🥗 Stop Screwing Up Ads

What Changed at Obvi: The Ultimate Ad Creative Shift

AuthorAuthor

Ron Shah & Ash Melwani
May 05, 2024

Hey there,

Welcome back for another bite to chew on.

If you've been with us for a while, you know we've focused heavily on tactical advice about ad creatives, testing strategies, and scaling campaigns. 

While constantly trying to improve, it’s important to take time to look at the past so you don’t make the same mistakes.

“Those who cannot remember the past are condemned to repeat it.” - George Santayana

Before we get into the past, let’s focus on the future out in LA,

It’ll be a little while until we’re back on the west coast, so come join in when you can!

Now, let’s talk about how we used to do things versus how we do them now to help you update your ad game.

Out with the Old, In with the New

Back then: Remember when we thought one creative could do it all?

We tried to sell multiple products, solve several problems, and speak to a broad audience with just one creative. 

We thought if we nailed the messaging, we'd hit gold every time.

That didn’t happen.

Our creatives were scattered. The messaging was broad and often didn’t resonate deeply with anyone.

Our user-generated content (UGC) was scripted and felt fake.

You know the kind we’re talking about, right?

Those “I’m obsessed with this product” ads that nobody trusts.

Then, there were the static ads we relied on with bright colors and bold headlines.

Sure, they delivered the message quickly but started tanking after overexposure.

Now: We’ve learned a lot since then. Here’s how we’re doing it differently today:

  • Persona-Specific: We focus on selling ONE product to solve ONE problem for ONE specific persona. 

We identify distinct customer personas and tailor creatives specifically for them.

  • Authentic UGC: We invest in authentic UGC that resonates deeply with the audience. 

Real customer testimonials that are unscripted (but effective) are golden. 

Invite your customers for a 10-15 minute call or in-person interview where they can genuinely share their experiences.

  • Iterated Statics: We mix and match static ads with high-fi and lo-fi visuals for variety. 

We also test winning static styles across reels, GIFs, and catalog ads.

It breaks down:

  • the hierarchy of our ads

  • what you should be focusing on to create yours and

  • actual Obvi examples

Behind-the-Scenes at Obvi

Back then: When we started, our creative testing process was all over the place.

We launched new tests daily or every other day, creating volatility in the ad account.

We often tweaked campaigns too soon, causing instability. 

Don’t get us started on the spreadsheets we used to analyze performance (it was a nightmare).

Now: We’ve got a much better system for refining our creative process at Obvi.

Pre-Ad Creation

Before creating an ad, we define personas and identify unique challenges. 

Then, we create anywhere from 3-6 ads trying to target those specific audiences. 

We test those 3-6 ads in specific ad sets.

That allows Meta to optimize to target those specific audiences.

Post-Ad Creation

Once those ads are created, we monitor spend levels as well as in platform data such as CTR, CPC, as well as ROAS to get some idea of performance level.

Then, we cross-check that against blended results and attribution inside of Triple Whale.

After that is all done, we iterate and create new concepts.

We spend 20% of our time iterating and 80% of our time on new creatives.

Obvi Testing Methodology

For testing, we have one campaign each, and then we use an ad set for each angle that we want to test. 

We drop 3-6 ads in those ads sets. 

These ad sets are ABOs, and we run them on cost cap.

We set the cap 20% higher than the normal bid amount that's applied across our account. 

If you're not using cost caps, you can still use this setup. 

Just make sure you have a budget set to generate 50 conversions in a 7 day period. 

Otherwise, you're not going to have any significant levels of spend to identify if something is scalable or not. 

  • If something is not getting any spend → this is an angle or creative design isn't strong enough. 

  • If it does get spend (typically on a cost cap, it will start to spend) → it'll probably start off at a high CPA, but then start to come down and level out.

  • If you are running automatic bidding → let this run for 3-4 days and then assess the CPAs. 

If performance is great, move to scale. 

If not, then restart this process and create additional tests to get your desired level of performance.

When evaluating your business's performance, cash will always be king.

Now, let’s move on to scaling.

From Manual Chaos to Meta’s Advantage+

Back then: Scaling budgets was a manual process that led to inconsistent results. 

Each product had isolated campaigns and scaling up often meant starting from scratch.

Now: We’ve developed a structure that works across brands like Obvi, Coffee Over Cardio, and Paw Rangers.

Each product has unique testing campaigns for different creative types run on ABO with 4-6 ads per ad set.

Once we find a winning ad, we use Meta’s Advantage+ Shopping Campaign (ASC) to run broad campaigns with winning creatives. 

Budgets increase by 19% every couple of days for stability. 

If you're spending over $100k/month, add Proxima AI lookalikes to your campaigns for cheaper CAC.

There’s a lot more to cost caps vs. Adv+, but that could be a whole newsletter on it’s own.

Reply back to this email with questions over account structure if we should cover that next!

The Obvi™ Testing Methodology

Here are our notes on how to run tests and figure out when to scale or not.

Check Your Metrics

Always check your blended NC-CPA, NC-ROAS in addition to in-platform metrics.

Are things at a healthy level and hitting targets?

Scale the Winning Ads

In order to scale, you need to know what makes a winning ad.

A winning ad creative is getting meaningful spend and close to 50 conversions on a weekly basis. 

We take the post idea of these winning ads and put them into our scale campaigns. 

If an ad doesn't have significant spend or close to 50 conversions in the 7 day window, we'd be wary about moving that into scale as it wasn't able to achieve the minimum threshold for getting out of learning within testing.

At first you may not see certain ad spend, but if you keep them on long enough, Facebook will adjust and recalibrate and start spending when it when it feels it needs to. 

In terms of scaling budgets, if you're on cost caps, you're basically sending a high budget with a certain bid and letting Facebook do its thing.

If you're running automatic bidding, and we suggest to scale by 19% every couple of days if things look good.

Refine Poor-Performing Creatives

If a creative isn’t hitting, analyze it:

  • Video Assets: Check the thumb-stop rate and hold rate. 

Improve the first three seconds if needed or adjust the hook and body of the video.

  • Static Assets: Analyze the click through rate and outbound CPC. 

Use eye-catching headlines, callouts, and USPs, and rework entirely if needed.

If performance is still poor, kill it.

Tool of the Week

If you're like us, you know it's tough to keep boosting your creativity to come up with fresh ideas.

This year alone, we've made almost 1,000 creatives for Obvi. 

When you're testing ads at that kind of volume, thinking of ideas takes up most of the time. 

Time we'd rather have back to focus on the thousand other things on our to-do list for the day.

That's where Foreplay has really helped out.

We've built swipe files with winning ads from the Facebook Ad Library and TikTok—all with just one click using their Chrome Extension.

This isn’t about copying ideas. We’re getting inspired to create original hits like our “PCOS Belly” ad, which is still performing extremely well.

Since we started using it, we're spending way less time brainstorming and more time refining ideas that work.

Thanks for Reading Along

Hopefully this behind-the-scenes look at how we've evolved our approach can help you update your ad creatives to be better than ever.

If you’re still trying to figure out where to start, remember this: test smart, iterate often, and keep your messaging crystal clear.

  • Persona-Specific Ads: Focus on solving ONE problem for ONE persona with ONE product.

  • Authentic UGC: Invite your customers to share their genuine stories.

  • Iterated Statics: Experiment with high-fi and lo-fi visuals to mix things up.

Use Dynamic Creative Testing (DCT) to find your winners, scale up the best-performing ads, and don’t be afraid to kill what isn’t working.

Let us know how these new tactics work for you, and don’t hesitate to reach out if you need a hand.

Schedule a MentorPass call for 1-on-1 help with Ash or Ron

All the best,

Ron & Ash