The link to the first part is here. This is the second one.

Personal SLA

As I mentioned before, people don't have documentation. Instead, we have to rely on personal perception.

If you are a human, you probably noticed that other people always expect something from you, but it may be not clear what exactly. Sometimes the way they think can surprise you:

– Tracy is the only one women in our team, so she should make a cake for our next event.

– Dan doesn't have a family, he should come to the office on Christmas holidays.

– Mike always forgets his glasses, we shouldn't offer him a promotion because he won't be able to handle it.

These things are hard to prevent, but I have two tips on how to make it less frequent:

1. Be conscious of what we do, because it gives people a hint about what to expect in the future.
2. Be consistent in it.

Sounds like two principles from the previous part, right?

Sounds like Personal SLA.

How to build a good Personal SLA?

As we already know from the dangerous part of SLA, one metric is not enough, we should have a set of them. It isn't enough to bring one mammoth per month, we need to add more details.

All the metrics and their target values should comply with the following rules:

1. They should be sustainable. It means that the longer we maintain a good SLA, the easier it is for us.
2. They should help to reach long-term goals.
3. It should be possible to achieve it.

Our set of metrics should cover 3 categories of behaviour:

1. Domain This category describes our area of responsibility: what we can do.
2. Availability This category has metrics that describe when we're doing it.
3. Quality The last category describes how we're doing it. It means not only the quality of a final result but also our attitude and approach.

Now I want to talk about each of these categories, show the examples of good metrics and describe how they comply with the rules we set earlier.

Category 1: Domain

Everything that describes what other people may expect us to do falls into this category.

If your keyboard doesn't work, you call a guy from the IT department. His domain – helping people with IT-related issues.

If you need proof of work for a visa application, you go to someone from the HR group. Most likely, it's their domain.

Sometimes it is not that clear:

– Tracy is the only one women in our team, so she should make a cake for our next event.

For someone, one of Tracy's domains is baking cakes. Perhaps it's true, but we don't know it now. If she disagrees, there is a discrepancy between her inner perception and what other people expect from her.

If someone constantly asks you to do something you don't like to do, shifts boring responsibilities and expects you to be happy about it, you may have the same problem.

Example

I had a problem when I just started working – it was hard to admit that I don't know something.

I felt that I had to find an answer every time someone asked a question. The fact that I worked remotely made it worse. Nobody saw what I was doing, so I had all my time to google or ask someone else.

Now I know that it just was Impostor syndrome, but back then I felt paranoia that someone would found out that I'm not good enough for this job.

To be honest, I can't complain much because it was good for everyone:

1. People could ask me anything and just wait when I handle it.
2. I've built a solid reputation because I've proven that I could work autonomously. My colleagues saw that I could find anything on my own – a crucial skill for a junior back then.

If it worked well, what is it about?

My biggest mistake was that I stuck to the same behaviour when I became a senior engineer and a team lead. When people asked me something, I either answered straight away or leveraged my network to find necessary information for them.

I built expectations that I will always find an answer.

Three reasons why it was bad:

1. I had less time to work on my own tasks.
2. I wasn't happy about this. Most of the time, people ask something when they don't want to search for an answer themselves. It isn't about complex problems and challenges, it's just about time.
3. I became a bottleneck in many situations. When people needed something from me, and I wasn't around, they waited for me. Usually, it wasn't necessary, and they could have solved it on their own.

What would I suggest to myself back then? Obviously, try to build more realistic expectations and let other people find answers. More specifically:

1. If you don't know something, just say "I don't know". Nobody knows everything, this is fine.
2. If you know where you can find this information, share this with a colleague. It can be a link to the documentation (or some keywords to find it), someone's email address, anything.
3. If you decide to find an answer yourself, share the information about how you did it. Make sure that similar questions can be answered without you: – send all the links you found useful – tell about people who can help in similar situations – include the colleague to all communication you initiated, so she can ask follow-up questions on her own

How does it fit the rules of good Personal SLA?

Let's compare two different expectations I've built:

1. I'll find answers on any question
2. If I know something, I'll tell you. Otherwise, I'll say that I don't know.

Rule 1: is it sustainable?

The more I follow the first approach, the harder it is. If people ask me questions, and I find answers for them, they will bring more questions. They will suggest other people talk to me when they aren't sure about something. At some point, I won't be able to handle it, so it is not sustainable.

In contrary, another approach gives people more information about the limits of my knowledge. Eventually, they will ask more about something I know and less about something I don't know, because it's easier for everyone. They will likely not be blocked if I'm not around because there is no point in waiting for me anymore – probably I just don't know the answer. It's worth to wait if only they're sure it's necessary. This approach is sustainable.

Rule 2: does it help to achieve long-term goals?

The answer depends on the goals.

My goals at the beginning were to prove (to myself) that I'm not an impostor and build a good reputation. It worked well, so even the first, not sustainable approach, helped me here.

Obviously, the goals of a senior engineer were different from junior's, so this strategy stopped working. Moreover, it made it worse, as my team was less able to do things in parallel, and sometimes I blocked professional growth for other engineers.

The second approach helped me to have more time for things related to my direct duties. At the same time, the team wasn't blocked by me, and we could achieve more. Saying "I don't know" was the better strategy for achieving my long-term goals.

Rule 3: is it possible to always stick to this approach?

What if I say "I always find an answer to your question"? Can I be confident that this is true?

Obviously, not all questions have answers. Some questions are so complex, so it's unreasonable to spend time on them. Eventually, it means that I won't be able to keep the promise and answer on everything.

On the other hand, saying "I'll answer your question if I know the answer" doesn't have the same problem. It's quite easy to follow this approach, and nobody will be disappointed if I don't know anything. This is definitely possible.

Now you see how finding the right domain and communicating it properly can change the scope of your daily work. After we found the right domain, all we need to do is to stick to it consistently, because this is the only way to build the correct expectations.

Category 2: Availability

Domain gives people around understanding what we can do. Availability has the metrics which describe when we can do it.

– When did I supposed to be in the office?

– How quickly am I answering on WhatsApp messages?

– Is it possible to call me on Sunday, if something urgent happens?

That's all about availability.

Frequently enough, it has nothing to do with what is written in the formal contract as people still rely on their own perception. This quote from the previous text illustrates it:

– Dan doesn't have a family, he should come to the office on Christmas holidays.

Someone thinks that Dan has to be on duty during the Christmas holidays.

Perhaps he already did it last year, or he said that he doesn't like Christmas for some reason. Anyway, he built these expectations, and it will be hard to change.

Example

As I said before, I spent a lot of time helping others with their questions. But I still had to do my work. How did I manage to do all of this?

The answer is simple – I worked extra hours.

The problem wasn't only about questions though, at least 3 standard situations come to mind:

1. If I spent a lot of time dealing with a simple problem, I felt obliged to finish my work no matter what. Imagine, that there is a task that I can do for 4 hours. I start it and then I encounter something unexpected. I spend the whole day debugging, and the problem ends up to be a wrong variable name or a missing comma. In situations like this, I usually worked extra hours to finish the task because I felt that I was my fault.
2. Fixed deadline and changing requirements. Another example – I work on something, and then the requirements change unexpectedly. I have to throw away some pieces and start again. It's worse when the deadline is tied to the start of a marketing campaign or a particular date, such as Christmas or the end of quarter.
3. Emergencies. Of course, nobody is safe from emergencies, and they always happen when you don't expect it. Something critical can happen at 3 AM or 30 minutes before you planned to go home – extra work is inevitable.

How SLA could help me to do it better?

The answer is simple: define a number of hours for work and stick to it.

It sounds strange: how would I work only 8 hours if I already worked 7.5 and we have an emergency? Should I just go home?

The answer becomes simple when you look at traditional metrics in SLA, such as Uptime. A system can't be 99.99% up at one particular moment – it can be either up or down. 99.99% is possible only on a distance.

Same with working time – it's really hard to work exactly 8* hours a day. But it's so much easier to work 40 hours a week. If I had to stay late on Monday, why can't I leave early on Tuesday?

Why is this important?

1. Working long hours negatively affects health: there are correlations between being overworked and problems with sleep, heart and symptoms of depression.
2. Tired people are less productive and tend to make more errors.
3. When you consistently overwork, you make it more likely to happen again because it becomes a part of your SLA that you communicate to the people around.

First two items sound quite straightforward. I'll spend a couple of words on the last one.

My friend works as a web developer in a digital agency. Some time ago, he shared a story about his problem at work.

They have a flexible work schedule, everyone works about 7 - 9 hours, but nobody checks it. Usually, they go home when they finish all planned work.

Recently my friend found himself in a very unpleasant situation – his manager started asking to start important and urgent tasks right before he planned to go home.

She used to do this before, but it was quite rare – once a month or so. Now she does it at least twice a week, and it's getting worse.

It wasn't just a story he told me – it was a complaint about his manager, to show how cunning she is, forcing him to work additional hours again and again.

It was easy for me to understand my friend, but I could imagine his manager's situation as well.

They both were hired to do the best for their company.

My friend's responsibility (one of) is to build good websites.

His manager's responsibility (also, just one of the list) is to organise processes in her teams to make everything done on time and with high quality.

The manager has a couple of dozens of developers in her teams, and she needs to talk to many of them during the day. At some point, she understands that if she rearranges her communication to talk to my friend at the end of the day, he works more.

We don't know how she came up with this – maybe she did it unconsciously, but it's also possible that she tried different approaches before this one. All we know is that she found a way to make her developers work more. It's always easier to measure hours than efficiency and quality, so in her opinion, she did the best for the company. Most likely, she didn't try to make other people feel bad, she just tried to do her work, using all the information she had.

Did my friend try to do his best as well?

I asked him a simple question:

– What would you say if your friend came to you in the middle of a working day and offered to drink 2 shots of vodka in a nearby bar?

– I'd probably say I'm alright.

– Why? Is it prohibited to drink in your company?

– Hm... I don't remember anything about it in my contract. But I don't think I'd be able to work on something after.

This is the point. Several researches show that being tired at work the equivalent of being drunk. For example:

Like with drinking, exhausted neurons respond more slowly, take longer and send weaker signals, according to the new research.

The study could explain why being very tired feels a little like being drunk. And it might also explain why some of the symptoms are similar, like memory lapses and an ability to concentrate after a night of no sleep or lots of drinking.

an excerpt from Independent.co.uk

Just rephrasing his situation: he found that something negatively impacts his performance and attitude. But he doesn't tell his manager about it. He doesn't try to improve the situation.

Instead, he keeps maintaining the same SLA: he will work longer hours if he has a new task at the end of the day. It's a lose/lose situation for everyone.

To be honest, we had this dialog much after my friend has resolved his problem.

I just put it here to look wiser, sorry :-)

Back then, I just suggested to talk to the manager about it to build a more sustainable process, but he found another solution.

He started maintaining the balance – working less if he had to stay late before. The best thing he did – he made it transparent for everyone.

If he came late, he explained why during their daily team meeting. If he planned to leave early, he made the team aware of it. Something like "I plan to leave at 3:30 today because I was fixing X till 10PM yesterday. Please let me know if you need me" made it explicit for everyone.

Even better, when people around started doing the same, it became less common to have new tasks at the end of the day. Their manager still tried to do the best for the company, but now she had more information to find another approach.

All engineers in that team changed their SLAs to maintain the same number of working hours per week. At the same time, they kept being flexible in case of emergencies. As you can see, they found sustainable balance and were consistent in maintaining a reasonable amount of working hours.

*8 hours is just an example because it's the most common value in contracts. The actual value depends on your situation: it can be 4 for one and 10 for another.

How does it fit the rules of a good Personal SLA?

My friend had 2 different tactics:

1. He works until all work is done
2. He works 40 hours a week, but it can be flexible from day to day if needed. If something important happens, he will work extra hours, but then he will be working less to maintain the balance.

Let's compare them as we did before.

Rule 1: is it sustainable?

The first approach is not sustainable, because the amount of work depends on the previous performance. The more tasks you finish, the more tasks you can plan for the future because you did it well in the past. In the previous example, the manager didn't have time to think that working extra hours can affect an engineer's performance in the long term. My friend, in turn, didn't give this information to her. Together they built unsustainable processes, and it was just a matter of time when and how it's going to burst up.

The second one is proven to be more sustainable, as the manager has more information about availability to prioritise the work wisely. The engineers stick to the schedule, that helps them to maintain optimal performance, while they're still flexible to handle emergencies. The more they follow this approach, the easier it is to plan ahead.

Rule 2: does it help to achieve long-term goals?

I can't answer this question with enough confidence because I don't know what goals my friend had back then.

I can suppose that his goal was to build a successful career in his company and, at the same time, enjoy the working atmosphere and don't sacrifice personal life. In this case, the first approach didn't work as he planned to look for a new job. The second approach works well and not only for him.

Rule 3: is it possible to always stick to this approach?

I don't think that it's feasible to go home only when all the work is done. Who defines how much work needs to be done? Who maintains the achievable level?

Obviously, it's a good intent to plan wisely and then successfully execute this plan. But in most situations, the level of uncertainty is too high, so being able to achieve long-term goals is much more important than finishing all tasks today no matter what.

How can we say that we have rest only when all the work is done when most of us can't even predict what we'll be working on tomorrow?

The second approach works surprisingly well in these situations. I can always say when I have my free time. I have a simple rule for emergencies. It doesn't depend on external factors.

Of course, sometimes we have more unexpected situations, and sometimes everything goes as planned. But if our SLA is clear for everybody, it's possible to always stick to this behaviour.

Category 3: Quality

We defined what and when we can do. The last category is about how.

What is this how means exactly? Several things, for example:

1. How good the final result is.
2. How we organise work.
3. Which methods we use to reach the goal.

These metrics, usually, are hard to measure.

Therefore, if someone has expectations about quality , it's difficult to change.

We had this example of expectations in the beginning:

– Mike always forgets his glasses, we should't offer him a promotion because he won't be able to handle it.

As you can see here, someone has an impression that Mike is unmindful and careless, so the expectations about his ability to handle the new position are low. It isn't about what he does or when he does it, it's about his attitude.

What can he do now to change the situation? It doesn't look like it's going to improve if he stops forgetting the glasses.

Therefore it's important to build the right impression from the beginning – define reasonable targets for the metrics and consistently achieve it.

There are plenty of different metrics that can define quality. Usually, most of the targets you can define for yourself will be sustainable and achievable. So instead of talking about one particular situation with good and bad targets, I want to show more examples, but spend less time talking about them.

Examples

1. You always aim for the best possible result. Yes, it will take more time. But it's possible and sometimes necessary. What if others lives depend on your work? What if you're building a space rocket and you have only one chance to make it right? If people around think that you always achieve high-quality results, you will be the best person for the most critical tasks.
2. You always try to make _ something working _ as soon as possible. There is a lot of situations when results shouldn't be perfect, and velocity matters more. Need to build a prototype to test a new idea? Or to come up with the fastest possible solution for the client's problem? There is a demand for people who can do this, so if this sounds like you, you will always find a good niche.
3. You can be good at working autonomously. It's very good if you can take a problem and come back with a solution. If you always do this, people around are sure that you will either make it done or return with the clear explanation of why it's impossible – no need for long meetings and constant status updates.
4. You can be good at working in a team. Organisations don't need bottlenecks, they need efficient processes, sharing knowledge and productive teams. No need to say that it's crucial to have someone who can help people achieve together much more than separately.

As you see, completely opposite approaches can be good in some situations, it's just a matter of aligning your goals with real needs and being consistent.

But what if I want to find something in the middle?

Is it really necessary to stick to the extreme?

What if I want to show that I can be flexible: work autonomously and in a team, make quick prototypes and work on high-quality products?

This is harder than you may expect because you never know what people notice:

1. You've built a perfect product, but your boss noticed only how long it took to make.
2. You've implemented a quick fix for an important client, but your colleagues noticed only how hard it was to work with your temporary solution after.

Usually, it happens not because they're bad people or don't like you – more likely they just don't have all the information. What can help in that case?

The answer is simple – communication.

If you want to be sure that people remember the right thing, you have to tell them about it.

Do you remember Mike, who forgets his glasses? Now you probably have some advice for him. If he wants to be perceived differently, he should talk about it.

Mike :

When I work on an important project, I'm very concentrated, so I don't notice anything around.

Yesterday I forgot to have lunch, today I can't find my glasses.

Being able to focus adds some benefits, but I try to step back and think about the big picture every Tuesday and Friday. That way, I can be sure that I use all my skills effectively.

If people hear this, they will see him as a person who cares about his productivity and works on his strengths and weaknesses. At least until he forgets something on Tuesday or Friday.

I'm serious here – communication is just an instrument here. Consistent behaviour is still required. As we discussed in the previous part, people rely on their experience more than on what's written or said.

Summary

A good Personal SLA covers three categories:

1. Domain ( what can do)
2. Availability ( when we can do it)
3. Quality ( how we do it)

All the metrics and their target values should be:

1. Sustainable
2. Helpful in reaching goals
3. Achievable

The best way to make sure that people build their expectations according to your SLA is to always stick to your metrics.

Sometimes maintaining the consistent SLA doesn't help, and people still make unrealistic expectations. Why?

Most of the time, they just don't have all the information.

In this situation, communication helps a lot: tell people what and why you're doing, give them more details.

There are different examples of good SLAs, and sometimes they're completely opposite. What works in one situation may not work in another.

So the best advice here is to build your SLA based on your goals, your understanding of sustainability and your own capabilities.

This article wouldn't be possible without all the people who talked to me after I published the first notes on this topic.

If you use the same principles, disagree with something or have interesting examples, please feel free to reach out to me on twitter, facebook or by email.