Pokupon.ua is a Ukrainian service designed to help its users find significant discounts on various goods and services. They are an online platform where partner sites find potential clients, and users pick discounts from 50 to 90% for a wide range of products.
On Pokupon.ua, about 4,000 promos and discounts are placed every day so that the website visitors have an opportunity to purchase goods and services at a reduced price. For sellers (partner companies), this is an additional channel of brand advertising that can potentially generate more sales.
As it appeared during the advertising campaigns, the largest volume of traffic was generated by the brand requests, e.g. people were searching for companies.
At the beginning, the team created ad groups, selected keywords, and wrote the ads for partner advertising manually, which caused the following issues:
The aforementioned nuances led to plenty of additional working hours for the assigned specialists, as it was necessary to manually process any change in ads and URLs on a daily basis, that already had ad groups created for more than 300 partners in Google Adwords. Manual updates required about two working hours every day.
We considered the project details and developed a two-stage working plan that included consult, audit, and technical optimization.
At the first stage of the project, the team developed a script to stop manual checking of the campaign relevance.
This script checked the launched promotional campaigns for their relevance on a daily basis. If one of the ads were overdue, the script automatically stopped the ad group which contained it to reduce the budget expenditure.
After the introduction of this script, there was no need to manually check every launched partner’s campaign which saved time.
Promodo used its own API* to automate a number of processes while working with Google Adwords. For instance, the API was used to process the generated feed that contained data on all the active ad campaigns. So, the team automated start, finish, and management of advertising campaigns for each of the Pokupon.ua partners.
*Currently, APIs are working for five major cities in Ukraine. Later, we are planning to expand the campaigns to other cities.
To set up a Google AdWords account configuration correctly, we created a feed with all the partners. This was in the YML format (Yandex Market Language based on XML format) and was updated once a day. This is a sample feed:
The relevant keywords, ads, and ad groups were automatically generated based on the feed with active partners. This process was repeated automatically on a daily basis considering feed updates. As a result, inactive ad groups were stopped and new ones were created using relevant keywords and ads.
This is a sample ad:
The issues we faced
While working on this project, the team faced a number of issues from which it was impossible to complete the full automation.
Keywords were generated based on the names of the partner companies, so the use of these requests could not result in targeted traffic and as a result, manual work was required to identify such partners and correct the keywords.
Another script was written to obtain more flexible settings and configurations for the project along with the same functionality.
The main advantage of this solution was the possibility of combining data from a few feeds and processing it according to certain criteria, which helped to avoid the problem of using brand keywords that were generated earlier for some partners.
With daily updates in the manual mode, it took about 12–20 minutes per day and helped to create more relevant ads, which allowed to increase their CTR.
Sample ads:
The achieved results as compared to other non-brand campaigns in Google AdWords are the following:
With the help of automation, the time for launching the advertising campaigns was reduced, and it also became possible to achieve an increase in transactions and, accordingly, the income.
Case studies
Check out how we help companies like yours overcome challenges and rise to the next level.
Conversion rate
Bounce rate
Work period
.avif)
.webp)
.svg){kind=logo}
