How to Read Your BFCM Data Without Being Misled by Last-Click

Sales look strong. ROAS is up. The channel breakdown looks clean. The problem is that if last-click attribution is your primary lens, you are probably drawing the wrong conclusions from that data.

Last-click attribution gives all the credit for a sale to whichever channel the customer clicked on immediately before converting. During Black Friday/Cyber Monday (BFCM), when purchase decisions are often made weeks before the sale period begins, that final click rarely tells you why the customer bought. It simply shows the last interaction before the purchase, rather than the channels and touchpoints that influenced the decision along the way.

This is how budget decisions go wrong. Email gets credit for a sale it only reminded someone of. Paid Search looks like a machine. Organic and upper-funnel activity look flat. When planning comes around, you optimise toward the channels that closed and away from the channels that built the demand in the first place.

 

Why BFCM makes last-click attribution especially misleading

BFCM is not a normal trading period. Shoppers start building wishlists and researching purchases weeks in advance. They visit brand sites, read reviews, compare prices, and join email lists long before the sale begins. By the time the discount code drops, many of them have already decided what they are buying.

That pre-decision behaviour is exactly what last-click attribution misses. The customer who converts via branded search on Black Friday morning may have been introduced to your brand through an organic blog post in October, nurtured through a welcome email flow in early November, and seen your retargeting ads every day for two weeks before that. Last-click gives all the credit to the branded search term. The other ten touchpoints get nothing.

Retargeting campaigns in particular look extraordinary under last-click during BFCM, because they fire at the precise moment a customer who is already primed to buy is revisiting your site. The campaign appears to convert at a high rate. What it is actually doing is reaching people who were going to convert anyway and charging you for the final impression. The distinction matters enormously when you are deciding where to put the budget next year.

 

Which channels last-click undervalues

The channels that consistently get undervalued by last-click during BFCM are the ones doing the awareness and consideration work: organic search, organic social, editorial content, and upper-funnel Paid Social. Email tends to look like a star performer, but often it is capturing intent that other channels built.

Here is what that looks like in practice. Your SEO content introduces someone to the brand via an informational search in October. They visit the site, do not buy, and join the email list. They receive a welcome flow. They see Meta retargeting. On Black Friday they search your brand name, click a Google Shopping ad, and convert. Last-click reports that as a Google Shopping sale. The SEO and Meta work gets nothing.

This has real consequences. If you reduce SEO investment because it shows poor last-click returns, you remove the channel that introduced the customer. If you scale Google Shopping because the last-click ROAS looks strong, you may be paying to capture demand that was already there regardless of the ad. Neither decision is based on what actually happened.

 

How to read your attribution data properly

Google Analytics 4 (GA4) has better tools than last-click, but they require you to know where to look.

Run a model comparison between last-click and data-driven attribution. GA4 offers data-driven attribution (DDA) as the default model in its advertising reports. DDA distributes credit across touchpoints based on their contribution to conversion, rather than giving it all to the last one. When you run the comparison for your BFCM period, the channels where DDA gives more credit than last-click are the ones doing upstream work. The channels where DDA gives less are likely capturing existing demand.

Check the path to conversion report. In GA4's advertising reports, the path to conversion shows the actual sequence of channels customers moved through before buying. Look at this for your BFCM period. If the same channels consistently appear in the first or second position before a conversion, they are doing meaningful work even if they never close the sale. Count the touchpoints. If the average path is four or more channels long, last-click is hiding most of the picture.

Look at time lag data. In GA4's model comparison report, you can see how long it takes from first touch to conversion. If the majority of your BFCM conversions have a lag of seven or more days, customers were in the funnel before Black Friday started. Someone who first touched your brand six days before converting via branded search is not a customer that campaign created.

Check your Meta attribution settings. The default attribution window on Meta is 7-day click plus 1-day view. That 1-day view setting means Meta claims credit for any purchase made within 24 hours of someone seeing (not clicking) an ad. During BFCM, when purchase rates are elevated for everyone, view-through attribution inflates reported results significantly. Compare what Meta reports with what GA4 shows as the channel of last interaction. The gap tells you how much of Meta's reported performance is view-through credit rather than click-driven conversion.

 

The data point your platform reports cannot give you

Platform attribution data tells you what the tracking pixel recorded. Post-purchase surveys tell you what the customer remembers. Neither is complete on its own, but together they give you a more reliable picture than either does alone.

A simple question at checkout or in a post-purchase email, asking how the customer first heard about you, consistently surfaces channels that platform data underreports. Word of mouth, podcast mentions, editorial coverage, and PR-driven search all show up in survey data and rarely get credit in GA4. During BFCM, when brand search volumes are elevated and many customers are acting on a recommendation they received weeks ago, survey responses can be the only way to understand what drove awareness.

Do not expect the numbers to match. The point is direction, not precision. If 30% of your post-purchase respondents say they found you through a friend or editorial coverage, and your GA4 gives that channel 2% of assisted conversions, that gap is worth investigating before you finalise next year's budget.

 

The BFCM metric most people miss: new versus returning customers

A BFCM with strong revenue numbers and poor new customer acquisition is a problem that last-click will not show you. The channel breakdown might look perfectly healthy while your business is essentially discounting its way through its existing customer base.

Segment your BFCM conversion data by new versus returning customers. If the majority of your BFCM revenue came from people who already knew and bought from you, two things are worth investigating. First, your acquisition channels may not be generating real new demand; they may be capturing returning customers who would have bought with or without the ad click. Second, your BFCM economics look better than they are, because new customer acquisition at BFCM cost of sale is very different from repeat purchase revenue.

New customer rate by channel is a useful cut to take. A Paid Social campaign with a 60% new customer rate is doing something different from a branded search campaign with a 90% returning customer rate. Both might have similar last-click ROAS. Only one is growing the customer base.

 

What to do with this before planning next year

Do not plan next year's BFCM budget using last-click ROAS by channel as the primary input. The channels that look most efficient in last-click are often the channels that capture demand, not the ones that created it. Scaling them further generates diminishing returns, because there is a ceiling on how much existing intent you can harvest.

The question to ask of every channel is: is this channel introducing customers to the brand, or is it closing customers the brand already had? DDA, path reports, time lag data, and new customer rate will tell you more about that than last-click ROAS ever will.

For channels that consistently appear in the first or second position in path reports, drive a high new customer rate, or show up in post-purchase survey responses despite low platform attribution, the case for protecting or growing budget is stronger than last-click would suggest. For channels that dominate last-click conversion data but appear mostly late in paths and drive high returning customer rates, the picture is more complicated. These channels may be essential for capturing intent at the moment of decision, or they may simply be taking credit for demand that existed anyway. Testing incrementality through controlled holdout experiments is the only way to know for certain.

The conclusions you draw from this period determine how you allocate budget, which channels get investment in the months before next November, and whether the channels that actually built your BFCM demand get the spend they have earned.

Ready to grow?

If you're serious about growth, we're serious about helping. No obligation, no pressure, just a conversation about what's possible.

DROP US A MESSAGE →