August 20, 2018

I ran $2 of Facebook Ads for 12 hours

The first Facebook advertisement was created from a quick link provided by FB.

SETUP - Ad #1 I chose to optimize for link clicks to my site https://www.portfoliospreadsheets.com and a daily budget of US$2.00.

It allowed choosing the targeted geographic locations, gender, and age range.

I chose Men and Women, aged 25-50 in Netherlands, Sweden, Finland, , Belgium, Russia, Iceland, Lithuania, France, Germany, Estonia, Latvia, Italy, Portugal, and India.

These countries were the non-US/Canada/UK ones that supported realtime stock quotes from Google Finance.

The ad text was, “Your stock portfolio performance at a glance. Global markets supported. Track your capital gains. Account for cash and dividends. 10 portfolios for $5 per month. Try for free.” It had a simple logo image with part of a newspaper image.

I stopped the ad after about 6 hours.

RESULT - Ad #1 - It cost about $1.00. I reached about 1400 people. 6 clicked the link. 2 liked the link. Cost per Link Click = 100/6 = 17 cents. 100% of the ads were located in India.

CONCLUSIONS - Ad #1 - If you optimize for link clicks, FB tries to maximize impressions. In doing so, it targets the most inexpensive country to advertise in.

SETUP - Ad #2 - I found a different FB link to create an ad that lets you choose an Audience and Placement. For the audience I chose people with the interest ‘Value Investing’. For placement I chose not to run Messenger or Instagram ads. The budget is $2.00 per day and it will run for one day.

For the other targeting parameters I chose Men and Women, aged 25 to 55, in India. It had the text, “Your stock portfolio performance at a glance. Bombay Stock Exchange supported.” Followed by an image of a spreadsheet.

I report the results after 12 hours below. The ad is still running after some modifications.

RESULT - Ad #2 - For $0.48, I got 946 impressions and 18 clicks making it 3 cents per click. All but one ad placement was on the Mobile News Feed.

CONCLUSIONS - Ad #2 - Indians are quite willing to click. However, my SaaS works best on the desktop so I need to target desktop only. I found how to change that easily using the iOS Facebook Ads app.

SETUP - Ad #3 - Ad #3 was run in parallel with Ad #2.

Audience was set to the interests ‘Value Investing’ and added ‘Warren Buffet’ because somehow FB said the audience was too small for an ad. The budget was set to $2.00 per day and set to run for one day.

Placement was on Mobile and Desktop News Feeds.

I chose Men and Women, aged 25-50 in Netherlands, Sweden, Finland, , Belgium, Russia, Iceland, Lithuania, France, Germany, Estonia, Latvia, Italy, and Portugal. (I excluded India.)

The text and image were the same as Ad #2 except it said “European stocks supported.” instead of “Bombay Stock Exchange supported.”

I report the result after about 12 hours of running the ad. The ad is still running after some modifications.

RESULT - Ad #3 - For $0.54 I got 218 impressions and 2 clicks from making the cost per click 27 cents. All the ads were placed on the Mobile News Feed. Various regions were placed with ads. I haven’t found a report that neatly organizes by country but Italy, Belgium, and Portugal seem to have the most placed ads.

CONCLUSIONS - Ad #3 - The EU is more expensive to target leading to lower impressions and higher cost per click.

FINAL THOUGHTS So far there have been no conversions. My landing page asks visitors to signup to beta test the service. 20-something visits is not enough to draw conclusions about the conversion rate.

Also, it’s useful to add something like /?ref=fb_eu to the end of the URL supplied to Facebook to track clicks yourself. However, something is hitting these links (maybe the CDN), and Facebook itself will test the link. Facebook is easy to detect in the logs because it has an IPv6 address with ‘face:b00c’ in it.

After some practical experience with Facebook Ads, I may start a Udemy course on Facebook advertising.

Thanks for reading! Any advice (landing page, advertising) would be appreciated.


  1. 4

    Also, it’s useful to add something like /?ref=fb_eu to the end of the URL supplied to Facebook to track clicks yourself. However, something is hitting these links (maybe the CDN), and Facebook itself will test the link. Facebook is easy to detect in the logs because it has an IPv6 address with ‘face:b00c’ in it.

    Just use Google Analytics and UTM tags for this instead of looking through logs

    https://ga-dev-tools.appspot.com/campaign-url-builder/

    1. 1

      Nice to know! One reason I look at logs is that I respectfully turn off Google Analytics tracking until they acknowledge the Cookies/Terms/Privacy Policy dialog. This is a requirement of Europe's GDPR law. In theory they can fine any company globally that interacts with EU residents.

      1. 2

        That’s incorrect RE GDPR

        Consent isn’t required for Google Analytics as it’s not personally identifiable data

        1. 1

          Maybe you're right. Not for GDPR. But it is needed for EU's Cookie Law. Google Analytics sets a cookie so I have to block GA until I have consent. https://www.cookielaw.org/google-analytics-eu-cookie-law/ also this link says IP address which is sent to Google is protected under GDPR: https://www.blastam.com/blog/5-actionable-steps-gdpr-compliance-google-analytics

          1. 1

            Firstly, I'd be mindful of advice from websites trying to push cookie consent software as they aren't exactly impartial

            It's up to you obviously but the ICO (UK) have said they are going to take a risk-based approach which means nobody is going to get in trouble for first-party analytics cookies. Especially if you have a clear privacy policy and aren't processing personally identifiable information.

            In fact, the ICO website themselves have an opt-out cookie banner rather than requiring explicit consent.

      2. 1

        All good.

        If you're planning on using more FB ads then you should definitely install the FB pixel so you can get far more data on the effectiveness of the ads and remarket to users who have interacted with your site.

        For example, you can show FB ads to users who have signed up but may have forgotten they signed up (very very very common use case)

        https://www.facebook.com/business/learn/facebook-ads-pixel

  2. 2

    Have you tried lead ads?

    Just found a random link to an article that describes them.

    https://sleeknote.com/blog/facebook-lead-ads

    1. 1

      I signed up for some free content on sleeknote.com and there's some good stuff there. Thanks for the link.

    2. 1

      Thanks, it may be useful for some people. However, I don't have the content to nurture leads by email.

      1. 1

        couldn't you just send the login details.

        1. 1

          They login with Google Sign-in because there are some permissions they have to give so the server can generate spreadsheets for them. This is another point of resistance - some people click the login link but then don't give permission/sign-in.

  3. 1

    One thing about facebook is that people is not into facebook to catch your ad (unlike google adwords where people actually searched for your keyword), they are there to see fun memes, or friends and family photos.

    thus a fecebook user is not "primed" into the idea of purchasing your product when scrolling the timeline.

    I found success (one cent a click) when my ad was actually engaging. A picture of a spreadsheet is not that catchy. Try something fun instead? how about a controversial question? and don't forget to re-target your audience with a facebook pixel.

  4. 1

    IPV6 address the starts with "face:b00c"?! Wow this is so cool!

  5. 1

    Thanks for this. I would love to engage with Facebook ads but I resent the fact that they force you to create a personal Facebook account to do so. This seems like madness to me.

  6. 1

    This is really useful. I'll also try little micro FB ad experiments!

  7. 1

    Thanks for this post. It was interesting to see how you went about it. I've run ads in the past for smaller projects with budgets of $50-$100 and had them hyper-localised. I've had some decent success with that method.

    I think maybe your spread + dollar amount will be too low to see any real results. It's a good starting point though.

    1. 1

      You're welcome. Yes, I've been told to create a narrow customer profile listing all the targetable attributes like city, interests and then customising the text to current interested topics like "Like The Avengers? Then..."

      1. 1

        While this is true, if you're targetting too tightly then the ads just won't ever be displayed. You should rarely use targetting that has an audience of less than 50,000.