Discuss the project
Leave your contacts in the form below and we will contact you shortly
Our contacts in Ukraine:
Kyiv, Popudrenko, 1A
+38 093 221 67 68
Our contacts in Canada:
Canada, Ontario, Toronto,
228 Queens Quay West St.
+1 647 393 5015
Kyiv, Popudrenko, 1A
+38 093 221 67 68
Our contacts in Canada:
Canada, Ontario, Toronto,
228 Queens Quay West St.
+1 647 393 5015
Increase in sales through the website by 254% in 6 months.
FARSH
Senior PPC Manager
Max Tarabashchuk
Services provided by adburo: advertising on Facebook
Region: Lviv
Terms of collaboration: September 2020 – February 2022
Budget for Facebook: $1200 a month
Website: farsh.in.ua
Region: Lviv
Terms of collaboration: September 2020 – February 2022
Budget for Facebook: $1200 a month
Website: farsh.in.ua
UA
SEND REQUEST
SEND REQUEST
About company
Fast food restaurant in Lviv. The restaurant prepares large burgers according to original recipes using sauces of its own production. It has its own delivery, loyalty program, regular promotions and good offers for companies with several employees.
Issues
Before starting we identified some key issues. Here are some of them:
- Lack of tracking of results. The client already had a Facebook ad set up. There were cute creatives of good quality, namely pictures. But the number of orders through the site was not tracked. Facebook launched a type of advertising campaign aimed at bringing the user to the site, but it was not clear whether the user made a purchase. Pixel, a tool from Facebook that needs to be connected and configured to track results, was not used. The company that previously set up and ran the ad could only show the number of visitors, but not the number of actual orders through the site.
- Lack of adaptation of creatives for stories format. Stories on Instagram provide big coverage. For the most part, ads that appear in the news feed along with stories give better results. Not all creatives on the client’s website were adapted to the vertical format of stories, so they did not look completely professional. This reduced the potential of advertising.
- Uneven sales in two restaurants. The client needed to increase the number of orders through the site with delivery in the central part of the city. Most of the orders for delivery were being made to a restaurant outside of the city center. The staff was loaded unevenly. In one restaurant, the staff was overworked, however, it lacked orders in another one.
Goals
- To maximize the number of orders with delivery.
- Customer acquisition cost (CAC) had to be $3,5-4.
- To increase the number of orders in the city center by 30-50%.
Tasks
- To set up tracking results using Facebook Pixel to understand the cost of one order from different advertising campaigns.
- To test the effectiveness of advertising by providing display in different areas of the city.
- To develop creatives in square and vertical formats, for news feeds on Facebook/Instagram. To test video format.
What was accomplished
1. Tracking of results was set up. In particular: Google Tag Manager, Google Analytics, Google Optimize, Facebook Pixel:
Google Tag Manager and Google Analytics make it possible to monitor various indicators of user behavior on the site: from which devices they enter, on which days and hours more orders are made, and much more. Google Tag Manager also makes it possible to create audiences from users who, for example, were on the site for more than 30 seconds, clicked on the required button or watched a video, but did not place an order. That means that the audience showed interest, but did not perform the intended action. Such people need to be shown ads again. In most cases, it is more profitable than showing ads to people who are not yet familiar with the brand.
Pixel provides an opportunity to transfer information about user’s behavior on the site directly to the Facebook advertising cabinet, as well as an ability to calculate the average cost of placing an order. Thanks to configurations on Facebook Pixel one can see in the screenshots below how many orders were received on the site for a certain period, and what was the average cost of each order.
Google Optimize allows to conduct A/B tests. A single changed phrase on the first screen of the site can improve or worsen the effectiveness of advertising by 20-40%. We run A/B tests to find out what users respond to best.
All settings are made according to our own checklist, which contains more than 50 items. Most of the settings are the same at the start of work. For example, Google Tag Manager and Google Analytics need to be set up on each project, no matter what ads we set up. In order not to miss any important configuration points, we have compiled a checklist for our team. By carrying out priority settings on it, we reduce the probability of making mistakes.
2. Searching for cheaper clients (testing different audiences). The company had already cooperated with a marketer. Traffic optimizing advertising campaigns had been set up. However, only indirect results were tracked: site visits, likes, comments. After setting up tracking of orders, we decided to test the type of advertising campaigns that were launched before us. Since, according to the client, after the launch of advertising (previous to us), the number of orders slightly increased. But we soon discovered that even if this type did provide orders, they were 3-5 times more expensive than the cost that was profitable for the client.
So we started testing a different type of advertising campaigns. Campaigns with conversion optimization. Facebook with conversion optimization finds those users who will not only click on the ad, but also perform the necessary action on the site. In our case, placing an order. This campaign immediately showed better results than the previous ones, so we started testing different audiences and targeting options to find the combination that would give the most orders at the required price of $3.5-$4.
We tested different audiences with different targeting options. For example: targeting interests, audiences similar to target customers (Lookalikes), site visitors, etc., or targeting a wider audience. For further work and optimization, we selected 3 audiences with a certain set of targeting and started working on making orders cheaper and further increasing their number.
So we started testing a different type of advertising campaigns. Campaigns with conversion optimization. Facebook with conversion optimization finds those users who will not only click on the ad, but also perform the necessary action on the site. In our case, placing an order. This campaign immediately showed better results than the previous ones, so we started testing different audiences and targeting options to find the combination that would give the most orders at the required price of $3.5-$4.
We tested different audiences with different targeting options. For example: targeting interests, audiences similar to target customers (Lookalikes), site visitors, etc., or targeting a wider audience. For further work and optimization, we selected 3 audiences with a certain set of targeting and started working on making orders cheaper and further increasing their number.
An A/B test was conducted. Site conversion was raised by 17%. One can develop the most remarkable ads and test out a number of targeting methods. But if we guide visitors through an unclear or simply unattractive site, advertising may end up worthless. Therefore, work on the site is no less important than work on advertising itself.
Our client’s website was done beautifully. Juicy pictures of burgers, a conveniently located basket, an array of promotions that drew attention of a potential buyer. Therefore, we decided to test which discount would catch more attention and convert a visitor into a buyer more quickly. For the test, we chose and agreed two phrases with the client:
1. Every 4th burger is a giveaway
2. 20% off the first order
1. Every 4th burger is a giveaway
2. 20% off the first order
These phrases appeared on the first screen of the directory page that we were forwarding people to from the ad. The test was conducted using the Google Optimize service, which we connect by default at the first stage of the set-up.
When visitors landed on this variant of the page with the phrase “20% off your first order”, they were more likely to place an order. In other words, one phrase made it possible to increase the percentage of purchases by 17.23%. This allowed us to further reduce the cost of one order.
(The test was conducted on this page. Some time after we did the work, the discount changed)
(Average order cost - $4.35)
Working with creatives
Creatives (picture or video and text) are one of the most important elements of advertising. In order for the Facebook algorithm to be able to find someone to show advertising to and for it to be effective, the algorithm needs to have something to choose from.
The result may vary by a huge margin. For example, creative 1 gave 206 orders at $2.88, while creative 2 – 5 orders at $8.62. The audience and settings are identical. The difference only lies in the content of the first picture of the carousel. And the difference in the cost of each order is almost threefold.
The result may vary by a huge margin. For example, creative 1 gave 206 orders at $2.88, while creative 2 – 5 orders at $8.62. The audience and settings are identical. The difference only lies in the content of the first picture of the carousel. And the difference in the cost of each order is almost threefold.
Creative 1
Creative 2
At the start of our work with a new project, we always develop at least 4-6 creatives to show to a cold audience, and as many to advertise for remarketing. We come up with several versions of texts for each audience.
This allows us to run different tests and find combinations of text and visuals that produce better results for less cost.
Examples of creatives and texts:
This allows us to run different tests and find combinations of text and visuals that produce better results for less cost.
Examples of creatives and texts:
- Orders through website: 334
- Average cost of the order: $2.7
- Coverage: 136 398 users (around 18% of the population of the city)
- Displays: 797 905
(The analytics shows data on orders from all advertising sources including Google Ads, which we set up as well)
Increasing the number of orders with delivery in the city center
We conducted several tests with geo-targeting. For example, we launched targeting on the radius around both restaurants. In one of the tests, the radius around the priority restaurant was larger.
After each launch, we received feedback from the client about changes in the number of orders and, as a result, with the load on the kitchen. After several tests, we discovered that even with an unequal load on the restaurants, it is more cost-effective to run ads targeting the entire city.
After each launch, we received feedback from the client about changes in the number of orders and, as a result, with the load on the kitchen. After several tests, we discovered that even with an unequal load on the restaurants, it is more cost-effective to run ads targeting the entire city.
Results
1. Cost of one order and secondary indicators. The results for April 2021 were selected as an example.
2. Dynamics of results. For example, there were 14% more orders in April than in March. The graph below shows the growth dynamics in the number of orders through the site over several months.
Summary and recommendations for the sector of food delivery
- The effectiveness of advertising should be evaluated approximately one week after its launch. Users complete the order after a few “touches” of the ad. Therefore, the results do not arrive immediately but roughly a week later, when orders start to appear on the website and profitability levels off.
- It is worth setting up search advertising on Google Ads. The most productive days in terms of the number of orders are Friday, Saturday and Sunday. In the first half of the week, the advertisement “catches the eye”, but the users do not complete the order. At the end of the week, due to not remembering the brand, customers would go on Google and search for “food delivery” or “burger delivery”. Google advertising increases the likelihood that such users will complete the order on our client’s website as the visitor recognizes the site they have already visited.
- A significant number of orders are emotional purchases. Therefore, it is necessary that the pictures of the food are done as professionally as possible. They should literally be mouth-watering. Therefore, it is worth cooperating with a professional food photographer.
First Step
We can start discussing your project. Or we can conduct an audit of your advertising campaigns
Our contact in Ukraine:
Our contact in Canada:
adburo © 2024
Schedule a call
We will discuss your needs in 30 minutes and prepare an offer
Social:
