Guessing the size of your user base
Without  download statistics
Step 1.  Define what you want to know 1.  Number of users (2.  Potential for growth)
Step 2.  Run a questionnaire <ul><li>What country are you in?
How many people are on your team?
How many of those people are writing code with Catalyst? If there are non Catalyst coders on your team, how many of the wh...
How many people using Catalyst on your team are subscribers to the Catalyst mailing list?
How many people writing Catalyst code on your team use the #catalyst IRC channel on irc.perl.org?
What do you see as potential for growth of Catalyst in your organisation? How many people do you think will be using Catal...
Step 3.  Remove stupid questions <ul><li>What country are you in?
How many people are on your team?
How many of those people are writing code with Catalyst? If there are non Catalyst coders on your team, how many of the wh...
How many people using Catalyst on your team are subscribers to the Catalyst mailing list?
How many people writing Catalyst code on your team use the #catalyst IRC channel on irc.perl.org?
What do you see as potential for growth of Catalyst in your organisation? How many people do you think will be using Catal...
Step 4.  Advertise <ul><li>Mailing list
Upcoming SlideShare
Loading in …5
×

How to estimate the size of your userbase without download statstics

609 views

Published on

Lightning talk, osdc 2008, sydney

Published in: Technology
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
609
On SlideShare
0
From Embeds
0
Number of Embeds
5
Actions
Shares
0
Downloads
9
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

How to estimate the size of your userbase without download statstics

  1. 1. Guessing the size of your user base
  2. 2. Without download statistics
  3. 3. Step 1. Define what you want to know 1. Number of users (2. Potential for growth)
  4. 4. Step 2. Run a questionnaire <ul><li>What country are you in?
  5. 5. How many people are on your team?
  6. 6. How many of those people are writing code with Catalyst? If there are non Catalyst coders on your team, how many of the whole team would you like to be writing Catalyst code?
  7. 7. How many people using Catalyst on your team are subscribers to the Catalyst mailing list?
  8. 8. How many people writing Catalyst code on your team use the #catalyst IRC channel on irc.perl.org?
  9. 9. What do you see as potential for growth of Catalyst in your organisation? How many people do you think will be using Catalyst in your organisation in 12 months? In 2 years? </li></ul>
  10. 10. Step 3. Remove stupid questions <ul><li>What country are you in?
  11. 11. How many people are on your team?
  12. 12. How many of those people are writing code with Catalyst? If there are non Catalyst coders on your team, how many of the whole team would you like to be writing Catalyst code?
  13. 13. How many people using Catalyst on your team are subscribers to the Catalyst mailing list?
  14. 14. How many people writing Catalyst code on your team use the #catalyst IRC channel on irc.perl.org?
  15. 15. What do you see as potential for growth of Catalyst in your organisation? How many people do you think will be using Catalyst in your organisation in 12 months? </li></ul>
  16. 16. Step 4. Advertise <ul><li>Mailing list
  17. 17. Perlmongers
  18. 18. Perlbuzz </li></ul>
  19. 19. Best practice: Send out three reminders Our practice: Send out no reminders Best practice: Concentrate on hard to reach groups Our practice: Just collect data quickly Step 5. Wait for responses
  20. 20. Number of responses 153
  21. 21. Number of potential respondents Asked for one response per team. Median team size 4, 1000 mailing list subscribers 88% of teams have 2 or less subscribers
  22. 22. Reached from: RSS/NNTP users Perlbuzz Perlmongers But what about non subscribers?
  23. 23. How many users across the list? 3.4 users per mailing list subscriber 1000 subscribers = 3400 users
  24. 24. Naïve estimation 3400 + 20% extra
  25. 25. Correction for Non-Response Bias (this is a technique from marketing) Solution: Pull numbers out of the air. 50% inflation? 100% inflation?? Run another survey with more aggressive recruitment to get a better estimate of NRB
  26. 26. Final estimate We know about 3400 users connected to subscribers. We know that this is at least a 20% under-estimate
  27. 27. 4000-7000 serious users + casual users Final estimate
  28. 28. Next year Validate our estimate with a more extensive survey.

×