On Weekly Reviews

As I talked about in my post on annual reviews, I use a weekly diary-type process to keep track of how everything is going.

My process is simple: I have a spreadsheet in Google Docs which has a number of rows containing questions, and every week I add a new column with the answers to those for that week. I usually set aside around half an hour to do this, though depending on what the week has been like for me it could take more or less time than that.

The exact questions will depend on what’s important to you and what you want to track. A few of my key ones are:

  • What went well this week?
  • What went badly this week?
  • Who did you enjoy spending time with?
  • Who did you not enjoy spending time with?
  • What’s your mood like right now?
  • What happened this week?

In the first two I usually put down things like work, personal relationships, sleep, exercise, nutrition, motivation, and progress on projects outside of work.

Three and four force me to be honest about the people I spend time with, and allow me be more aware of whose company I enjoy consistently and who I should be spending more or less time with.

The last two are great when looked at across an entire year. While the mood question tends to have a one word answer, the other one I treat like a free-text area where I just write down in a few lines all the things I can remember happening, in no particular order. This part is more like a traditional diary, and gives some insight into what I was thinking about/doing each week.

I think everyone should take the time to answer the above six questions on a weekly basis. This system not only provides them with greater insight into themselves, but is also a useful tool looking back across a period of several months.

In addition to the above, I’ve also integrated a few extra things into my weekly review process:

  • External data sources — RescueTime measures how I spend my time on my phone and computer, and sometimes faces me with the grim reality that I spent 5+ hours on WhatsApp in a single week. Beeminder keeps track of my more short-term personal goals and habit commitments.
  • GTD cleanup — clearing my In list and reviewing my Next Actions list to ensure that everything there is still relevant. More details here.
  • Meta goals — spending 30 seconds thinking about how I can improve my weekly review, and another 30 seconds checking my calendar and finding a convenient time to do it next week.

The last point is crucial to keeping up this or any other habit, and is something I will elaborate on in a future post.

Having done this mini review every week for a year now (except one where I was on a two week road-trip holiday and completely forgot) I feel like it’s added a lot to my life and I expect it to evolve more as I keep doing it. I’d like to thank Cal Newport for giving me the idea to do this.

Enjoyed this article? Please share/recommend/follow 🙂

On Annual Reviews

Reviewing One Year; Planning The Next

There is one ingredient (other than coffee) that is key to a successful annual review — and that’s acknowledging that a year is a long time. Our minds aren’t built to compare events across such long time frames, and a severe subconscious distortion takes place when you try to think about everything that happened in an entire year. The solution to this is to make reviewing your life not just something you do on a yearly basis, but an ongoing process.

Looking Backward

Weekly Reviews and The Availability Heuristic

The Availability Heuristic is a cognitive bias that was introduced to me in the book Thinking, Fast and Slow. It’s the phenomenon which causes people to make judgements about the probability and significance of events by how easy it is to think of examples.

In practice, it means that more recent events are far easier to remember than distant events, and so we see them as more significant and more frequent. So when we do an annual review without prior planning, we often end up reviewing the end of the year rather than the year as a whole.

To avoid this bias, and to try to separate recency from importance or intensity, it helps to review the year’s events as if they just happened.

The way I’ve started doing that is to keep a kind of log, a very compact diary, on a weekly basis. I’ve been doing this for a year now and I feel like it’s really helped me keep things in perspective. Reading through this I’ll often be surprised at how significant things which happened a long time ago seemed at the time. I’ve written about my weekly review process in more detail here.

Looking Forward

Use Systems, Not Goals

Gym attendance statistics are one piece of evidence to suggest that New Year’s resolutions don’t tend to work — the general defeatist attitude towards them is another.

Goals like “exercise more”, “eat less”, “learn to speak French” and “lose 5kg” are almost impossible to realise by themselves. Scott Adams has written and talked about this quite a bit. Describing a system as a way to “continually look for better options”, he writes:

Throughout my career I’ve had my antennae up, looking for examples of people who use systems as opposed to goals. In most cases, as far as I can tell, the people who use systems do better. The systems-driven people have found a way to look at the familiar in new and more useful ways.

To put it bluntly, goals are for losers. That’s literally true most of the time. For example, if your goal is to lose 10 pounds, you will spend every moment until you reach the goal — if you reach it at all — feeling as if you were short of your goal. In other words, goal-oriented people exist in a state of nearly continuous failure that they hope will be temporary.

To combat this, I use a combination of vague guidelines and rigid systems. Guidelines are idealistic rules which give guidance in decision making, whereas systems are there to implement these rules in a realistic and sustainable way.

A guideline might be something like Michael Pollan’s “Eat food. Not too much. Mostly plants.” The associated system could be a combination of Beeminder goals to eat less meat and less sugar.

Another set of guidelines could be: “Spend more time with people who help you grow. Spend less time with people you’re indifferent about, and people who hold you back.” This could be implemented as a system by forcing yourself to think consciously on a weekly basis about who you enjoyed spending time with, and who you didn’t.

This year was the first time I implemented these new tools, spending a morning in the local library with a lot of coffee, a notebook and a laptop. I got a lot of clarity out of those few hours, and it got me excited about the year coming up. If you’d like to try this out you might find my other post about weekly reviews helpful.

Enjoyed this article? Please share/recommend/follow 🙂

For more on cognitive biases, check out the book Thinking, Fast and Slow. Buy it now on Amazon (UK/US) or compare prices across multiple stores (UK/US).

Accuracy Over Precision

At school I had a physics teacher who was extremely pedantic. Nothing would upset him more than if you used the wrong term to refer to something. Even small mistakes most people wouldn’t pick up on would incense him.

I didn’t really get why, and in fact we would take great pleasure in saying the wrong thing so we could watch his face turn red as he shouted “Gravity is not a force! It’s an ACCELERATION!

But one of the distinctions he made has stuck with me. One that has turned out to be useful in many things in life, and not just physics. And that’s the distinction between accuracy and precision. While some people use these almost interchangeably, they have very different technical definitions.

Both apply to problems of measurement. In measuring things, we rarely have perfect tools, and most measurements will have some degree of error in them. Accuracy and precision are words we use to talk about different aspects of those degrees of error.

Accuracy vs Precision

Accuracy is how close our measurement is to the real value we’re trying to measure.

Precision is about the level of detail in our measurement.

It’s not immediately clear why these are different concepts, because in many cases they have the same magnitude. For example, measuring length with a ruler will let you read off an answer to the nearest 1mm (precision), which will probably be within 1mm of the real length (accuracy).

But this isn’t always the case. A watch might show the time to the nearest second (precision), but if you haven’t set it properly it could be out by a few minutes (accuracy).

It gets worse. When we start dealing with estimates and predictions in the presence of noise, these numbers can be much further off.


For example, this article in the Telegraph claims that socialising increases our happiness by 6.38%. Yes, it might be true for the set of data points collected on the sample of people surveyed. But the decimal places hardly seem relevant, as the ‘real value’ (if there even is such a thing in this case) probably can’t be measured that accurately in a single study.

What’s happening in the Telegraph article, and in other places, is that by giving results to a high degree of precision, they create the illusion of accuracy and confidence in the results.

It’s not particularly damaging here, as there probably aren’t many people basing their actions on the results of that article. But what about more significant areas?

There’s nothing wrong with predictions or estimates, but giving precise estimates and not stating the variance or the level of confidence can be misleading.


If you’re making a decision based on data, don’t mistake precision for accuracy. If in doubt, ask questions:

  • How many measurements were used to come up with the number you’ve been presented with?
  • What is the range in the sample of measurements? What was the smallest value, and what was the largest value?
  • How close does the person presenting the data think it is to the real value?

The last question will force them to be open about the accuracy of their data, and should allow you to decide how much weight to place on it. Sometimes it can even be best to just ignore the number completely and focus on more important things for your decision!

Thanks for reading! If you’ve found this article useful I’d really appreciate it if you could share or recommend it so other people can find it too. Follow me for more of the same, or check out some of my other articles.

Life, Death, and a Missing Pellet

Photo © John Krzesinski, 2013

I love hearing good stories. It’s a waste when they fade and you forget the details, so I’ve started to record the best ones I hear. I recently went on a climbing trip where one of the more experienced climbers had the whole group listening to his every word with a story he told. I’ve tried my best to retell that story here.

Our friend, Berny, was in Yosemite partway up a multi-day climb of the famous Half Dome with his climbing partner when they came across two Swiss climbers. The Swiss climbers seemed to be struggling, and it quickly became obvious that one of them had completely lost his head. At that point they had already passed a few one-way pendulums so there was no way back — they had to go on. Berny and his partner agreed with the other climber to climb as a four and help out.

As they climbed together Berny and his partner put some time and effort into helping the others, so when night came they didn’t quite make the ledge they were planning on sleeping on. They were one pitch (one stretch of rope) below it on a much narrower ledge, and the Swiss climbers another pitch below them.

During the night Berny woke up and needed the toilet. Yosemite protocol dictates that when nature calls and it’s not liquid, you can’t just do it over the edge. You do it in a paper bag, and then throw the paper bag over the edge.

So he leant back over the edge, hanging off his sling, and did his best to aim into the bag. At this point, because of the dehydration, he said that what came out was not very different to rabbit pellets. He counted one… two… three pellets coming out, but when he was about to throw the bag over the edge he could only locate two of the pellets.

Searching frantically for the third yielded no results, and he soon gave up. He decided that if it happened to be in his sleeping bag then so be it, he needed to get to sleep.

A few hours later he was awoken again, this time by movement.

The whole wall was moving.

They were to find out that it was an earthquake of magnitude 5.6. As they climbed on later that day, they saw that the earthquake had caused a severe rockfall, which had buried the entirety of the ledge above them — the ledge on which they had been planning to sleep. In a way, the Swiss climber had saved their lives.

He never told us what happened to the third pellet.

Thanks for reading! Please share, recommend, or leave a comment if you’d like to see more stories like this.