More budget, more sales. Did the change work?

Imagine the owner of an online lighting shop. On Monday, she raised the ad budget. By Friday, orders were up. She is already tempted to message her ads specialist: “Let’s add a bit more.”

But on Wednesday, a delivery of desk lamps arrived. They had been out of stock for almost two weeks. On Thursday, the shop emailed customers to say the popular model was available again.

“Did the orders pick up after we raised the budget?” the owner asks.

“We need to check. When did the lamps go back on the website?”

She remembers the delivery date. She is less sure when the stock status changed on the product page. And she thinks the email went out in batches.

This is a fictional example. The question is a practical one: how do you assess a change to your ads when the shop changed at the same time?

First, piece together what happened

The owner opens the change history in GetProfit, then the messages from the warehouse and the email campaign history. She puts three events in one note: the budget change, the lamps returning to sale and the customer email.

Now “we raised the budget and sales went up” sounds different. Both things happened. How much of the increase came from the budget is still unclear.

Our GetProfit platform has a section called “What changed” for this: a history of changes to the ad account, including the budget. You can see when it changed and the values before and after. You can also open a change history in Google Ads. Keep your reason for making the change, delivery dates and email dates alongside it in your own notes.

You do not need to recreate the ad account in a spreadsheet. You need a short record of the decision: what you wanted to test and what else might have affected the result.

Check your stock before the next step

After the episode with the lamps, the owner checks whether any other bestsellers have sold out and still have not returned to sale.

In GetProfit’s “Products” section, the “Top missing” block lists products that used to bring sales but are now absent from ads. It gives a reason, including “out of stock”. Some products with that reason may be your bestsellers: start your restocking list with those.

Start by checking them against your stock records, then ask the supplier about lead times and terms. For lamps that customers still need and that remain worthwhile for the shop, the next action will be to place an order with the supplier. A higher ad budget alone will not bring an unavailable product back into stock.

The owner now has two connected decisions: which lamps to reorder and what budget to test once customers can order them. She will note the date they return alongside the ad changes, so the next review shows both parts of the story.

Write down the question before you see the answer

Next time, the owner starts with a sentence: “I want to check whether we can get more desk lamp orders at an acceptable cost of selling them.”

That is more useful than “let’s try raising the budget”. There is a product group, an expected change and a condition under which it would not be worth continuing.

She then agrees with the specialist which metrics they will review. Ad campaign data will go on one sheet and actual shop orders on another. The shop’s sales include other channels; you cannot attribute every extra order to one change in Google Ads.

For the decision itself, a record like this is enough:

Write the limit as a specific amount or a metric based on your own costs and margins. “If it gets expensive” will mean something different each time. The budget setting is no substitute for actual spend, either: check how much was really spent.

Do not rush the verdict, but keep watching

A few days later, the specialist sends the first report. Spend has already risen, but some purchases may still come through later.

Google specifically warns about conversion delay: someone may buy some time after interacting with an ad. A recent period can therefore look worse than an older one: spend is already visible, but not all conversions have reached the report.

That does not mean you should always wait exactly a week. Check how long purchases usually take to appear in your account, and consider how much data you have. You can set the first review date in advance; you do not have to reach a final conclusion by that date.

Waiting does not mean you stop watching the shop. If checkout stops working, the lamps sell out or the agreed spending limit is reached, investigate straight away. Add another entry to the log: what happened, when you noticed and what you did.

Finish with a decision, not a score of “good”

The owner returns to that first week. She now has the event dates, a fresh ad data export and the shop’s orders. She can see growth, but these records do not let her separate the budget’s effect from the delivery and the email.

So that is what she writes: “Orders are up. We also brought a popular model back into stock and sent an email. The effect of the budget alone has not been established.”

The next step is clearer. First, make sure the lamps are available. Then choose a quieter stretch for the next change, record the conditions and monitor them. If you need to estimate the causal effect itself, a before-and-after comparison is not enough: that needs a separately designed experiment.

Even a careful log does not switch off seasonality or competitors’ actions. But a month later, you can check your own decision instead of debating what you remember.

Open the change history in GetProfit and find the latest budget change. Next to it, write why it was made and what was happening with the product, its price and delivery. Then look at which bestsellers are out of stock and what it would take to bring them back. You will have both a record of the past decision and a concrete action before the next investment in ads.