How to increase orders even during a pandemic
It’s time to summarize the results of the spring and summer season for one of our clients, ArvenAuto car service in Kharkiv. It is worth noting that we have been working with this client for quite some time, starting back in 2018. During this period, we started working with both contextual advertising and SEO promotion. SEO played a significant role in today’s case study.
Straight to the numbers
Compared to the previous year:
- Conversions increased by 16.95%;
- Increase in the value of conversions (their actual quality) by 72.97%. Even the client himself noted that there were more orders and their quality had improved;
- The conversion rate has improved by 25.38% (more users than ever before converted);
- Some behavioral metrics increased: the bounce rate decreased by 3.94% and the number of pages per session increased by 6.94%. This indicates that the quality of traffic has improved;
- The number of users (and new users) grew slightly – 2.80% and 2.62%, but herein lies the main catch – all these figures took place during the pandemic and strict quarantine, so even keeping them at the same level was a significant result.
From the client, we know about the seasonality period, which is the period between March and August, when car service is most often used. We have always emphasized this time of year and were shocked when quarantine was announced right before the season.
Decisive action
After a short pause, the client decided that he wanted to continue working, as he would still need the services of a car service. We agreed, but decided to change our tactics. The entire advertising account was set up to drive traffic (this was the main goal). In the conditions of self-isolation (when everyone spent a lot of time on the Internet), there was already enough traffic, so we had to work on its quality. First of all, we turned off all media campaigns, except for remarketing.
Then it was decided to go through all the semantics, use more accurate keys with more relevant landing pages. For example, there used to be several campaigns that had a division into services:
- Car service;
- Repair of the internal combustion engine;
- Diagnostics of internal combustion engines;
- Walking;
- etc.
Each group had both broad keys “+car repair” and narrow keys “+KIA +RIO +Kharkiv”. This made the keys difficult to manage. Some had many results, while others were idle for months. Also, due to this system, there were often overlaps between groups of ads, which blurred the statistics and generally caused inconvenience.
In addition, the client gave preference to certain areas, for example, he was much more interested in one repair of an internal combustion engine than several repairs of a steering rack. The same advantages have emerged for car brands. The decision was made to distinguish several separate campaigns.
The result was this picture:
- General services (without specifying makes and models of cars);
- Brands and models of cars (indicating the service, for example: “+repair +engine +KIA”)
These campaigns were carefully disguised. This way, we were able to control all the directions we needed in just a couple of clicks, and we also removed the intersection.
The next step was to abandon campaigns with keys without modifiers. Their essence was that they were disjointed from regular campaigns and did not overlap with exact queries, but they drove more or less useful traffic. In any case, traffic was no longer a priority, so we decided to go the other way: arvenauto.com has good SEO optimization, and there is a separate page for each type of repair, which is in the top 10 for its keys. We decided to take advantage of this advantage and created campaigns with dynamically generated headlines, and as it turned out in the end, this decision was the most successful.
As a final step, we added as many audiences as possible to all search campaigns for monitoring and started collecting statistics on them.
Customer feedback and optimization
Within a few days, we saw that the quality of traffic had indeed improved (but the quantity had decreased):
The client was also generally satisfied and suggested raising the advertising budget slightly to bring back the previous volumes. Next, we started to optimize the account, and the following work was done to this end:
- Working with bids for keywords;
- Adding semantics from queries;
- Cleaning up queries;
- Adjustment of demographics and location.
The next step in the optimization was to reduce remarketing costs, but we decided not to reduce daily budgets and the maximum CPC, but went the other way: we reassembled audiences to more targeted ones. We returned only those who had performed some micro-action, for example, started filling out a contact form but quit for some reason. This helped reduce costs and improve remarketing conversion.
As we continued to work with our audiences, our search campaigns eventually collected enough data to create combined lists. Here’s an example:
Some of them were so successful that we switched several campaigns to targeting only these combined lists.
Then it was just a matter of tracking campaign performance, making adjustments where necessary, and almost completely abandoning the purchase of PC traffic. We also tested how the CPA strategy would work on this account, but we were not satisfied with the result. It was decided to manage the client manually. After that, no major changes were made.
Conclusions
To be honest, we weren’t sure that we would be able to reach the level of the previous year, let alone improve our result. However, the experience we have gained helped us to get out of this difficult situation. The client was also satisfied.
We continue to work with Arven Auto and look forward to effective cooperation in the future. This was a case study on how to optimize your Google Ads account and get benefits even where it would seem impossible to get them.
We hope you found this case study useful. Thank you for your attention!
