Contents:

Free course: "Quick start in Python"
Learn more
Senior iOS developer at VKontakte. He has experience as a full-stack developer, back-end developer, and DevOps engineer. He managed the mobile development department and taught iOS development for three years at GeekBrains, where he also served as the dean of the faculty. He serves on the program committee of the Podlodka iOS Crew conference and runs a YouTube channel with Flutter tutorials. On Twitter, he shares his thoughts under the handle @tygeddar.
Links play a key role in the structure of the internet, providing connections between different pages and resources. They help users find the information they need and also contribute to improving the visibility of websites in search engines. It is important to use links correctly, including internal and external ones, to improve the SEO of your content. Internal links help users navigate your site and deepen their engagement with the content, while external links to authoritative sources can increase your site's credibility. Optimizing the anchor text of links is also important: it should be informative and relevant to the content it links to. Effective use of links improves user experience and improves search rankings.
I've been working remotely for a long time, while most people started this journey just over two years ago. The pandemic has left many managers worried: how to supervise employees in a remote work environment? In fact, managing a team in a virtual setting is no different from in the office: trust and effective communication remain key factors.
Who are we trying to fool?
In an office environment, there is no oversight of employees. Over the course of six years of work, including constant presence and occasional visits, I have observed colleagues freely coming and going at any time, arriving late, or spending long periods at lunch. No one monitors this process. This creates an atmosphere of trust and autonomy, allowing employees to manage their time and work responsibilities.

You You might argue that work time controls have become stricter now: turnstiles record arrival and departure times, as well as the total number of hours spent at the workplace. This is true, but even in the office, employees are not constantly monitored. You can spend time in the PlayStation area, in the gym, or simply chat with colleagues in the open-plan office, discussing cars or concerts of your favorite artists. However, management will not be monitoring your every move.
During the trip, I noticed a young man sitting quietly in the corner, almost constantly immersed in his personal affairs. From the outside, this looked quite acceptable: he was sitting with headphones on, his back to the wall, staring intently at the screen of his device with a serious expression. It seemed like he was working hard. This approach to working on public transport commands respect.
Remote work requires the same supervision as in the office. It is important to monitor the volume of tasks, the quality, and the speed of their completion for each employee individually. All other discussions are of no practical significance. Effective remote team management relies on a clear understanding of tasks and deliverables, which contributes to increased overall productivity and employee satisfaction.
How I was under the hood

I worked with a time tracking system, and the experience was extremely unpleasant. It's important to understand that a developer can't work without breaks for an entire shift. It's simply impossible. Of course, there are professions where the work consists of routine tasks performed for eight hours. However, even a layout designer needs to take breaks from monotonous layouts from time to time. Without this, the cognitive load becomes excessive, which negatively impacts productivity and the quality of work. An effective approach to time and task management requires flexibility and consideration of the individual needs of employees.
The tracker only tracked time spent coding and phone calls, which created additional inconvenience. If a break was needed, for example, for tea or a smoke, the user was required to stop the tracker. Moreover, all employees were required to record their actions every half hour, which added unnecessary workload and distracted from the main work.
Of course, I'm ready to help with text editing. Please provide the source text that needs to be revised.
If I spent 40 minutes discussing work issues with a colleague outside the main area so as not to distract others, the question arises of how to document it. In what task should I record this conversation? We could discuss not only current matters but also potential problems or future projects that are also related to our work. Therefore, we often have to be creative in structuring such meetings, which creates certain inconveniences. Properly organizing such discussions is important for time tracking and increasing efficiency.
There are trackers that capture screenshots at certain intervals, for example, every 10 minutes. If you receive a message in Telegram and open it to read, a screenshot is created at that moment. This leads the system to believe that during these 10 minutes you did not work, but spent time on the messenger. Proving otherwise can be extremely difficult.
Some continue to track user keyboard activity. This is quite complex and inefficient, as most of the time an experienced developer spends is spent thinking through decisions. This process can't be measured by any counters, but it is crucial for code quality and the successful completion of projects.
I found a way to effectively use this tracker for personal needs. I set it to eight hours to avoid overworking. It helps me manage my time and maintain a balance between work and leisure.
Time is not money

Predicting the results of a factory is easy when 30 employees are cutting identical bolts on identical machines. If 29 of them produced 500 bolts each, and one produced only 400, then questions about that employee's productivity are likely to arise. This approach allows for the effective evaluation of each employee's work and the identification of bottlenecks in the production process.
Working with programmers requires an individual approach. Each specialist has his own tasks, varying in complexity and volume. It is often impossible to accurately determine in advance how much time will be required to complete them.
The workflow is often interrupted due to missing requirements or the need to wait for other developers to complete tasks. Situations may also arise that require discussing details with the designer. These factors significantly increase time costs and complicate the evaluation of work performance. As a result, I have to rely on intuitive performance assessments.
In most places I've worked, colleagues have expressed satisfaction with my assessments. Perhaps I'm just lucky and join strong teams.
In my experience, I've only encountered subordinates with extremely low productivity twice. One of them consistently demonstrated the slowest results and made mistakes in his work. When asked to make corrections, instead of fixing the problem, new problems emerged. After three months of ineffective work, without completing a single task, it became clear that the collaboration had to end.
The second employee seemed to be coping with the tasks, but many suspected he wasn't giving his all. His deadlines were constantly slipping. Something would always happen that would cause him to push the deadline back: he'd promise to finish it in the evening, but then push it back again. I couldn't approach him directly and ask if he was lying, as that would have been impolite. We discussed the situation with the manager and a colleague. We discovered that where others needed just a few hours, he was taking two days. However, he always found convincing excuses for the delays and demonstrated in detail what he did and when.
Over the course of several weeks, I closely observed his work and came to the conclusion that he was genuinely slow or skillfully faking it. Nevertheless, he was constantly busy with various tasks, pondering solutions, and trying different approaches. Ultimately, we concluded that he was simply a low performer and accepted it as a given.
Agile is fair and understandable
There are several estimation systems in Agile, but the most effective is the story point methodology. This system allows teams to more accurately estimate the complexity and scope of work, which contributes to better project management and increased productivity. Story points help teams focus not only on the completion time of tasks, but also on their complexity and risks, which makes the development process more flexible and adaptive. Using story points also facilitates sprint planning and helps in monitoring project progress.
The first thing to consider is the need to break down the overall task into smaller components. The smaller the task, the easier it is to estimate and the less likely it is to be estimated. The second important note concerns the use of story points, which are a kind of time equivalent. Estimating one task against another allows you to more accurately determine its complexity and the resources required to complete it. This approach improves project planning and management and minimizes the risks associated with underestimating or overestimating tasks. Let's say you completed a simple task in four hours. Now you consider the next task and realize it will take five times longer. Based on this, you assign story points and estimate the team's velocity. Proper task allocation and time estimation are critical to successful project management and achieving effective results. Effective story point estimation helps teams optimize workflows and increase productivity, which in turn contributes to the successful completion of projects on time.
If you estimated the total workload for a two-week sprint at 200 story points, but your team only completed 130 and continues to complete 130 in subsequent sprints, this indicates your team's actual velocity is 130 story points per sprint. In this case, it is advisable to reconsider sprint planning and set more realistic goals to ensure effective task completion and improve team productivity.

While these metrics are not absolute values and do not allow us to evaluate individual achievements, they do provide an idea of the team's performance as a whole. It may happen that the entire team is working actively, but one person is less active. In such cases, it becomes necessary to apply various analysis methods to determine the reasons for this phenomenon. This is no different from situations that might arise in an office environment. You can't simply evaluate programmers if they are working on different tasks. This analogy with bolts emphasizes the importance of considering the context of each team member's work.
Why KPIs and OKRs Don't Work
KPIs and OKRs are no longer just tools for evaluating performance, but also mechanisms for rewarding and setting goals. However, in practice, a clear and understandable system that is accessible to all participants is often lacking. Instead, we are faced with unclear and abstract metrics developed without taking into account the real needs of the business. This creates confusion and reduces employee motivation, as they don't understand the criteria by which their work is measured. It's important to develop a transparent and logical system of KPIs and OKRs that is understandable to everyone and helps them achieve their goals.
I once worked for a company with a flexible approach to KPIs. Each employee had the opportunity to set additional goals for themselves in addition to their primary responsibilities. This could include learning a new technology, developing an innovative solution, or writing a report on achieved results. This approach contributed to the development of professional skills and increased the overall effectiveness of the team.
No additional time was allocated for these tasks, despite a lack of resources to fulfill current responsibilities. Employees had to constantly struggle with the workload to avoid overtime. The main question remained: why is this necessary at all?
Developers often face the need to grow and master new technologies. For beginners, this can be difficult, as everything around them seems unfamiliar and incomprehensible. However, if you already have experience working with 15 different technologies, the question arises: is it worth mastering a 16th? It is important to consider that the use of a new technology must be justified by the actual needs of the project. If you cannot apply the new technology, this may raise doubts about its necessity. A smart approach to training and development will help you avoid overload and focus on truly relevant skills.
One company had a complex system of growth points that was causing confusion among employees. One of these points was documentation maintenance, despite the presence of a dedicated analyst responsible for this task. Employees could turn to the analyst and try to prepare the materials themselves, but this seemed strange: it’s hard to imagine a programmer aspiring to the role of analyst. Usually, it’s the opposite—analysts strive to move into development. As a result, the team wasted time coming up with irrelevant schemes and diagrams to fulfill goal requirements, instead of focusing on the real tasks of the project. This emphasizes the importance of a clear structure and understanding of team roles for improving team efficiency and achieving real results.
While working on the railway, I encountered the implementation of the Toyota 5S system, which involves workflow optimization. One of the key requirements was that employees submit a certain number of their own rationalization suggestions each month. For this purpose, a special system for recording these initiatives was developed. However, many of the proposals raised doubts and seemed impractical. Instead of focusing on completing assigned tasks, employees were forced to waste time coming up with ineffective ideas.
Conclusion
Automated systems for evaluating the effectiveness and efficiency of developers may not deliver the expected results. For such systems to be truly useful, clear and specific criteria for measurement must be established. However, in practice, I have never encountered such criteria, which calls into question the effectiveness of such tools.
Testing systems are often prone to simplification and inefficiency. For example, if you decide to implement testing in your project, this is commendable, and having a metric such as coverage percentage helps track progress. At first glance, it is enough to simply start the process, and if coverage grows, then you are on the right track. However, there are many approaches to writing tests. You can quickly create multiple tests, increasing coverage by 20%, but that doesn't guarantee their usefulness. On the other hand, a single high-quality test can provide much more reliable results, even if it only slightly increases coverage. The question is, which will you choose if your premium depends on these metrics? High-quality tests are always preferable, as they provide true confidence in the functionality of your code.
Reworked text:
Read also:
- Who is a blockchain developer and how to become one
- 7 books on DevOps for beginners and advanced engineers
- Andrey Ershov: the father of Soviet "computer science" and one of the first programmers in the USSR

