Day 96: Learning-Phase Reset And Doubling Down On Creatives

Day 96, and it has been one of those slightly frustrating days where a well-intentioned change has sent the campaign backwards, at least in the short term.

I made some adjustments yesterday to try to bring CPM down and ended up pushing the new funnel back into the learning phase – exactly the state I’d worked hard to escape.

Yesterday’s numbers were okay but not spectacular.

We finished at $668 in sales, with ad spend of £399 (around $518), leaving a profit of about $140.

That is technically profitable but nowhere near the doubling-of-ad-spend outcome I’m aiming for.

Today, by about 1:45 p.m., we were sitting at $319 in sales, so the day could still go either way.

The bigger issue is not today’s revenue figure, but the instability caused by being back in learning.

The change that triggered this was my attempt to broaden targeting and give Meta more room to find cheaper traffic.

I duplicated the ad set, expanded the geography, opened up gender, and removed interest targeting.

In hindsight, that level of change was significant enough to force the new ad set back into learning, even though it lives inside the same campaign.

Early metrics suggested CPM was heading in the right direction, but the cost per purchase is still too high and needs to come down substantially.

To offset this, I have pushed harder on creative volume.

The new funnel now has nine ads running, including several fresh short-form videos that speak directly to beginner guitarist pain points, such as sore fingers.

The idea is to position myself as just one step ahead of the complete beginner, sharing small but real lessons I’ve learned.

These new videos are already showing strong engagement numbers; now it is a question of whether that engagement can convert into affordable purchases.

What You’ll See

• Yesterday’s sales of $668 and ad spend of £399 ($518)
• Profit of around $140, just scraping into the black
• Today’s mid-afternoon revenue at $319
• How a targeting change pushed Funnel 2 back into the learning phase
• Details of the new, broadened audience: more countries and both genders 50+
• Removal of guitar interest targeting to open up the ad set
• Current cost per purchase still too high and target range for CPA
• Total of nine creatives now inside the Funnel 2 ad set
• Two new short-form videos added yesterday, with more in the queue
• Positioning strategy of “average guitarist just beyond beginner”
• Early engagement and CTR signals from the new videos
• Plan to upload more of the seven or eight recorded clips over the next days

Strategy Breakdown

The main strategic event today is the unintended reset of the new funnel’s learning phase.

My goal was to reduce CPM and give Meta more freedom to optimise by broadening the audience and simplifying targeting.

Instead of editing the original ad set, I chose to duplicate it and adjust the new version.

That felt safer than tampering with something already working, but the combination of changes was large enough to re-trigger learning.

Originally, Funnel 2’s ad set was focused on the US and UK, targeting men aged 50+, with an interest layer around guitar.

The new version covers the US, Canada, UK, Europe, Australia, and New Zealand, targeting both men and women aged 50+, and removes interest targeting entirely.

In other words, it is almost fully broad, with geography and age as the only constraints.

The logic is that the historic data in the campaign should allow Meta to find the right people within that big pool, and that a broader auction will often produce cheaper CPM.

The early signs on the upper funnel metrics are encouraging.

Looking at the last seven days, CPM appears to have dropped by around a third, which is exactly the direction I wanted.

The problem is that cost per purchase is still too high. I want that CPA to sit closer to £15–£20 at most, and right now it is still significantly above that.

Because the campaign has been pushed back into learning, that CPA volatility is to be expected, but it is not comfortable.

The key now is to avoid making further major changes and allow the system to stabilise around the new audience definition.

Creative volume is the other half of the strategy.

I now have nine creatives running in Funnel 2: long and short videos, multiple static images, two carousel variations (one starting with a video, one all static), and the classic pack shot that has proved to be the strongest converter.

Yesterday, I added two new short-form videos and recorded a batch of seven or eight more to roll out over the next few days.

Each of these new clips addresses a specific beginner issue – like sore fingers – and then gently introduces the “Your First Song” course as the next step.

This “average guitarist” positioning is deliberate. Instead of presenting myself as an expert shredding at a level most beginners cannot imagine, I’m deliberately leaning into being just a little further along than they are.

That makes the transformation from where they are now to where I am feel realistic and achievable.

Early metrics support this: the new videos are picking up strong engagement rates and healthy click-through numbers in under 24 hours, and even gathering early likes and interaction.

If CPM continues to fall and these videos remain engaging, they should begin to drive cheaper landings and, hopefully, lower CPAs.

Focus

My focus today is on letting the broadened ad set settle back out of the learning phase while feeding it a larger library of short, relatable video creatives, so that CPM continues to fall and cost per purchase moves into a sustainable range.

Insight

Sometimes the price of a better long-term structure is a short-term step backwards.

Re-triggering the learning phase feels like a mistake because the immediate numbers usually look worse.

But if the change gives the campaign a broader, healthier foundation – more creative options, a bigger audience, lower CPM – then it can be worth absorbing a few unstable days.

The key is not to panic and start frantically changing things again.

Once you’ve made a structural shift, the smartest thing you can do is keep feeding the system good creative and give it enough time and data to recalibrate.

........................

Follow my daily journey to $240K as I build, test, and scale profitable funnels.

If you would like to learn how to create powerful Facebook Ads, check out my Course: https://www.jonathanhowkins.com/the-course

And if you're looking for a platform to build your sales funnel, the one I use and recommend is Systeme.io. You can set up and start making sales with their free account:

https://systeme.io/?sa=sa005182666557b5ccf999dfc8d9877207dc47da35

And you can ask questions and get support in my Facebook Group here: https://www.facebook.com/groups/coursecreatorads