How to calculate the effectiveness of advertising in the niche of employment abroad

The client was engaged in the employment of Ukrainian citizens in Hungary.
The task of the agency’s team was to generate leads for this service. The cost of an employment transaction (an already closed transaction, not a lead) should not exceed $200.

The monthly budget for targeted advertising was not less than $2000, which allowed them to constantly test different campaigns and offers.
But the challenge was to determine which ad settings were effective and cost-efficient and worthy of further scaling.

The main problem was that the duration of the employment decision often reached a month, two months or more.
For quite objective reasons. Someone is still planning to work abroad and is looking for conditions for the future. Someone is just starting to prepare documents and it takes time. Many people are still choosing a country and an employer, so after the first contact, they may return to continue the dialog in weeks or months.

This particular area of activity was relatively new for the client, so they did not have their own analytical data on the conversion of leads into deals.
The client was also just planning to implement CRM. So there were some problems with calculating the effectiveness of advertising.

For example, at the end of April, 6 people were hired. This is unsatisfactory because the cost of “leaving” is too high, and advertising is unprofitable.
The following month, in May, 20+ people left, which is a good indicator.
However, client managers note that many of those who signed a deal in May first applied in April, and some even in March.

At the same time, there was no structured data on the quality of leads, reasons for refusals, or decision-making timeframes.
We, on the other hand, saw the number of leads we received in the advertising office. The client only provided information on the number of visits during the reporting period.
Without knowing the date of the lead, it is impossible to calculate the real effectiveness of advertising.
Nor is it possible to determine which advertising campaigns are producing effective leads. After all, there were 4-5 campaigns running at the same time.

Therefore, the following scheme was proposed for implementation.
Each lead is entered into a Google spreadsheet. The main identifier is the phone number. This is what users left in the lead collection form.
Be sure to put the date of receipt of the lead. The date of the first contact (call) during business hours on the same day or the next business day.
Usually, after the first contact, the manager can determine the relevance of the lead. A corresponding mark is made in the table.
The results of all subsequent communications with each lead are also recorded, indicating the results.

Careful maintenance of such a table has provided enough statistics and other useful information for the client within two months.

The most important thing is that now we have the opportunity to focus on the actual results in our advertising analysis.

First of all, when testing new campaigns, we, as executors, can actually see the quality of the leads we receive in real time.
So it reduces the time for testing and simplifies decision-making on the effectiveness and continuation of tests.

Secondly, we started to accumulate more accurate statistics on the conversion of leads into deals, which made it possible to take a more realistic approach to cost per lead (CPL) estimation.

And finally, we were able to attribute the cost of each lead to the reporting period in which it was received.
This, in our opinion, is the correct scheme for calculating efficiency.

So, let’s summarize.
Of course, the presence of a customized CRM should have removed all the issues listed in this case study. For objective reasons, this project did not have a CRM at the time of cooperation, but it was necessary to work and show results.
I am sure that today, in 2024, there are many, many such examples.

In order not to turn advertising analysis into fortune-telling, you need to develop and implement a custom quality assessment system for the specifics of the project.
To be honest, in the project described above, the client did not accept the scheme of work we proposed at first.
It took three presentations before we convinced them that it was necessary and would provide correct results and calculations.
Subsequently, the client confirmed all this himself when he saw the real results of cooperation.