mysociety / alaveteli-experiments

A collection of experiments and research to learn about what makes an Alaveteli grow
3 stars 0 forks source link

Is advertising Alaveteli sites using positive imagery/wording more effective than using provocative imagery/wording? #113

Open Gemmamysoc opened 7 years ago

Gemmamysoc commented 7 years ago

Q: We want to test out which adverts are more effective in bringing new users to Alaveteli sites: those that describe the Right To Know in a positive way, or those that describe it in a more provocative way.

Explain how you will measure this

Set up Facebook or Twitter ads for RightToKnow (Australia), AskYourGov (Uganda) and AsktheEU comparing positive adverts with provocative adverts.

What do we need to do to be able to take the measurements?

Stats on Facebook Ads Manager and stats on Google Analytics. Customised URLs need to be used for the adverts, so we we can track results in Google Analytics.

Adverts set up

Positive adverts:

righttoknow_facebook_ad_positive_dec2016

askyourgov_facebook_ad_positive_dec2016

Provocative adverts:

righttoknow_facebook_ad_provocative_dec2016

askyourgov_facebook_ad_provocative_dec2016

Results so far: Positive adverts more effective

RightToKnow (Australia):

Positive advert was a lot more effective than negative one

Results from Facebook Ads Manager:

Positive advert was most popular with a reach of 6,596 and 96 link clicks = 1.45% click through rate

Negative advert had a reach of 1,606 and 12 link clicks = 0.75% click through rate

Results from Google Analytics:

The positive advert bought 116 sessions to the site and 83 new users. These users then went on to make 78 goal completions: 2 new requests, 5 visits to their own requests, 22 request annotations, 22 updates of statuses, 22 request reports, 5 messages to users. This is a goal conversion rate of: 67.24%

The negative advert bought 15 sessions to the site and 11 new users. No goals were completed.

Comparison to Right To Know Day adverts experiment (which ran for one day and cost the same amount of $50): This experiment which was over a few days and cost the same amount bought 131 sessions compared to 69 sessions on Right To Know Day, so it’s worth spending the same amount but allowing the ad to run over several days.

AskYourGov (Uganda):

Positive advert was a lot more effective than negative one

Results on Facebook Ads Manager:

Positive advert had a reach of 59,481 and 4,247 link clicks = 7.14% click through rate

Negative advert had a reach of 18,958 and 1,493 link clicks = 7.87% click through rate

Results on Google Analytics:

The positive advert bought 1437 sessions to the site and 1074 new users. Bounce rate is 74.81%. No goals are set up in AskYourGov’s Google Analytics so goal conversations couldn’t be tracked.

The negative advert bought 476 sessions to the site and 324 new users. Bounce rate is 79.83%. No goals are set up in AskYourGov’s Google Analytics so goal conversations couldn’t be tracked.

Comparison to Right To Know Day adverts experiment(which ran for one day and cost the same amount of $50): This experiment which was over a few days and cost the same amonut bought 1906 sessions compared to 476 sessions on Right To Know Day, so it’s worth spending the same amount but allowing the ad to run over several days.

AsktheEU:

TBC, still need to set up Twitter adverts.