5 ms·

I'm CTO at a local newspaper chain. I ran a seminar on how to digital advertising, then ran $500 in advertising to promote the event -- all as a test to see whi

I'm CTO at a local newspaper chain. I ran a seminar on how to digital advertising, then ran $500 in advertising to promote the event -- all as a test to see which advertising mediums I could use to get people to show up to learn about advertising. Here's the data and a summary: https://docs.google.com/spreadsheets/d/1DsJYUYETwQ7a-hISFPcCI7ojGtk2WtyGdxwQXSN6szQ/edit?usp=sharing https://docs.google.com/spreadsheets/d/1DsJYUYETwQ7a-hISFPcC... http://www.davisenterprise.com/special-projects/how-to-waste-less-money-doing-digital-advertising-next-time/ http://www.davisenterprise.com/special-projects/how-to-waste...

searine 10y ago
This is some incredible data. It's really refreshing to see someone share it.
NicoJuicy 10y ago
Nice summary, thanks for sharing!
rattray 10y ago
tldr?
- r00fus 10y ago
  Fta: "What the ROI chart says is that the return for the social ads was zero. They were a complete waste of money. They generated a lot of “action” but it was at best the wrong group of people, and at worst some type of fraud. This held true for facebook.com and its audience network. Ditto Google search ads, display ads and the Twitter campaign."
  - johnking 10y ago
    "Ditto Google search ads" checks article He spent less than $5 on Search ads... Hard to be drawing conclusions with that data set.
jackgolding 10y ago
I'd want to see the ad copy to believe this. The remnant inventory may have cost $2.00/$4.00 CPM but if the ad copy is highly relevant to the readers of the particular site (i.e. listen to our CTO talk about digital marketing) this is quite deceptive to draw industry wide conclusions based on this - some other readers have questioned the sample size which I agree with.
tn13 10y ago
I am not surprised at all. But the numbers you have posted are indicative of nothing. $500 is not statistically significant to show any results on Facebook. Most of my customers spend at least $10K for tests like these.
babuskov 10y ago
Paid $200+ for e-mail marketing? What does that actually mean? You bought someone's mailing list?
- manuelflara 10y ago
  Or paid someone to create content.
- robmcvey 10y ago
  Or paid for an email delivery service? (such as MailChimp)
cerved 10y ago
$500 is not enough or a serious test, esp if you divide it up between multiple channels.

searine 10y ago

This is some incredible data. It's really refreshing to see someone share it.

NicoJuicy 10y ago

Nice summary, thanks for sharing!

rattray 10y ago

tldr?

r00fus 10y ago
Fta: "What the ROI chart says is that the return for the social ads was zero. They were a complete waste of money. They generated a lot of “action” but it was at best the wrong group of people, and at worst some type of fraud. This held true for facebook.com and its audience network. Ditto Google search ads, display ads and the Twitter campaign."
- johnking 10y ago
  "Ditto Google search ads" checks article He spent less than $5 on Search ads... Hard to be drawing conclusions with that data set.

jackgolding 10y ago

I'd want to see the ad copy to believe this. The remnant inventory may have cost $2.00/$4.00 CPM but if the ad copy is highly relevant to the readers of the particular site (i.e. listen to our CTO talk about digital marketing) this is quite deceptive to draw industry wide conclusions based on this - some other readers have questioned the sample size which I agree with.

tn13 10y ago

I am not surprised at all. But the numbers you have posted are indicative of nothing. $500 is not statistically significant to show any results on Facebook. Most of my customers spend at least $10K for tests like these.

babuskov 10y ago

Paid $200+ for e-mail marketing? What does that actually mean? You bought someone's mailing list?

manuelflara 10y ago
Or paid someone to create content.
robmcvey 10y ago
Or paid for an email delivery service? (such as MailChimp)

cerved 10y ago

$500 is not enough or a serious test, esp if you divide it up between multiple channels.