Part 3: The B-I-G 10K

This is it... this is what everyone is after... 10,000 supporters! You've been checking your stats every hour, posting on social media, reaching out to long-lost cousins, showering only when convenient... all in a bid to leave no stone unturned on your quest to glory. Let's see what insights we can glean from the data.

---

Click here for a popup on what makes up the data: 

---

Question 1: 

How can I tell if I'm on track for 10K supporters?

Turns out, the best indicator is probably just what you thought it would be; how many days did it take for your submission to reach each a given milestone. But how to arrive at the numbers is somewhat tricky. I ended up using something called a Box and Whisker plot in order to give a range of values. You can read the details AFTER we look at the results first.

Cutting to the chase, the numbers below are probably what you should care about. There are 10K and Non-10K columns and for each milestone there is a range of days. That range represents how many days 75% of the corresponding submissions took. For example, if you look at the1K milestone you'll see a 10K submission normally took 4-13 days while a Non-10K took 27-255. 

Just compare the number of days for your submission (how to is in Question 2 below) and see where you stand. Is it perfect? Of course not, someone who has taken 27 days to get to 1,000 may indeed still get to 10K. However, the chances go down the further away from the 10K range (4-13) it is.

Range of days typically needed for 10K and Non-10K

Items of note:

• There are distinct differences between 10K and Non-10K
• Days needed for 100 supporters gives an initial indication, the longer it takes... the less likely to make 10K
• They're so close together (10K: 0-1 days, Non-10K: 2-23 days) that anything taking from 1-4 days really is not enough to bet the farm on. It's just an indicator that you've gotten off to a really good start.
• Days needed for 1K is really the best indicator... 4-13 for 10K and 27-255 for Non-10K. 
• These numbers do not take into account the effect of Staff Picks; we'll cover that in the next post

If you'd like to know why I chose box-and-whisker read on. Otherwise, you can jump to Question 2.

I examined three different ways of looking at the data... averaging, median, and box-and-whisker. 

Averaging: not-so-good

Averaging all the submissions that end up at a milestone is simple, but it could include outliers (anomalies) which would skew the results. For example, you wouldn't include someone who cramped up and took six hours to complete a marathon in averaging out the top performers. Instead, a better way would be to look at the median of each. The median represents the point where, for a given milestone, half the users are above the number and half are below. For example, consider the following pretend milestone and how many days it took for five different projects:

example: Showing how Median can be better than Average

If we average the values, we get 210, but that's not very common because four of the five entries were below that. The one outlier of 500 has skewed our results! On the other hand, the median is smack in the middle and gives us 150. Much better.

Median: definitely better

So, let's look at JUST the median first in the chart below.

Items of note:

• 10K achievers seem to be achieving 100 supporters before the first day is up, and one thousand after a week... that's pretty quick!
• We also see some wide separations between the days needed for10K versus Non-10K

Question 1 - Median Data:

The median is helpful, but it doesn't tell us what happens the further we get away from it. For example, looking at 1K was can see the 10K median is 7 days, while the Non-10K is 92. But what about all the numbers between them? Say 30 days... is that more in line with 10K or Non-10K? 

Box and Whisker Chart: we have a winner!

This is where a Box and Whisker chart can help us. It accounts for outliers and displays where 75% of the entries occur (the box) and what the minimum/maximum (the whiskers) are for a given milestone. If we're within that 75% range (fancy name is Quartile1 to Quartile3), then chances are good that we're tracking with the corresponding 10K/Non-10K for a final outcome.

Below is just such a box and whisker plot. Let's look at the specific milestone DaysTo_1K (in red). We can see a distinct difference on where the 75% ranges are for 10K and Non-10K entries... 10K achievers reach the milestone much quicker (4-13 days) than Non-10K (27-255 days). We can use this difference to help project what track we are on. The ranges produced by this chart are much more helpful than a simple chart of the medians!

So, the table I showed at the start of this post was really nothing more than showing the ranges (i.e. the boxes) of a box and whisker plot in table form. 

Question 2: 

How can I tell how many days it's been since my submission was published?

We can get this from the submission itself. Login to LEGO Ideas, go to your profile and click the submission. Then follow the steps below. 

1. We can get the published date of your submission (when it went live) in the upper left-hand of the screen.
   
   
   
   
   
   
2. If you've already reached a given milestone, then click the tab for Official LEGO Comments, find the entry for your milestone and note the date.
   
   
   
   
   
3. Now head over to your favorite search engine and enter the following in the search bar:
   Days from <published date> to <milestone date> 
   
   
   
   
   

Hint: you can use the word "today" for the second date to get the
 number of days since your submission was published

In this case, it took 26 days to reach the 1K milestone. So, if we consult the chart from Question 1, we can see this doesn't quite fit 10K or Non-10K. However, it is really close to the Non-10K mark. Therefore, I would say this build (at the time when 1K was reached) was more likely to be a Non-10K submission.

Prolific Success: users with most submissions reaching 10K

Every... Single... Time:  users whose every submission has made it to 10K (top 10)

Lickety-Split: project with shortest time to reach 10K

Slow and Steady: project with longest time to reach 10K

up next: Staff Picks