Black Friday data checklist: what to prepare before October

Preparing your data for Black Friday is September work, not November work. Once the campaign goes live, a source that came in crooked stays crooked forever: no later fix gives back the information the lead did not bring. That is the difference between arriving in December knowing which offer worked and arriving with a big number that nobody can explain.

This checklist is not about inventory, shipping, or pricing. It is about the part that tends to be remembered last and that decides whether the result will be analyzable: tracking, naming, integration, and dashboard. There are seven items, all doable before October.

One note before starting: almost nothing here is exclusive to the date. These are the same checks that sustain any media operation all year round. What changes in November is the margin for error. With ten times more volume coming in during a single week, a missing parameter stops being a nuisance and becomes the difference between knowing and not knowing where the quarter's revenue came from.

1. Standardize naming before launching the first campaign

On Black Friday, the volume of campaigns explodes, and with it the number of people working in the account. It is the perfect scenario for the same channel to become three different lines in the report.

Lock in the convention now: name structure, separator, allowed values for objective and offer. Include a tag that identifies the promotion, something like bf2026 in a fixed position in the name, because that is what will let you separate the result of the date from the result of the whole month. If the standard does not yet exist in your operation, we wrote about the naming mistakes that break your report and what to do about each one.

The payoff shows up in December, when someone asks for performance by offer. With a standard, it is a filter. Without one, it is an afternoon of spreadsheets.

2. Test the form and the integration end to end

Do not trust the form just because it worked in August. Make a real submission, with test data, and follow the record all the way through: did it reach the marketing tool, did it reach the CRM, and did it arrive with the source filled in?

The two points that break most often are the expired credential and the field renamed by the website team without notice. Both fail silently, and the gap only shows up at closing.

Test each entry point separately: website form, campaign landing page, chatbot, and WhatsApp. One of them is usually out of standard.

3. Check lead deduplication

Black Friday behavior multiplies records. The same person clicks the ad, enters through WhatsApp, comes back through email, and fills in the form again on the promotion landing page.

Without a merge rule, each pass becomes a new lead. The result is an inflated funnel and an acquisition cost that looks lower than it really was, because the investment was divided by people who do not exist.

Before the peak, check how your database treats the same contact with different emails and with a phone number registered with and without the ninth digit. If the merge is not automatic, define the primary key now and agree on who runs the cleanup during the campaign.

Resource · In PortuguesePrompt Guide for Marketing and Sales70 prompts with questions about marketing, sales and behavior to ask your data.Download the guide

4. Tag what is not paid media too

The classic mistake of the season is to polish the ad's UTM and forget the rest. On Black Friday, a good part of the traffic comes from email, WhatsApp, the bio link, and banners on your own site.

Without a parameter, all of it falls into direct traffic, and the channel that converted the most becomes invisible. Build the links for these pieces together with the media ones, in the same standard, and have them all ready before October.

5. Define the daily tracking metric

During the campaign there is no time to invent analysis. Decide now, with the team, which number will be looked at every day and which decision it unlocks.

In most operations there are three: volume of leads or orders per channel, cost per result per channel, and a quality metric, such as acceptance rate or average ticket. The third is the one that usually goes missing, and it is what keeps the team from celebrating volume that does not turn into sales.

Also write down the threshold that triggers an action. Something like “if a channel's cost per result goes above X for two days in a row, we review the offer.” Agreed beforehand, it becomes a quick decision. Agreed during the campaign, it becomes a debate.

6. Have the dashboard built before the volume arrives

Build the dashboard in September, with September data. It seems early, and that is exactly the point: you find out now that a field is missing, that a source is not connected, or that the chart does not break down by offer.

The dashboard needs to answer, without manual work: how much came in per channel, how much it cost, and what turned into sales. If it depends on someone building a spreadsheet every morning, it will not be updated on November 28.

Also leave a comparison view ready, with the same period of last year alongside. Without that comparison, every number in the week looks good, because volume goes up either way. It is the reading against the historical baseline that shows whether the operation was really better, and it is the main reason to organize your Black Friday data in advance.

7. Freeze structural changes once the campaign goes live

This is the item almost nobody writes in the checklist and that saves the analysis. From the moment the campaign is live, do not rename campaigns, do not swap parameters, and do not change the account structure.

Besides breaking the history in your report, structural change has a cost inside the media platform itself. The Google Ads documentation explains that significant changes send the campaign back to the learning period, and that the bidding strategy can take up to 50 conversions or three conversion cycles to adjust. In Black Friday week, you do not have that time.

If something is wrong, fix it by creating the new one the right way and letting the old one finish. Renaming in mid-flight is the worst of both worlds.

The schedule until November

September: naming convention locked, end-to-end tests done, deduplication verified, and dashboard built. It is the month of boring work.

October: links for all pieces ready and reviewed, daily metric defined with the team, warm-up campaigns live with the standard already applied. It is the month in which you validate the standard at low volume, when mistakes are cheap.

November: operation. Nothing in the structure changes, and the team looks at the dashboard every day with the agreed action already written down.

The schedule above assumes your operation already has working tracking and only needs to reinforce it. If the starting point is a messy database, with a stalled integration and lost sources, add a month: data cleanup for Black Friday in an operation in that state starts in August, and it is better to do the honest diagnosis now than to discover it in October.

It is worth running a full check of the database in October before the peak. The five data hygiene routines cover exactly that and fit in one morning.

What changes when the database is already organized

A good part of this checklist exists because collection is manual in many operations, and Black Friday only amplifies the problem that already existed in August. When UTMs are organized automatically, the same customer becomes a single record, and media, form, and CRM data arrive in the same database, preparation shrinks to two things: defining the daily metric and building the dashboard.

If your operation is going to face November with spreadsheets and manual exports, it is worth at least locking in the standard beforehand. The Data Management Playbook brings campaign naming, the UTM manual, and the custom fields guide in a single document, ready to adapt.

Frequently asked questions

What is the real deadline for preparing data for Black Friday?

September for the structure and October for validation at low volume. What determines the deadline is not the date of the promotion, it is the time you need to discover a tracking error and fix it before it gets expensive.

Can tracking be fixed during the campaign?

You can fix it going forward, but not backward. The lead that came in without a source came in without a source, and that information does not come back. That is why validation needs to happen beforehand, with low traffic.

Do I need to create new campaigns or can I use the ones already running?

Both work, and the decision is about media, not data. What matters for the analysis is that the Black Friday campaign is identifiable in the name and in the parameters, so you can separate the result of the promotion from the result of the month.

Which metric should I watch during the peak days?

One for volume, one for cost, and one for quality, always broken down by channel. Volume alone misleads, because the peak brings many curious people, and the team only finds out in December that the cheapest channel was also the one that sold the least.