mattandrews/dropout-missing-docs.md

## dropout-missing-docs.md

      
    Raw
  

              dropout-missing-docs.md
            
          
    I was really excited to read about Dropout as it seemed great for my project. Unfortunately though the documentation is a bit sparse and, in a couple of spots, wrong. These are the steps I took over a few hours messing around with it to get it to work:
(you'll need pip and virtualenv installed for this to work)

mkdir dropout; cd dropout
pip install gapy
git clone git@github.com:annapowellsmith/dropout.git

At this point you'll need to do some configuration work: in goals.csv you should have something like:
service_name,service_id,goal_id,goal_has_funnel
foo,XXXXXX,X,TRUE

The service_id took me a while to figure out as "your Google Analytics ID" isn't immediately obvious. This is not the UA-XXXXXX-XX number. Go to Google Analytics, select your property and click Admin. In the third column ("View") there's a link to "View Settings'. In there you'll see a "View ID". This is the number you want.


virtualenv ve


source ve/bin/activate


pip install -r requirements.txt


Ignore the dropout instructions about configuring gapy as it turns out you don't need to do anything.


Definitely ignore this line:
python dropout.py -f [GOALS_FILE] -s [PATH_TO_STORAGE_DB]


After agonising trying to create a storage.db file, I looked at the source and it turns out the -s flag is for the SECRETS_PATH, eg. the directory containing client_secrets.json.
You can generate this file at the Google Developers Console:

Create a new Project.
Once created, add the Analytics API to the Enabled APIs list.
In "Credentials", click "Create new Client ID".
Choose "Installed Application" and type should be "Other" (or "Playstation 4" if you're feeling experimental)
Once it's generated your ID, click the new "Download JSON" button and save this file as client_secrets.json (I saved it in the same directory as dropout.py.
Finally, you can run this: python dropout.py -f goals.csv -s .

Congratulations – hopefully you now see a console full of delicious data.