All language subtitles for 02 - Interpret the results of your analysis

- [Instructor] Businesses run on data.

Which is both good news and bad news.

The good news is that it has never been easier to

collect data about your customers, products and sales.

The bad news is that you still have to decide

what all that data means.

As an example,

suppose you ran a web advertising campaign

with a goal of bringing in at least 70

new qualified customers per week.

You ran the ads on 25 different sites for four weeks.

So you have a sample size of 100.

On average,

each ad brought in 73 new customers

with a standard deviation of 15.1.

The central question is though,

whether the ads met your goal of bringing in

70 new customers with each campaign.

Your sample size is 100 and your mean is 73.

But the standard deviation is pretty large.

Over 15.

So you have to ask yourself one question,

did I just get lucky?

I've used the numbers I collected

to calculate my standard error

and what's called a Z-score.

The Z-score lets me calculate how likely it is

my results were due to chance.

And in this case, again,

based on all my values,

the probability is 2.35%.

And

2.35%, okay.

Is that good?

Bad,

neither?

Most analysts say that a result is significant

if the probability your results were due

to chance is below 5%.

In other words,

if your data would show that the conclusion will be wrong

less than one time out of 20,

then you can claim the result is significant.

So in this case,

97.65%,

or 100% minus 2.35,

appears to be significant past the 95% confidence level.

Of course, there is still the possibility

your results actually were due to chance

and that your ads did not have a meaningful effect.

That's where your knowledge of your business,

intuition and willingness to continue analyzing

comes in handy.