How to test cheaper

How to test an offer with a minimum budget?

Tests are the basis of the basics in arbitration, somewhere like drinking water, that is, they are vital. Everyone knows how to do them, because any launch is already a test. But not everyone can do it optimally and effectively - and we will talk about how to do it later.

What is “optimal” in this case?

Spend as little budget as possible - and get the desired test results - whether it is successful or not. You can find out that the landing page does not work for 2000 USD - or maybe for 500.

So how to calculate the minimum?

Let's say our goal is CPA. Then the budget that should be set is at least 1.5 of the CPA price.

This does not mean that they must be spent to the last cent - if you see clicks for $10, low CTR or CR - perhaps you should stop advertising earlier. Everything here is optional.

However, if everything is going plus or minus normally - the minimum budget should be 1.5 CPA - ideally - 2.

How does it look from the media buyer's side?

Suppose we have an offer in the crypto vertical - which will convert by 5% to Tier 1 geo - but the lead price is low, and we pour plus or minus into zero. This is a low envelope for Tier 1 - and we want to improve the situation.

To do this, we can:

a) make new creatives

b) test a new lead

c) try a new offer

Option A - new creatives

We made new creatives - we will find out the lead price for them in 2-3 days. We pour 1.5-2 CPA - if the price is adequate.

It is important to give this spend in a few days - because the results in 1 day may be biased - the lead may close later - and we are interested in objectivity.

We received 0 deps - we are short

1 dep - if there is a budget and status norms - we can still shed, if not - we are short

2+ deps - we try to carefully scale, or we look at the same connection at a longer distance.

Option B - a new landing page

I'll say right away - if you choose between a creative and a landing page - then the landing page has a stronger impact on the envelope. The goal is to find a landing page that “opens” the creative page and leads the user to conversion.

You take 2-3 creative pages - you launch them on a new landing page - and spend the same 1.5-2 CPA for several days.

Option B - a new bundle.

When do we make a new bundle? In short, always =)

Because it is worth testing not when the bundle stops working, but constantly, so that everything works 24/7, ideally. Here the same logic applies, only we test the combination of 2/3 creative + landing page right away.

This is if the CR is normal.

If the CR has dropped significantly, and there were many cases of this - given that the approach to advertising has not changed - then perhaps it is worth stopping the test immediately.

In the results, we again focus on the number of conversions.

You can even put the same landing page in a split with different headlines - and see where the CR is better, and switch to the best one in the live broadcast.

The main thing is to have a clear plan, and not just pour traffic and see what happens, and so on endlessly. You test chaotically - you expect a miracle. And if you have a strategy, you will spend money effectively, save both time and nerves.