Facebook advertising is undergoing the largest change in its history: iOS14.5+.

Adopted users are now 70% of its spend.

Everything. Is. Different.

Here are my 4 recommendations to keep winning in this new world:

(source @ampush’s 8 figs in MONTHLY $ with top DTC brands)
Recc #1 - Test consolidated account structures and narrowed targeting strategies RIGOROUSLY.

The change in privacy means less data for FB. So, 14+ years of an algorithm built on user data has to change and so does your strategy…
Let’s start with an example: before iOS14.5+, Facebook could track “Sara” from her clicks on an ad to her specific activity on a brand’s site, store this data and continue to collect, building up a database on Sara’s patterns.

Now what happens when Sara opts out of tracking?
FB is no longer allowed to directly follow Sara from clicking on an ad to making a purchase.

The data gets sent to Apple before FB even sees it, who then sends back very limited data to let FB know that someone made a purchase from an ad at 11am on June 5th.
So, FB is losing impt user data that makes their algo smart. Without knowing Sara’s patterns, they cant properly deliver her ads that she will engage with and convert on.

Advertisers saw this translate into a drop in conversion rate of >20%, mostly in broader audiences.
So, we began testing and found two things…

First, a consolidated account structure increases signal and therefore, efficiency. When you over-segment your campaigns, there is a higher chance that you compete with yourself and confuse the algo...
...When you consolidate, batching the data into one place, gives FB a better signal of what’s working and what’s not.

Second, narrowing targeting pushes the algo into a specific direction. Since the algo is losing data, narrowing your targeting gives FB a good starting point.
Partner anecdote: CR decreased 20%, so we cut our campaigns in half, even removing broad and narrowed our audiences. This improved efficiency by 20%.
Recc #2 - FB reporting is inaccurate and they are under-attributing conversions, so use another tracking source to improve your optimizations.
FB has been forced to predict your daily reporting instead of giving you actuals. Why?

Because now the data goes straight to Apple who only sends it back to FB after 3 days. So, FB tries to predict what’s gonna happen day-to-day.
The main problem with this setup is that even when Apple sends back the data, it is inaccurate.

This is proven by comparing to Google Analytics and finding they are capturing 35% more conversions than before the update. The implications to this are…
Daily optimizations are challenging bc the data is inaccurate and tracking such as Google Analytics do not give you the level of detail needed to properly optimize.

@ampush has our own pixel so we can track real-time data across campaigns ad sets and ads to optimize effectively.
So figure out a 3rd party pixel solution you can use to track and optimize daily.
Recc #3 - Data is limited not just for FB, but also for advertisers, so segment campaigns by demographics to keep deriving important insights that inform your business decisions.
Before iOS14.5, we could break down all data by demographics, like age, gender, region.

Now, that information is unavailable to us due to the tracking limitations. But, there is a workaround…
Segment your ad sets by the demo you are looking to solve for:

You cant see which age group is performing well on FB. @ampush develops tests where we deliver ads to each demo separately enabling us to see full funnel data for each and re-orient our creative and target mkt strat
You can also do this by device: for a partner at Ampush we excluded OS devices from their Brazil campaigns and drove a 40% higher conversion rate and 30% lower cost per acquisition.

They discovered that Android-only was able to hold scale in this country at higher efficiencies.
Recc #4 - platform bugs are increasing, so keep a close watch, communicate with your FB reps and take necessary action to mitigate performance issues.
Facebook is going through a learning curve as they juggle how to operate with limited data, so platform issues are more common.

The most recent bug was an unwarranted spike in delivery costs on June 11 & 12, where partners saw upwards of a +100% increase in CPM. In response, we
Communicated across partners to ensure this was an advertiser-wide issue, flagged to our FB reps, scaled back spend to mitigate loss and facilitated refunds from FB to the advertisers.
iOS14.5+ is top of mind for all advertisers, but it doesn’t have to be a scary topic, rather another problem to solve. And as #entrepreneurs, that's what we do best.

Follow @jspujji for more top of mind #growthmarketing, DTC, entrepreneurship and facebook.
TLDR
1. Consolidate and narrow targeting; test rigorously
2. Create your own attribution source
3. Segment demographics to uncover insights and improve performance
4. Track platform bugs closely & put a response plan in place
Follow @jspujji @ampush for more!

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with Jesse Pujji

Jesse Pujji Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @jspujji

13 Jul
In the last 100 days, I've gone from 2k to ~22k twitter followers.

Fast right?

Not really if you account for the year prior where all I did was debate the best way to share my entrepreneurial learnings.

Here is:
- What I did
- WHY I did it
- What I learned

👇🏽👇🏽👇🏽
First, my personal WHY: I want to help others learn/grow to be the best versions of themselves through the power of entrepreneurship.

It took me many years of coaching and reflection to realize this is what "I can't not do."

I love entrepreneurship. I love learning/growth.
Earlier this year, I started @GatewayX to live this personal Why.

It's early but the gameplan for now is:

A venture studio/holdco where 1) all the companies are bootstrapped (i.e. get profitable fast) and 2) its a single culture/org. We invest/buy/build. Mostly build.
Read 23 tweets
3 Jul
Is Facebook listening to your conversations?

No, they are not.

They are doing something MUCH more effective!

Here’s how it works👇🏽👇🏽👇🏽
The two most valuable pieces of software on earth are: 1) the $FB pixel and 2) the $FB newsfeed.

When you wonder, how come FB is worth $1T and Twitter is only $55BN, those two pieces of software are your answer.
The FB pixel is a tiny piece of code that nearly every website on the planet has embedded.

It feeds data back to FB (in aggregate, anonymized) for the list of websites visited, how much time was spent, did you buy or not, etc.
Read 13 tweets
29 Jun
A 50 year old immigrant monk bootstrapped a biz from nothing to over $1 BILLION in sales with 30%+ profit margins.

It took only 8 years.

The crazy part?

It doesn't even have 200 employees.

Buckle up for this one 👇🏽👇🏽👇🏽
1) Manoj Bhargava was born in Lucknow, India in 1953.

In 1967, his family moved to the US so his dad could get his PhD at Wharton.

His family went from Indian affluence to living in an $80 per month West Philly apt.

At age 17, he started his first business.
2) He bought a $400 truck and cleaned out debris in rough Philly neighborhoods.

Despite the danger and rough work, he made $600 that summer. After turning his first profit, he was hooked.

In 1972, he got into a little college called Princeton.

A year later, he dropped out.
Read 19 tweets
19 Jun
Over the years, I get 1 consistent Q when it comes to Zuckerberg.

Is he a “visionary” or did he just get lucky?

The answer is visionary.

How do I know?

Bc in 2005, we spoke for an hour and he shared the 10 year $FB plan w/me.

The kicker? It all came true.

Thread time👇
In the 2004, @aniketkshah, @chrisamos and I were amongst the first Facebook users (we were in one of the first 8 schools where it started).

Being ambitious young entrepreneurs ourselves, we thought: let’s start HIGH SCHOOL version of Facebook!
At that time, like many, we believed FB would never go beyond colleges because it would alienate its core user base.

In the summer of 2005, we would work 9am to 11pm at our finance internships.

And then nights and weekends, we worked on building this website.
Read 24 tweets
18 Jun
This is the most successful thing I've ever done.

And it is not directly tied to Business/Entrepreneurship.

If you are a new/expecting parent, it will change your life.

🧵 about how to get your newborn 👶🏾 to sleep through the night by 4 mo old without cry it out or coddling
Some background: when @tweepika and I were fam planning, we met a lot of young parents. We asked questions, listened, observed and learned.

It seemed parental happiness was directly correlated with:
+ how well their baby slept
+ Childcare
+ Alignment with spouse.
Regarding baby sleep (topic of this thread), we were told there are only two approaches, either:

A) At 6 mos old, "cry it out" - put baby in crib at 7pm and don't return until 7am regardless of their cries (there are modified versions)

or
Read 25 tweets
16 Jun
Not every business can be bootstrapped.

For 10 years, this business raised and burned >$2BN with ZERO dollars in revenue.

Then, it became the most important business in the world.

👇🏽👇🏽👇🏽
Today, we released a business breakdown on Moderna

Here are 3 amazing facts about about it:

1) The actual vaccine was developed in less than a WEEK.

Normal vaccines take ~3 years to develop.

The rest of the time pre-release was spent on clinical tests.
2) Moderna purposefully built itself as an mRNA platform vs the “norm” of a single drug therapy.

This platform allows them to quickly code new dna/mRNA sequences via software that within days shows up as a testable physical protein.

This platform approach was key for the vax
Read 5 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!

Follow Us on Twitter!

:(