jeorj_euler's profile

8K Messages


183.6K Points

Sat, Aug 19, 2017 12:08 AM

Why doesn't IMDb staff ever consult with the contributor base before making changes?!

It's a simple question, but I'm prepared to give examples of "upgrades" made since the beginning of this year. (I also acknowledge that once in a rare while, contributors are consulted.)


Official Solution



4.4K Messages


143.6K Points

4 y ago

First for context please see and 

We are listening to contributors and non-contributing customers all of the time. This feedback includes: reading posts on Get Satisfaction; reading contacts at the IMDb help desk; conducting online customer surveys; conducting off-line focus groups; invite only beta testing programs; along with detailed analyses of traffic patterns and even analyses of contribution patterns. IMDb is built by our customers (who include our staff team) for our customers (who include our staff team). We are also running a continuous stream of experiments where different customers see different versions of features while we test out variations -- this is an implicit form of consultation giving hard data on what works and what does not. 

The challenge is that we have hundreds of millions of customers worldwide and hundreds of thousands of contributors worldwide. Each individual has different needs and expectations.  We have to take the fixed time and resources which we have available to meet those needs / expectations as best we can. It is not easy, but it is very rewarding when it works. We are by no means perfect, and we often make mistakes, or have to take inefficient detours on the path to where we are ultimately heading. I like to think that we are always learning and improving along the way and the customer feedback tends to support that -- I recommend reading our annual letters to contributors from 2016 back to 1995 which are archived at for a bigger picture, long term view.

The top piece of feedback we receive from contributors is that they want their contributions processed as quickly as possible.  The top pieces of feedback we receive from other customers is that they want IMDb to be as complete, accurate and as easy to use as possible. This is good because all of those things are aligned, and are the core part of our mission. We currently have a record number of contributors, record volumes of contributions, and yet we are processing data faster than ever.  The processing times page at is a joy to behold these days -- we are approaching 1,000,000 updates per week and yet look how consistently low the unprocessed item count has become. We have people around the world working seven days per week (in shifts, nobody works all seven days :-) to achieve this. 

So far, so good. I think I can now start to get at the spirit of your question even though I disagree with the exact wording.  IMDb is a complex system which has evolved over the last 27 years -- this is both good and bad news. The good news is we are still here and still growing -- we are getting the IMDb database published and accessible on more services, devices and in more countries than ever before; from a contributor perspective this means your contributions have an impact on more people than ever before.  The bad news is that 27 years has given us plenty of time to build up old software systems / processes which need to be updated; for example, you will see this in things like certain parts of the web site updating in seconds after a new piece of data has been approved, yet other taking days to catch-up.

To stay on top of all of this growth, we have to evolve and we have to move fast. We feel we have a good picture of what needs to be done and why.  However, we are the first to admit that, from the outside, it is not always easy to see why any single particular change makes sense at a particular time.  Often these things only become clearer over the long term. 

Pulling this together: we do listen to customer feedback, however, beyond the mechanisms listed above, it is often not practical to consult in advance on changes, sorry. The important point though is that IMDb is all software, process and data -- nothing is fixed in stone.  If something is not working in a major way then we can roll back the change or make fast fixes after launch; if something is not working in a minor way then we can build a list of future fixes and work upon them over time. 

I hope this helps.