Day 6: Launching the First Split Test. Quick Strategy Shift.

Yesterday ended at $840 in sales on $416 in ad spend, leaving around $425 in profit. Today, by lunchtime, sales are at just $130—a reminder that daily numbers rise and fall.

The focus now shifts from results to action: the first live split test of this journey.

Sales Recap → A Profitable Day, A Slower Start

• Friday, Sept 5th

  • Revenue: $840

  • Ad Spend: $416 (£307.53)

  • Profit: ~$425

• Saturday, Sept 6th (so far)

  • Revenue: $130 by 1:20pm UK time

  • Sales tracking lower than previous days

Profit yesterday, uncertainty today. But the daily numbers aren’t the real story—the testing is.

Funnel Focus → Shifting Strategy

Yesterday I shared my plan to test a new order bump at $27 in place of my $17 backing tracks. Overnight, I had a rethink.

Instead of focusing on Average Order Value (AOV), I decided to target Cost Per Acquisition (CPA).

Why? Because if I can reduce the cost of winning a new customer, profit margins improve across the board.

So here’s the revised plan:

• Backing tracks → move them from the bump into the main course as a free bonus (boosting perceived value and hopefully increasing conversions).

• New order bump → a Guitar Tone & Effects Setup Guide at $19, sourced from AI-driven product ideas and matched with an existing product I already had.

The aim? Add value to the front-end offer while keeping the bump attractive and affordable.

The Test → A/B In Action

To run the test, I duplicated my main sales page inside systeme.io, my funnel builder.

• Control page: Original offer with the $17 backing tracks bump.

• Variant page: New bonus (backing tracks included in the main course) + $19 Guitar Tone Guide bump.

Traffic from Facebook ads is now split evenly between the two. Over the next week or so, I’ll start to see early indicators of which version is winning. A month of data will make it conclusive.

If the new version works, I’ll see:

• Higher front-end conversion rates (thanks to the added bonus).

• Lower cost per acquisition (CPA dropping below the current $38 average).

• Stable or higher bump take-up, despite the slightly higher price point.

If it fails, conversions could dip, bump take-up could fall, and CPA could rise.

That’s the risk—but that’s also why we test.

“Split testing is the only way to replace guesswork with clarity. You don’t know until you measure.”

Key Lessons (Day 6)

• Don’t get attached to one lever. Improving CPA can sometimes be more impactful than pushing AOV.

• Perceived value matters. Adding bonuses to the core offer can lift conversions without lowering price.

• AI is a creative ally. Product ideas from AI tools can spark fresh, profitable bumps.

• Set it, then wait. A/B tests take time—early results are tempting, but only long enough data tells the truth.

What’s Next

Tomorrow, I’ll shift focus from the funnel to the Facebook ads themselves—the metrics I track daily and the signals that show whether campaigns are on the right path.

I’ll also share what I’m preparing to split test next on the ad side.