How do I decide how much to invest in online ads?

"How much should I invest in ads?" is one of the hardest questions to answer well, because it is almost always answered backwards. An amount that seems reasonable is chosen and then people hope it works. The safer path is the opposite: start from what a new customer is worth to you.

Step 1: what a new customer is worth

Work out, on average, how much margin a new customer leaves you, counting what they buy once and what they buy again. That number is the ceiling of the calculation: everything you spend to win them has to stay below it, otherwise you are working to lose.

Step 2: what you can pay for a new customer

From the value in the previous step, decide what part you are willing to spend to win one. There is no single rule; it depends on margin, sector and appetite for growth. What matters is giving it a limit, so you do not decide on emotion.

Step 3: how many enquiries become customers

Not every enquiry becomes a customer. See, out of every ten people who ask for information or book, how many end up buying. This is called the conversion rate, and it is a number only you know, because it depends on how you answer and what you offer.

Step 4: the maximum cost per enquiry

Now do the sum: if a new customer leaves you X in margin and you close, say, one in every five enquiries, then you should not pay for each enquiry more than a fifth of what you are willing to spend on a customer. That is the limit of what each enquiry can cost you.

Step 5: start with a small test

With the limit set, the starting budget is what is needed to get enough enquiries to learn from, over a few weeks. It does not need to be large. It serves to see which messages and audiences respond, and how much each enquiry really costs.

• If the real cost comes in below the limit, increase gradually.
• If it comes in above, do not increase: change the message, the audience or the landing page first.
• If there are no enquiries, before spending more, check that the page the ads lead to says clearly what you do.

What changes the cost of each enquiry

• The landing page. A good ad that leads to a confusing page wastes money.
• The clarity of the offer. Whoever does not quickly understand what you do does not get in touch.
• The reply. Whoever takes days to reply loses enquiries that were already paid for.
• Competition in the auction. There are sectors and seasons where every click costs more, and that does not depend on you.

An honest warning

Ads speed up what already works; they do not fix what does not. If the website does not explain what you do, or if replies take long, the money is spent and the result falls short. That is why, when we work on Digital Marketing, we first look at what whoever clicks receives, and only then at how much to spend. To see the B2B channels that tend to work, we have the guide to lead generation in Portugal.

The sum, in words

If a new customer leaves you X in margin, you can decide to spend a part of X to win them. If one in every five enquiries closes, each enquiry can cost a fifth of that part. Whenever the real cost goes past the limit, the problem is in the message, the page or the reply, not in the size of the budget.

In short

• Start from what a new customer is worth, not from an amount that seems reasonable.
• Maximum cost per enquiry: what you can pay for a customer, times the share of enquiries that close.
• Start with a small test and only increase when there is a signal.

Quick questions

What if I do not know what a customer is worth?

Make a conservative estimate, with the average margin of a sale and an idea of how often the customer comes back. Even imperfect, it is better than having no calculation.

Is Google or Meta better?

It depends on where the customer decides. Whoever actively searches for what you do tends to be on Google; whoever did not know they needed it can be reached on social media. Start with one and measure.

How long should I leave the test running?

Long enough to get enquiries that allow conclusions, usually a few weeks. Changing everything every two days stops you learning.