serverspace approach
Serverspace.us
Serverspace.us

Posted on • Originally published at reddit.com

Resolving user requests in 30 minutes. Serverspace approach

Receipt of requests
Each of us has contacted technical support at least once in our lives. As a rule, we call support by phone, write an email, or send a message in a special chat room. Indeed, these are the most common channels for receiving requests. However, support staff can be disturbed not only by users but also by robots; the smooth functioning of the cloud provider's infrastructure is monitored by automated monitoring systems, which are sure to sound the alarm if something goes wrong. Specialists have to respond to such signals promptly; thanks to this, the problem can often be solved even before the customer finds out about it.

Practice shows that users often choose email to send a request. Our users prefer to put a problem or ask a question in writing rather than in a dialogue with an operator because of the possibility of attaching screenshots to the letter for the situation's demonstration.

One more popular way to send a request is through the "ticket system" in the control panel. Serverspace users have access to this system via login and password. Any registered user can ask our specialists a question directly in the "Support" section of the personal cabinet.

Tickets get into the management system, and the support team sees the requests at once. This method simplifies communication and reduces the response time of tech support. Less frequently, customers call tech support by phone. This method is the most important priority for solving urgent problems.

Regardless of the channel, all requests are processed according to the same scheme. The nuance is that these requests are different by themselves, and therefore they may reach different specialists.

Types of requests

Depending on the situation, all requests are divided into two global categories: incident or service request. The first category includes all kinds of technical failures. For example, if a virtual machine suddenly stops responding or the administrator cannot log in to the server.

The second category includes all kinds of technical advice and user requests, such as a request for help figuring out how to increase resources on the server or how to install and configure software, etc.

Service requests are fulfilled according to the approved protocols, but incidents have a much more interesting fate. Each request received by the technical support service has its own urgency. The most urgent tasks are those that directly affect the security and operation of the infrastructure. High-urgency tasks include all cases of significant degradation of service quality that may harm the user. Medium-urgency tasks are characterized by a minor degradation of service quality, which may increase to a significant degradation. The category of incidents of low urgency includes various inconveniences in the use of the provider's services that are not critical to their operation. Finally, the wishes of clients have no urgency.

In addition, each request has an impact. This characteristic determines how many users and systems are involved in the incident. If all users or systems are affected, it is a significant incident. High-impact incidents are those in which a large group of users or systems are affected. Incidents affecting a small number of users and systems have a medium impact; if only one user or system is affected, the impact is low. Finally, if an incident does not affect users or infrastructure, there is no impact as such.

All incidents are prioritized according to their urgency and impact. The priority has four gradations: low, medium, high, and very high.

Support lines

Although from the outside it looks like the technical support team is just a group of people sitting on the phone, the support team is divided into three support lines inside the company. They differ not only in the level of qualification of the engineers working there but also in how they fulfill slightly different functions.

The first line processes the incoming flow of requests, registers them, evaluates their urgency, influences and prioritizes them, and performs initial diagnostics if necessary. Serverspace has a set of specially developed instructions for solving typical, frequently occurring problems. When a problem can be solved using such instructions, the first-line technical support does it ASAP. If there is no suitable instruction or the problem turns out to be too complicated, it is passed to the next level.

The first line can be staffed by specialists with minimal IT experience and basic knowledge of system administration, virtualization, and containerization. Ideal candidates are senior students at technical universities. Knowledge of English is also important for assisting in two languages. All first-line employees go through a special training course that familiarizes them with Serverspace's services and technologies.

Second-line technical support specialists are experienced system administrators who can deploy a virtual machine, configure an operating system, configure protocols, and understand the configuration of the server. As a rule, engineers who work here have been working on the first line for some time and have accumulated relevant knowledge. Even if an experienced sysadmin comes to work in technical support, he is still offered to work a little at the first line to get into technological processes and get used to the infrastructure of Serverspace.

If the knowledge or competencies of the second-line employee are not enough for some reason, the request is transferred to the third line of technical support. At this level, vendors and specialists with narrow profiles (equipment repair and configuration, networks, or containers) are involved in solving the problem. Such experts deal with complex incidents and customized client tasks, the solution of which requires special knowledge and skills.

Request statuses

Each request has its own lifecycle, and the client can always find out the current status of his request in the control panel. As soon as the request has been received and registered in the admin panel, it will receive the "New" status. If additional information is required to resolve the issue, the status "Information Required" is displayed. When the issue is resolved, it requires verification and confirmation from the user, and the panel will show the status "Resolved". The status "Closed" is automatically changed to "Closed" if there is no activity in the ticket for 3 days.

Incidents also have their own lifecycle; the statuses are displayed in the system only for support specialists. As soon as a request has been received by the support team, classified, and registered in the system, it will receive the status "Registered". Then a specific executor is identified to work on it, and the status changes to "Assigned". When the performer starts to solve the task directly, it receives the status "In Progress".

If information from users is required to solve the incident, technical support requests the necessary data. There are situations when the cause of the incident is third-party software or equipment; then the request is redirected to the technical support of the corresponding company, and the request is assigned the status "On External Support". Finally, the result of the request processing can be the status "Completed" (the work has been completed), "Closed" (the problem has been solved and the client is satisfied with the actions of the support team), or "Rejected by the user" (the problem solution did not satisfy the client). In the latter case, work on the request is resumed, and the cycle is repeated.

Technical support performance criteria

The quality and efficiency of the technical support service are evaluated by several criteria at once, and one of the most important is the response time. Users most often pay attention to this parameter. When an unforeseen situation occurs, a person is in a state of stress, and it is extremely important for him that the specialist is engaged in solving the problem ASAP.

Criteria for technical support work and quantitative and qualitative targets for services are defined in the document called "Service Level Agreement" (SLA). According to this Agreement, the support team should respond to incidents within 20 minutes; a typical request, i.e., a request formalized in the standard Serverspace instructions and not requiring a call to the 2nd and 3rd line support, is processed within 30 minutes. Non-typical requests—those for which it is necessary to involve employees of the 2nd and 3rd lines—are processed within an hour.

Serverspace's technical support service is in touch with clients 24/7, including weekends and holidays; this is the schedule of the 1st and 2nd lines dealing with incidents, standard requests, and most non-typical requests. Line 3 specialists work five days a week, helping customers during business hours.

Serverspace's support service is continuously developing; if earlier the speed of response to requests was about 30 minutes, now the company strives to fully solve the user's problem within this time. This is achieved by introducing new protocols for request processing, standard scenarios, and strategies of interaction with clients. Another important aspect that is given special attention is the continuous training of employees.

"It is important for us that every Serverspace user receives the most prompt and high-quality support in any situation. That's why we are constantly improving our support service and aim to solve user queries in 30 minutes. We are confident that this will help us to improve interaction with our clients and provide them with the best possible service", Dmitry Semin, head of Serverspace customer support service, told RBC.

Top comments (0)