01 / Introduction

First, the receipt. Then the actual news. 

Last issue ended with a prediction: the Labor Day effect would give something back, and I would report it either way. Here is the scorecard. Search snapped back on cue, with Google's nCAC down 4% to $172 and conversion up on both Google and Microsoft. Meta, meanwhile, never gave anything back. Third straight week of falling new customer cost, now at $78. The full grids are below.

But the real story of the week is not the two channels everyone watches. It is the one most brands quietly gave up on measuring. This issue tells that story, then walks the same four verified grids as always: new customer acquisition, first click versus last click, overall channel performance, and new customer lifetime value. All first-party, order-verified, aggregated across hundreds of Wicked Reports accounts. No modeling, no surveys.

02 / Analysis 

05 / Overall Channel Performance

Where the money goes, and THE TRUE COST OF A NEW CUSTOMER 

Look at YouTube's markup column against the search channels. A week ago video carried a 79% penalty between blended CAC and true new customer cost. This week it is 25%, tighter than Google's 58% and Microsoft's 70%. Search still posts the pretty ROAS numbers, and search is still where blended CAC most flatters the real cost of growth. The standing lesson of this grid does not change: the channels that look most efficient on ROAS are the ones borrowing most heavily from demand that already existed. 

06 / New Customer Lifetime Value

What a new customer becomes over a year

This grid is what turns YouTube's cheap week into a real economics question. A YouTube customer starts at $79 and nearly triples inside 30 days to $233, the steepest early payback curve of any paid channel, reaching $261 at one year. Judged on day-zero math, a $360 nCAC against a $79 first order looks indefensible. Judged on the 30-day number, the gap nearly closes on click-credited revenue alone, before any view-through contribution. The standing caveat applies: this grid blends brands at very different price points, so it is a directional market benchmark, not a promise for any one store. 

07 / Conclusion 

Put video on the watch list, with a number attached 

The September story so far: the holiday came and went, search snapped back, Meta keeps getting quietly cheaper, and now video just posted the biggest verified efficiency move of the month while most dashboards handed the credit to the search box. None of that is visible on last click, and none of it is visible on ROAS.

The move this week is specific. Pull your own YouTube new customer cost, verified against orders, not platform-reported. Write down this week's number and the new-customer share next to it. If both hold through next week, you have an early position on a channel most of your competitors stopped measuring years ago. That is the entire advantage: not spending more, just knowing sooner. I will publish what the aggregate did either way.

See your own version of these four grids.

Your real nCAC next to your aCAC, your first versus last click gap, your new customer LTV by channel, in your own account.

Get the Paid Traffic Truth Report directly to your inbox.

How this week's numbers were built. Aggregated across hundreds of Wicked Reports client accounts for the week of September 13 to September 19, 2026, except first click vs last click, which uses a rolling 90 day window. New versus repeat is verified at the order level against first party order IDs, not modeled and not surveyed. The new visit to new customer conversion rate credits the channel that originated the new visit. Channels without cost data, including email, SMS, organic, and influencer, are left out of the cost comparisons. Snapchat is excluded for negligible spend. Pinterest is marked with an asterisk because it ran on a small number of new customers this week, so it is not used to anchor any headline. The channel labeled Facebook in the underlying platform data is shown here as Meta. Charts and tables carry meaning through direction, labels, and contrast rather than color alone.

01 / Introduction

The two CACs, revisited 

Every week we aggregate first-party, order-verified data across hundreds of eCommerce accounts and publish the four grids: new customer acquisition, first click versus last click, overall channel performance, and new customer lifetime value. One story per issue. This week the story is the markup between aCAC, the blended cost of acquiring any customer, and nCAC, the cost of acquiring a customer who has never bought from you before.

We ran this comparison in our very first issue. I am running it again on purpose, because the honest news is persistence. This is not a weekly anomaly you can wait out. It is a structural property of how blended CAC is calculated, and the size of the distortion depends entirely on which channel you are looking at.

02 / Analysis 

05 / Overall Channel Performance

Where the money goes, and THE TRUE COST OF A NEW CUSTOMER 

Read the markup column top to bottom. Every single paid channel costs more per new customer than its blended CAC admits. There is no channel where the blended number is honest. The only question is how big the markup is, and the answer ranges from a rounding error on Meta to more than double on YouTube. If your growth model uses one CAC for the whole business, your model is built on the most flattering version of every channel simultaneously.  

06 / New Customer Lifetime Value

What a new customer becomes over a year

Here is the fairness clause for YouTube. Its new customers start small at $84 and nearly triple to $244 by one year, still the steepest growth curve in the set, a pattern we covered in depth in Issue 008. That back-end value is real, but it does not erase an $826 nCAC. It means YouTube is a channel you evaluate on a payback window, not on week-one math, and definitely not on blended CAC. One caveat that applies to this whole grid: it blends brands at very different price points, so treat it as a directional market benchmark, not a promise for your store.  

07 / Conclusion 

One business, two CACs. Budget off the right one. 

The blended CAC on your dashboard is answering a question you did not ask. You want to know what growth costs. It is telling you what activity costs, with your cheapest, most loyal buyers averaged in to make every channel look better than it is. Nine issues into this series, the markup between the two numbers has shown up every single week, on every paid channel, without exception.

The fix is not more reporting. It is verifying new versus repeat at the order level, pricing each channel by its true nCAC, and then making Scale, Chill, or Kill decisions off that number. Meta at a $88 nCAC and 87% new is a different decision than YouTube at $826 and 39% new, even though blended CAC says they are $77 and $319. Know which number you are looking at before you move the budget.

How this week's numbers were built. Aggregated across hundreds of Wicked Reports client accounts for the week of August 23 to August 29, 2026, except first click vs last click, which uses a rolling 90 day window. New versus repeat is verified at the order level against first party order IDs, not modeled and not surveyed. The new visit to new customer conversion rate credits the channel that originated the new visit. Channels without cost data, including email, SMS, organic, and influencer, are left out of the cost comparisons. Snapchat is excluded for negligible spend. Pinterest is marked with an asterisk because it ran on a small number of new customers this week, so it is not used to anchor any headline. The channel labeled Facebook in the underlying platform data is shown here as Meta. Charts and tables carry meaning through direction, labels, and contrast rather than color alone.

01 / Introduction

One story, four grids, hundreds of verified accounts

Every week I aggregate first party, order verified data across hundreds of eCommerce accounts and publish what the ad platforms will not show you. One story leads, and the same four grids follow: new customer acquisition, first click versus last click, overall channel performance, and new customer lifetime value. No modeled conversions, no surveys, no platform grading its own homework.

This week's story lives in the lifetime value grid. YouTube posts the ugliest new customer cost of any major paid channel, and it also grows the most valuable customer curve of any major paid channel. Both things are true at once, and if your measurement stops at day zero, you only ever see the first one.

02 / Analysis 

05 / Overall Channel Performance

Where the money goes, and THE TRUE COST OF A NEW CUSTOMER 

Meta and Google together carry 91% of tracked spend, so the market has voted for cheap day zero acquisition. Fair enough. But look at the ROAS column with fresh eyes: the best weekly ROAS in the set belongs to Microsoft at 1.89, a channel where only 59% of buyers are new and last click over credits it by 0.46. High weekly ROAS keeps correlating with closing existing demand, not creating new customers. Cheap and compounding are different things, and this grid only shows you cheap. 

06 / New Customer Lifetime Value

What a new customer becomes over a year

Sorted by growth multiple, the video and social channels own the top of this grid while search sits near the bottom. A Google customer arrives spending $164 and grows only 36% in a year. A YouTube customer arrives spending $104 and more than doubles. Pinterest posts the steepest multiple at 2.65x, but it runs on very little spend and very few new customers this week, so treat it as a footnote, not a finding. One standing caveat: this grid blends brands at different price points, so it is a directional market benchmark, not a promise for any one store. 

07 / Conclusion 

Stop grading a compounding asset on its opening day

Every channel in your account is being judged by a number, and for most operators that number is built from day zero revenue and last click credit. This week's data shows exactly which channels that math executes: the video channels acquiring your freshest customers with the fastest growing value curves. Nobody decides to kill their best prospecting engine. They just use a measuring stick that makes killing it look responsible.

The fix is not faith in video. It is measurement that follows the customer past day zero and past the last click. Know your real nCAC by channel, know what a new customer from each channel is worth at 30, 90, and 365 days, and make Scale, Chill, and Kill calls on that, not on a screenshot of week one ROAS. The channels compound. Your measurement should too.

HOW TO TARGET HIGH-INTENT YOUTUBE AUDIENCES BY SEARCH HISTORY IN 2026 (AND PROVE WHICH ONES ACTUALLY PAY OFF)

Most YouTube advertising is a spray of impressions at people who never asked to see you. There is a sharper way. Google lets you show your video ads specifically to people who have recently searched for the things you sell, which is about the highest-intent YouTube audience you can build. But targeting the right people is only half the job. The other half, the half most advertisers skip, is proving which of those audiences actually turned into profitable customers rather than just cheap views. Here is how to do both.

HOW TO TRACK YOUTUBE ADS AND MEASURE REAL ROI FROM VIDEO 

The oldest problem in advertising still bites hardest on YouTube. As the saying goes, half your ad budget is wasted and you just do not know which half. Video makes that worse, because a YouTube ad so often does its work early — someone watches on their phone during a commute, forgets about it, then searches your brand three days later on a laptop and buys. Last-click attribution gives YouTube zero credit for that sale. If you are judging YouTube on what Google Ads reports, you are almost certainly underrating it. Here is how to track YouTube ads properly and measure their real ROI.