Do you like pineapples? Do you remember the taste? Surely you do, even if you’ve tasted it only once. And you can easily recognize delicious pineapple among average ones — instantly, in a twinkling of an eye.

We have the ability to recognize. And it seems that recognition is a function that does not depend on the nature of input: it can be fruits, images, political speeches, languages, toothpastes, gestures and so on.

But in practice the precision of recognition differs very significantly for different inputs. How many times in history we’ve been fooled be politicians. But we’ve never been fooled by fruits. It is worth to think about this.

On the tip of iceberg, I think, is one very big difference. Our perception of taste is not affected by sophisticated belief systems and social interdependencies.

It is pure intuition, and it works — that is very important point. We don’t need any deliberate logical reasoning to make conclusions about taste of pineapple, we don’t need to ask for advice, we don’t need to die in the name of our perception, we don’t need to vote to recognize the best.

Leading photo by disneymike

No related posts.

Related posts brought to you by Yet Another Related Posts Plugin.

Taking action without thinking is the cause of every failure

When it comes to decisions, I think, making decisions without evaluation is the cause of many failures.

You can’t control the things you can’t measure. The ground truth.

Once you have started a project, which goal is more than a waste of time, you should take care of evaluation.

Here are the main steps of effective evaluation process (most basic and very quick):

• Define the goals of project.
• Define measurable metrics for each goal. Metrics can be direct, e.g. weight measuring, and indirect, e.g. IQ test.
• Define the desired results via chosen metrics – exact values that indicate that a certain goal is met.
• Define the factors that influence each of metrics, e.g. food consumption influences weight.
• Identify the factors you can control. Can you control your daily food consumption?
• Define the periods between evaluations.
• Set up a kind of database for storing the measurements – it can be a Moleskine or MySQL – whatever you prefer.
• Set up a convenient evaluation environment, e.g. if you’re doing the evaluation twice a week at monday and thursday evening, make sure it is the time you are waiting for – prepare a coffee, favorite music, etc.
• Use convenient tools. Dealing with plain numbers can be truly boring. Find the tools for creating eye-candy graphs, tables, for doing statistical analysis in one-click.
• If you don’t like numbers, hire someone who likes, the analyst.
• Carry out atomic experiments to find out the most significant factors.
• Do not tweak too frequently. Each decision must be supported with statistically valid data.
• Each time you tweak something, make sure you keep a record of your changes.
• From time to time think of your goals, metrics and factors – look for something to add, also remove irrelevant ones.
• Look what other people do, what kind of tools they use, what do they measure, learn from the best practices.
• Keep eye on technology and technology will be your ally.

Proper evaluation is a good foundation when you are about to make decisions about your project. Not only it provides you with data to rely on, but also it gives you a clear view of your project’ state.

Imagine the improvements you can achieve if you had the perfect evaluation system by your side. It is quite useful to write down these improvements and start implementing your perfect system. Not such a difficult task, once you care about it.

Stay tuned. In upcoming posts I will explain each of the points above in detail.

No related posts.

Related posts brought to you by Yet Another Related Posts Plugin.