The first week was not the verdict
We opened the account with no warm retargeting lists, no past creative winners and no sense of which audiences would actually buy. The campaign opened with its weakest result. More than a third of the total budget was used in the first seven days, while the return finished below one euro back for every euro spent. If we had judged the whole event from that week alone, the obvious conclusion would have been that demand was not there.
The next four weeks told a different story. Together they returned 3.46 times what went into the ads and accounted for 122 of the 144 purchase events tracked during the run. The opening result still belongs in the total, but it did not have to define everything that followed.
How the five weeks unfolded
| Period | Purchases | Back per euro |
|---|---|---|
| Opening week | 22 | 0.81× |
| Week 2 | 15 | 3.64× |
| Week 3 | 32 | 3.93× |
| Week 4 | 39 | 2.66× |
| Final 7 days | 36 | 3.95× |
Better did not mean perfectly smooth
After the difficult opening, the return rose to 3.64 times and then 3.93 times. The fourth week dipped to 2.66 times before the final seven days climbed back to 3.95 times. That movement matters because successful campaigns rarely improve in a neat straight line.
People buy when their interest, plans and the event date line up. Our job is to read those changes without overreacting to one day or one week, then keep the budget closest to the audiences giving us the clearest evidence.
Two audiences, two very different jobs
Warm audiences closed the sale. New audiences kept filling the pool.
This does not mean new audiences were unnecessary. Some people in the warm group may have first met the event through those broader ads. Discovery and conversion support each other, even when the final purchase is credited to the later touchpoint.
The strongest segment is not the whole campaign
One smaller audience group returned more than six times what went into it. That is useful evidence, but it is not the event result. Smaller groups can look exceptional until more budget reaches them and the available audience begins to run out.
We prefer the honest blended number: 2.48 times across every row and every week. The standout segment tells us where to test next. The blended result tells us how the campaign actually finished.
What this means for your next event
1. Treat a weak opening as information, not a final answer. Some events reveal demand immediately; others need time to gather it. A disappointing first week should lead to closer reading and careful changes, not an automatic decision to stop or spend more. Hugel's next four weeks were much stronger, while the opening week remained part of the honest total.
2. Give warm and new audiences different expectations. People who already know the event are naturally more likely to buy. New audiences often need more than one encounter before they are ready. Judge each group by the role it plays, while still setting a clear limit on how much discovery you can afford.
3. Let current evidence guide the budget. The final week was strong here, but the week before it was softer. That is why a countdown alone should not decide how aggressively to spend. Stay present for late buyers, then increase support when the latest results justify it.
4. Build the next test from the full result. A high-return pocket can point toward a useful opportunity, but it should be expanded carefully. Start with the campaign-wide return, then use the stronger audience and location signals to shape the next test without assuming they can absorb unlimited budget.
Figures are rounded and measured in the Romanian advertising account from March 28 to May 1, 2025. Purchases are checkout events, so one purchase can include more than one ticket. Returns are advertising-platform figures rather than the ticketing ledger.