Black Friday Measurement Guide

₀₁

Setting the right goal before the campaign starts    

Before Black Friday starts, you need to be able to answer a fairly simple question. “What does a successful Black Friday actually look like for the business?”

“Have a big Black Friday” isn't an answer. Neither is “grow revenue” on its own.

You might want to acquire thousands of new customers. You might want to generate as much cash as possible before Christmas. You might want to clear a particular product range, reactivate existing customers or use the period to introduce more people to the brand.

All of those can be valid goals. But they lead to different decisions about your offer, marketing budget, products and how you judge performance afterwards.

The important thing is deciding which outcome matters before the numbers start coming in.

Turn the goal into something measurable

Marketing goals have a habit of becoming ratios.

ROAS. CPA. MER. Conversion rate. AOV.

They're useful measures, but none of them is the actual objective of the business.

Start with the outcome you want Black Friday to create and put a number against it.

For example, “We want Black Friday to generate £400,000 in revenue while acquiring at least 1,500 new customers.”

Or, “We want to sell £150,000 of this range before Christmas without spending more than £30,000 acquiring those sales.”

Now you have something you can work backwards from.

You can determine how many orders you need, how much traffic that is likely to require, what you can afford to spend and which metrics actually matter while the campaign is running.

The metric is there to help you reach the goal. It isn't the goal itself.

Goals that conflict with each other

The problem comes when you try to optimise Black Friday for everything at once.

You want more revenue, but you also want a higher ROAS.

You want thousands of new customers, but you don't want acquisition costs to increase.

You want to clear stock, but you want to protect average order value.

You want to maximise sales during the biggest shopping period of the year, but you don't want to waste media spend.

Those goals can pull the campaign in different directions.

If your priority is acquiring new customers, you may be prepared to accept a higher acquisition cost because you're reaching people who haven't bought from you before.

If your priority is clearing a particular product range, overall ROAS might matter less than whether that stock actually moves.

If your priority is generating the maximum possible revenue, protecting an exceptionally efficient ROAS can become the thing that stops you spending into profitable additional demand.

This is why the primary goal needs to be decided in advance.

Otherwise, it's very easy to move the goalposts afterwards and judge the campaign using whichever metric happens to look best.

You can't decide what success means after you've seen the result.

Agree what success looks like with the business

In a privately owned business, this conversation shouldn't need a complicated planning process.

If you own the business, decide what you actually need Black Friday to deliver commercially and make sure the marketing plan reflects it.

If you run the marketing, sit down with the person who does.

Don't start asking “What ROAS do you want us to hit?” Start with “What do we need Black Friday to do for the business?”

Then turn that answer into numbers.

How much do we want to sell? How important are new customers? Is there stock we need to move? How much are we prepared to invest to achieve it? What result would make us say in January that Black Friday was worth doing?

Once you've agreed to that, the marketing metrics have a job to do. They help you understand whether you're moving towards that outcome and where you might need to change course.

Without it, the marketing team can deliver a campaign that looks successful in the ad platforms while the person looking at the bank account has a completely different view.

Before you decide how you're going to measure Black Friday, agree what you're actually trying to achieve.

₀₂

Why ROAS falls apart during peak     

“The platforms seem fine, but the money isn't showing in the bank.”

It's a problem we hear about during Black Friday. You open Google or Meta and the numbers look healthy. Revenue is up. ROAS is within target. The campaigns appear to be doing their job.

But when you look at the business as a whole, the picture doesn't feel quite as good.

That doesn't mean ROAS has suddenly become useless. It's still valuable for understanding what's happening within your advertising. The problem comes when you treat it as the scorecard for whether Black Friday is working for the business.

During peak, three things make that particularly dangerous.

1. Discounting changes the economics underneath the revenue. 
2. Multiple channels compete for credit during a compressed buying window. 
3. Some of the revenue you're looking at won't survive the returns period.

ROAS is still useful diagnostically. It just isn't the final score.

What discounting does to the ratio

ROAS is a simple calculation. 

Revenue attributed to advertising ÷ advertising spend

The problem is that the ratio doesn't tell you what revenue is worth to the business.

Imagine you normally spend £20 to generate a £100 order. That's a 5x ROAS. During Black Friday, you discount that product to £80 and acquisition costs fall to £16. You're still generating a 5x ROAS.

The platform sees exactly the same efficiency. But before the promotion, that £100 order generated £60 of gross profit before advertising and £40 after the £20 acquisition cost. During Black Friday, gross profit falls to £40 and you're left with £24 after the £16 acquisition cost.

Same 5x ROAS, but the £40 contribution becomes £24.

Nothing looks worse in the ROAS number, but the commercial outcome is 40% lower.

That doesn't make the discount wrong. You might deliberately accept a lower margin to acquire new customers, clear stock or generate additional volume. It does mean that a ROAS you'd be delighted with during the rest of the year can represent a very different commercial outcome during Black Friday.

The ratio hasn't necessarily broken. The economics underneath it have changed.

Attribution during a compressed window

Black Friday also compresses a lot of customer activity into a relatively short period. A customer might see a Meta ad earlier in the week, search for the brand on Google later, receive an email on Friday morning and then buy that afternoon.

That's true throughout the year. During peak, there's simply much more happening at the same time.

More advertising. More emails. More branded searches. More direct traffic. More retargeting. More customers who were already waiting for the sale to start.

Imagine the business generates £80,000 during Black Friday. Meta attributes £50,000 to its campaigns. Google attributes £60,000 to its campaigns. That doesn't mean those two channels generated £110,000 between them.

Some customers will appear in both numbers because each platform is applying its own attribution rules to the same buying journeys. Both platforms can therefore report healthy ROAS at the same time without the business seeing an equivalent increase in total revenue.

The danger is adding together the stories each platform tells and assuming you've found the truth.

You haven't. You're looking at different measurements of the same customer journey. This is where measurement needs to become diagnostic, not a scorecard.

Use the platform numbers to understand what's happening. Where are costs increasing? Which campaigns are generating demand? Where is conversion changing? Where might budget be constrained?

But don't ask each platform to tell you whether Black Friday is working for the business.

The question isn't “Which platform gets the credit?” It's “What is all of this marketing doing to the business?”

Returns, and the number you saw in December being wrong

Then there is the part you can't properly measure when Black Friday finishes. Returns.

If £50,000 of Black Friday revenue is eventually returned, the ROAS you were celebrating in December was calculated using revenue the business ultimately didn't keep. 

So the platforms can look perfectly healthy in December while the commercial result deteriorates in January. 

This matters when comparing years too. If last year's Black Friday performance is based on final retained revenue but this year's is based on gross orders before returns, you're not comparing the same thing.

It also matters when judging acquisition. A campaign might appear to have acquired a customer profitably until their first order is refunded.

Black Friday doesn't finish when the advertising stops. Neither should the measurement.

Step above the platforms

None of this means you should stop looking at ROAS. It's useful for diagnosing what's happening within your advertising and helping you make decisions inside individual channels. 

The mistake is turning it into the final score for Black Friday. Ultimately the question isn't whether Meta reported a good ROAS or Google hit its target.

It's whether the total amount you invested in marketing generated enough revenue for the business.

To answer that, you need to step above individual channels and look at the relationship between your total marketing investment and total revenue. That's where MER becomes useful.

₀₃

Marketing efficiency ratio

If ROAS helps you understand what is happening inside your advertising channels, Marketing Efficiency Ratio helps you step back and understand what is happening to the business as a whole.

That's particularly useful during Black Friday.

You've already seen why individual platform numbers can become difficult to interpret during peak. Google and Meta can both report healthy performance. Discounts can change the value of the revenue being generated. Multiple channels can claim a role in the same sale.

MER removes some of that noise by asking a much simpler question:

How much revenue did the business generate for every pound invested in marketing?

It doesn't replace the numbers inside Google, Meta or your other channels. It gives you another level of measurement above them.

What MER is and how to calculate it

The basic calculation is straightforward:

Total revenue ÷ total marketing spend = MER

If your business generates £500,000 in revenue from £100,000 of marketing spend:

£500,000 ÷ £100,000 = 5x MER

For every £1 invested in marketing, the business generated £5 in revenue.

The important difference from platform ROAS is what goes into the calculation. You're not asking Meta how much revenue Meta generated or Google how much revenue Google generated.

You're looking at what the business generated in total against what you invested in marketing in total.

For the purposes of measuring Black Friday, keep it simple: use total media spend across your marketing channels.

You can build a broader measure that includes agency fees, salaries and other marketing costs, but if you do, you need to use the same definition every time you compare performance.

The important thing isn't finding the perfect definition. It's making sure the number means the same thing every time you use it.

ROAS looks inside the channels. MER looks at the relationship between marketing and the business.

Why MER holds up better than ROAS during peak

Imagine the example from the previous section. Meta reports £50,000 of revenue. Google reports £60,000. But the business actually generated £80,000.

Looking at the channels individually creates a problem. There's £110,000 of attributed revenue attached to £80,000 of actual business revenue. MER doesn't need to decide which platform deserves the sale.

If you spent £20,000 across marketing and the business generated £80,000:

£80,000 ÷ £20,000 = 4x MER. That's the business-level reality.

This becomes particularly useful during Black Friday because you're rarely running one channel in isolation. Paid social might create awareness. Search captures people actively looking for you. Email brings existing customers back. Organic, direct and affiliate traffic can all play a role.

Trying to assign every pound of revenue neatly between those channels can create an illusion of precision. MER steps above that argument.

It doesn't ask which channel gets the credit. It asks what happened to the business when you invested the money.

There is an important limitation, though. MER doesn't solve the economics of discounting by itself.

If you generate £500,000 from £100,000 of marketing spend, you have a 5x MER whether that £500,000 was sold at full price or at a heavily reduced margin.

So during Black Friday, MER still needs to sit alongside the commercial measures that matter to the goal you set at the start: margin, contribution, new customers, stock cleared or whatever you've decided success looks like.

MER gives you a cleaner view of marketing efficiency across the business. It doesn't make revenue the same thing as profit.

Setting a MER target for peak

Don't start with the MER you want and build the Black Friday plan around it. Start with what the business needs Black Friday to achieve and work backwards to the MER that makes that commercially viable.

Peak changes the economics. You might deliberately accept lower efficiency because there is substantially more demand available. You might be prepared to spend £100,000 to generate £400,000 rather than £50,000 to generate £250,000.

MER falls from 5x to 4x. Looked at purely as an efficiency score, that appears worse.

But the second scenario generates an additional £150,000 of revenue. Whether that's a good trade depends on the margin you're making, the customers you're acquiring and what you wanted Black Friday to achieve in the first place.

That's why the target needs to come from the business rather than from what last month's dashboard happened to say.

Start with the commercial goal you've already set. How much revenue are you trying to generate? How much are you prepared to invest to achieve it? What margin will you retain after the offer? How much efficiency are you prepared to trade for additional volume?

Then work backwards. If the goal is £600,000 of revenue and you've agreed that the business is prepared to invest £150,000 in marketing to achieve it:

£600,000 ÷ £150,000 = 4x MER

You now have a business-level reference point for the period. Crucially, it isn't an arbitrary marketing target. It's the consequence of the commercial plan.

Run it alongside channel metrics, not instead of them

MER tells you whether marketing efficiency across the business is moving in the right direction.

It doesn't tell you why. If MER starts deteriorating on Friday afternoon, you still need the channel-level data to diagnose what's happening.

Has Meta CPM increased? Has Google CPC risen? Has conversion rate fallen? Is one campaign absorbing more budget? Has a product sold out? Is mobile conversion deteriorating?

This is why replacing ROAS with MER would simply create a different measurement problem.

You need both levels. Use MER to understand the relationship between total marketing investment and total business revenue.

Use channel metrics to diagnose what is driving it and decide what to change.

There will even be times when the two appear to disagree.

ROAS in an individual channel might deteriorate as you increase spend, while MER remains relatively stable and total revenue grows significantly. Stopping because the channel no longer hits its historical ROAS target could mean giving up profitable growth for the sake of protecting a metric.

Equally, platform ROAS could look excellent while MER deteriorates. That should prompt a different question. If the channels are apparently becoming more efficient, why isn't the business seeing the benefit?

That's where measurement becomes useful. Not as a collection of scores you're trying to keep green, but as a way of understanding what's happening and deciding what to do next.

Business first. Channel second.

₀₄

What to watch during the week      

Once Black Friday is live, the temptation is to watch everything.

You keep refreshing your dashboards to check for new sales. The numbers don’t follow how you forensically forecasted and panic sets in. A good afternoon suddenly becomes the new benchmark.

This isn’t going to help you make better decisions.

During peak, separate the numbers that can tell you something useful now from the ones that need time before you can trust them.

The daily numbers

There are a handful of numbers worth keeping close during the week.

Start this at a business level:

• Revenue
• Orders
• Marketing spend
• MER
• Conversion rate
• AOV

Together, they tell you whether the business is broadly tracking where you expected it to be and, importantly, whether there might be an opportunity to do more. 

Revenue and orders tell you whether you're delivering the volume you planned for. Spend and MER tell you what it's costing you to generate it. Conversion rate helps you understand whether you're making enough from the traffic arriving on the site, while AOV tells you how much each order is worth. 

None of those numbers is particularly useful in isolation. Revenue being ahead of plan is less exciting if you've had to massively overspend to achieve it. A falling MER might be perfectly acceptable if you're deliberately pushing harder into profitable demand.

Then use your channel metrics to understand what's driving that performance.

Your Black Friday Ads Guide peak week checklist covers the channel-level checks in more detail. The important thing here is to keep bringing those signals back to what is happening to the business overall.

The ones that lag

Some of the most important Black Friday numbers won't be available during Black Friday.

You need to wait for returns to happen, for new customers to have the chance to buy again and attribution to settle after any data lags. 

Those numbers matter, but they shouldn't necessarily determine what you do at 2pm on Black Friday. A customer acquired that morning can't tell you what they're worth over the next six months, and revenue generated that day hasn't had time to become retained revenue. 

This means your live view of Black Friday is incomplete, so don't force a conclusion from a metric that hasn't had enough time to mature.

Some numbers will help you run Black Friday. Others tell you whether Black Friday was worth running. 

When to intervene and when to leave it

A metric moving isn't automatically a reason to change something. You see more volatility than usual during peak. Costs change, conversion rates vary throughout the day and customer behaviour can look very different from normal trading periods. 

The opposite mistake is being too cautious when the numbers are telling you there's more opportunity available. The purpose of monitoring performance isn't just to spot problems. It's to understand when to protect the plan and when to push beyond it.

If revenue is ahead of plan, MER remains commercially viable, conversion is holding and there is still demand available, you may have room to increase spend.

If revenue is behind plan, don't immediately cut budget. You need to work out why.

If traffic is there but conversion has fallen, look at the website, offer, stock and customer journey.

If conversion is holding but traffic is below plan, look at whether you can capture more demand through additional media spend.

If spend is increasing but revenue isn't moving with it, look at the channel-level data to understand where the additional investment has stopped producing enough incremental value.

And if revenue, MER and conversion are all broadly where you expected them to be, sometimes the right decision is to leave things alone.

This is where the targets you set before Black Friday become useful. They give you boundaries within which you can make decisions rather than reacting to whether an individual metric happens to be green or red.

The question isn't simply “Is performance good or bad?” It's “What are the numbers telling us to do next?”

Sometimes that means fixing a problem. Sometimes it means leaving things alone. Sometimes it means recognising that performance is holding, there is more demand available and you have an opportunity to push harder.

The point of measuring Black Friday while it's happening isn't to score the campaign. It's to make better decisions while there's still time to change the result.

₀₅

The review that decides next year      

Black Friday can be the biggest customer acquisition event of the year. Thousands of people might buy from you for the first time, often attracted by the biggest discount you'll offer all year.

The immediate numbers tell you what it cost to acquire them. The more interesting question is what those customers turned out to be worth.

Did they become valuable customers, or did Black Friday simply generate a large number of discounted first orders?

Those answers shouldn't just sit in a Black Friday report. They should change what you do next year.

The numbers you can only see in January

The revenue number you celebrated during Black Friday isn't necessarily the revenue you eventually keep.

By January, you can start replacing some of the assumptions you made during Black Friday with actual outcomes. 

Returns and cancellations have worked their way through. You have a clearer view of the revenue you actually kept and enough time has passed to start understanding whether the customers you acquired behaved differently from those you normally attract.

Go back to the targets you set before the campaign and reconcile them against what actually happened.

How much revenue survived returns? What happened to margin once discounts, acquisition costs and returns were accounted for? How many genuinely new customers did you acquire? Did you clear the stock you intended to clear?

The live numbers tell you how Black Friday is going. The January numbers tell you what Black Friday actually did.

New customers acquired, and what they did next

This is where Black Friday measurement gets much more commercially interesting. Promotions can make customer acquisition look extremely successful. You might acquire more new customers than at any other point in the year. But acquiring a customer and acquiring a valuable customer aren't necessarily the same thing.

Someone who buys for the first time because you've offered 25% off may behave very differently from somebody acquired at full price in March.

You need to build a Black Friday customer cohort and keep watching it.

How many made another purchase? How quickly did they come back? What did they spend on their next order? What proportion of their subsequent purchases were made at full price? How does their return rate compare with customers acquired at other points in the year? 

Then compare their value over time with the cost of acquiring them.

Imagine your normal customer costs £25 to acquire and is worth £120 over the following 12 months.

During Black Friday, you acquire customers for £18 and initially celebrate the lower acquisition cost. But if those customers only ever place the discounted first order, the £7 saving on acquisition isn't particularly meaningful. 

The opposite can also be true. You might accept a higher acquisition cost during Black Friday because those customers subsequently become highly valuable repeat buyers.

The cheapest customer to acquire isn't necessarily the most valuable customer to acquire.

This is why the lifetime value of a discount-acquired customer matters.

You don't need to wait several years to learn something useful. Compare Black Friday cohorts after 30, 60, 90 and eventually 365 days. Over successive years, you'll build a much clearer picture of what a Black Friday customer is actually worth to your business.

That changes how much you should be prepared to spend acquiring the next one. 

I've seen this happen in practice. One client acquired thousands of new customers on their peak trading day. On the surface, it looked like a huge acquisition success. When we looked at those customers later, very few had bought again. The business had used a significant promotional discount to acquire them in the first place, but that initial purchase wasn't translating into valuable repeat customers.

That changed how the business viewed peak entirely. Rather than paying to acquire the same volume of discount-led customers the following year, they decided to switch off non-brand acquisition activity during the period and rely more heavily on the strength of the brand to capture demand more efficiently.

The acquisition number hadn't changed. Their understanding of what those customers were worth had.

What to change before September

The purpose of the review isn't to produce a Black Friday report that gets filed away until next November. It's to make next year's decisions better.

By the time planning starts again in September, you shouldn't be starting from assumptions. You should have evidence from the previous year.

You should know which products genuinely created value, which offers produced sales without destroying too much margin, where you ran out of stock, where the customer journey failed and where additional marketing investment continued to generate worthwhile returns.

Most importantly, you should have a much better idea of what a Black Friday customer is worth. That can change the plan significantly. 

If discount-acquired customers rarely return, you may decide that aggressive new-customer acquisition isn't as valuable as it appeared during the event.

If they become strong repeat customers, you may be willing to accept a higher acquisition cost next year and push harder while the opportunity is available.

If particular products acquire customers who subsequently become unusually valuable, those products may deserve a very different role in next year's campaign.

If you discovered you could have spent considerably more while remaining commercially viable, that's something you want to know in September, not halfway through the next Black Friday weekend.

Every Black Friday should leave you with more than revenue. It should leave you with better information for the next one. So when planning starts again in September, don't start with last year's media plan.

Start with what last year's customers, orders and economics taught you.

₀₆

Measure what matters to the business      

Black Friday gives you no shortage of numbers. The harder part is knowing which ones actually tell you whether the business made more money. Google and Meta can both look successful while competing for credit for the same revenue. ROAS can hold while discounting changes the economics underneath it. A cheap new customer can turn out to be worth very little once you see what they do next.

The answer isn't another platform report. It's connecting what happened in your marketing to what happened in the business. Because when you're optimising towards an incomplete view of performance, it's very easy for budget to end up in the wrong place. 

We typically find 15-30% in wasted marketing spend in the first 3 months. Our Growth Intelligence Audit is designed to find it. 

We connect your marketing performance with your commercial, customer and behavioural data to understand where budget is being wasted, where the numbers you're optimising towards are constraining growth and where the biggest opportunities actually sit.

That's exactly what we found with Stone Refurb. 

Their ROAS target looked like a sensible way to protect marketing efficiency. In reality, it had become the constraint. Strict targets meant decisions were being made to protect the platform metric rather than grow the business, limiting how much profitable demand they could capture. 

We rebuilt the strategy around commercial efficiency instead, changed where budget was being invested and used broader attribution modelling to give the business the confidence to scale beyond the old ROAS ceiling. Overall revenue increased by 91%, with no change in MER.

Read the Stone Refurb case study

The number you're optimising towards should help the business grow. It shouldn't become the thing that limits how far it can. 

If you want to understand what your marketing is really contributing to the business and whether the numbers you're using to judge performance are helping or constraining growth, start with a Growth Intelligence Audit.