Lesson Tag: IMP

“DON’T BITE OFF MORE THAN YOU CAN CHEW”

• Retargeting: Create custom audiences, exclude conversions, target with ads to overcome common objections & catch the “ones who got away”
• A discussion about manual bidding
• A discussion about campaign budget optimization (CBO)
  • 10-12 ad sets w/ min. 200-300/day budget, 3-4 ads per ad set (ALWAYS contains a proven winner), larger audiences, & be VERY careful of overlap
• This is not intended to be comprehensive or even used in this form — just a preview of what lies ahead

Feeding your cash cow

• After several weeks of following these practices, your pixel will get smarter and smarter
• Assuming you don’t corrupt or “cross pollenate” the data, you can use this to your advantage
• Launch test ads to larger interest and demographic-based audiences, and your pixel will be MUCH more effective than when it was brand new
• Sometimes you can even revive test ad sets that failed early on

Taking winning audience:offer matches abroad

• This alone can take one “winner” to 5-6 figures of profit
• Apply the same principles and tests
• Scale your winners via cloning and LAAs
• Dig deeper and expand further into every country that shows profit potential
• SOMETIMES you can take CAs from one country to build LAAs from others, but not always — it all comes down to testing and monitoring your data

• Digging deeper into LAAs…
  • As you spend money to collect data, you want to hone in on valuable groups:
    • Website clickers
    • VC (viewed content) for particular pages/products
    • Leads
    • Adds to cart
    • Checkout initiations/complete registrations
    • Purchases
• For a “young” pixel, transition out of cold traffic into LAAs as soon as you can
  • This changes as you get a REALLY developed pixel, but early on…
  • Once you’ve accrued enough valid, valuable data — transition into LAAs
  • Create a custom audience for those valuable “groups”
  • Then create lookalike audiences for 30, 90, and 180 days using that CA as a seed
  • Stick to the country you’ve gathered the original data from
• How many LAAs should I create?
  • As many as you can assuming the “data set” you’re using as a seed is large enough
  • At least 300-500 members, but ideally over 1,000
    • 1,000 website clickers, 1,000 adds to cart, etc.
  • Begin launching test ad sets using the same rules & daily budget as your main tests
  • Let Facebook’s algorithm and insights do the heavy lifting for you
    • You’ve paid for the data — LEVERAGE IT!

Scaling simply…

• Reminder: All you need is breakeven
• Don’t adjust your winning test ad set — let it keep running
• Clone that ad set and increase the clone’s budget by 50 – 100% per day
• Do this until you find the “threshold,” and let it run as long as it remains profitable

Picking up where we left off…

• The hardest part: finding a winning audience:offer match
• Scaling those campaigns can be complex, but it’s a blast
• A ton of ways to do this
• No such thing as “perfection” — what I’m teaching here are time-tested, straightforward strategies

• How do you know when something has succeeded or failed?
  • If it’s just absolutely TANKING with an astronomical CPC (relative to each industry) or really low CTR, just turn it off even if it’s only been 24-48 hours
  • If you’re running your ad sets at $5/day, let it run at least 5-7 days with decent metrics before “calling it”
  • $10/day per ad set with decent metrics = “call it” after 3-4 days
• Profitable (Revenue > CPA & COGS), or at least breakeven with one of your ad sets?
  • Congratulations! You may be on to a winner!
  • This is where things can get REALLY fun
  • We’ll discuss how to expand from here in the next section ?
• Not profitable on at least one of your ad sets?
  • Form hypotheses based on your data to determine what went wrong
    • Not targeting your audience accurately?
    • Not a compelling enough offer?
    • Not a well-optimized landing page?
    • Not good enough ad creative?
  • Then wash, rinse, and repeat to test your hypotheses until you find a winner
    • This is why it’s so important you have money to spare
    • It’s distinctly possible that it will take thousands of dollars in testing & revising to get your first “winner”
    • But once you find one, assuming your audience is large enough, it can be worth tens of thousands (if not hundreds of thousands) of dollars as long as you keep “striking the piñata” the right way

What makes FB advertising veterans so successful?

• They know that NO MATTER WHAT…
• Even if they’re “losing money”…
• They’re always buying data that is well worth its cost as long as it’s used correctly
• Regardless of whether you use it to rule something out or you’ve stumbled on a winner you want to expand… it’s still a step forward and money well spent.

Putting together these pieces to launch your own campaign

• Remember: Only spend money you can afford to comfortably not see financial ROI from
• Create a new campaign — set your objective (I strongly recommend conversion events vs. traffic or engagement)
• Create your first ad set ($5 or $10/day each) — compile based on the audience interests and/or demographics you identified with the Audience Insights Tool
• Add at least one creative (headline/text/imagery or video), and you’re off to the races!
  • A note about adding multiple creatives to the same ad set
  • A quick note about “Dynamic Creative” as an alternative to the above
• Repeat for as many campaigns & ad sets you want to create
• Let’s look back at a couple of my accounts for examples