Google Ads recommends us to use the last option (UAC) for everything related to installations and registrations in applications. Already in case it is suspicious that we lose control of the keywords or the platforms where our ad will appear. That is why I question your recommendation …
Alternative campaign to UACWhat I propose as an alternative to UAC is to use search campaigns as if it were not a mobile application. In this way we make sure to get out in the search options for the keywords we decide.A priori it is a risky option since the user is going to do a search thinking that he is going to find a type of results that are far from the “tedious” download of an app.
In the case of the example, we were focused on professional categories, so that a user who searched for “plumber Madrid” would be difficult to end up downloading an app …
What do we need?To carry out search campaigns aimed at downloading an app, we don’t need many things, I will focus on the landing page, the campaign settings and the different cell phone number database for sale objectives and bids.
Landing page mobileThe first, most basic thing is to have a mobile landing page . If you don’t know how to easily design an optimized landing page, I’ll tellyou about it in this article.😉We need a VERY optimized recruitment page gulf email list for mobile phones that makes it clear that to get the servicewe are looking for we have to download an app. If we are going to “force” the user to download an app to get this service, we have to convey the advantages of our app (free, fast, simple, …). In the example below, links were included in all the images to facilitate the user’s arrival at the store.
The configuration of the campaign will have to be ad hoc to each project, but there are a number of obvious sections that we have to take into account.
DEVICES: The campaign should only be oriented to mobile (and tablet if applicable).
KEYWORDS: The selected keywords cannot be redundant and we will have to refine a lot in those that we believe can generate a download and subsequent registration (if this is our objective). This means achieving less results, but gaining in quality.
LOCATION: It depends a lot on the app and the project, but in many cases we may be interested in focusing this effort on large cities and forgetting the less populated areas. I repeat: this is something that depends a lot on the app, but in my test the installations in sparsely populated areas were much more expensiveMarketing objective and bid strategy
The marketing objective and the bid strategy are included in the campaign settings, but it seems like a delicate point to me and so I want to treat it separately.First of all, we must decide the main objective of the campaign Sales: Our campaign will be focused on the users most likely to make a purchase in our app. If the objective of our campaign is to generate a download that later reports a sale to us, we may be looking at the best selection.
Sales opportunities: If we are looking for leads that bring us a profit in the form of sale later, this is our goal. We could say that it is one level less than the previous objective, since here it would be “potential sales”.
Traffic to the website: In the case of wanting to optimize our campaign to gain more traffic to a website where we will later interact with the user, it would be our option.
Remove objective recommendations: What has been not setting a goal to the campaign.
Based on my experience, and taking into account that it is based on campaigns targeting only one type of application, I have not found differences between the objectives. Maybe the “Traffic to the website” has worked a little better for me because of using a landing page, but nothing significant.If you have a budget, it is best to do tests to find the best optimization objective, but if not, I advise not to use any and take advantage of bid strategies to better define the strategy.
Second, we must choose the bidding strategy that will give us the best results:
Objective CPA: If we are clear about our CPA (or CPL in the case that a record is a lead) we can try this option. This strategy ensures that we have an objective cost and it will not be exceeded. We run the risk of not getting any results
ROAS objective: I honestly find it difficult to see the difference compared to the first one. At a theoretical level it is clear, optimizing to make the advertising investment profitable, but on a practical level I find it difficult to see that there are cases in which it is better than optimizing to Target CPA.
Maximize clicks: It is a strategy oriented to CPC, that is, more clicks at the lowest cost. In my experience, for this type of campaign it has brought me more results but of poorer quality.
Maximize conversions: Optimize to get the highest number of conversions. Here it depends on what we have configured as a conversion in the app. It is important the previous configuration and that the app or Firebase send events that we consider the most important. For Firebase, Google Ads etc, a conversion equals our most important KPI.Search page placement targeting : A strategy geared towards ad position. I don’t see it as anything appropriate for an app experiment. In general I don’t like this option, which does not mean that there are people who will work for you.
Target Higher Ranking Percentage: Same as above is not among my favorites. If we have a very similar competitor and we are in an oligopolistic market, perhaps it could be useful. I have not tried.
Improved CPC: Adwords automatically optimizes our CPC to achieve the best results.
Manual CPC: The user is the one who decides the CPC manually. If we have time and a great knowledge of the market it can be a good option. By the way, it is the best way to learn😉
For this type of campaigns to applications and taking into account that my campaign objective was the installation and subsequent registration, the ones that have worked best for me have been: Target CPA, Maximize Conversions and improved / manual CPA.
Results of my campaign
After testing the UACs for 2 or 3 months, I decided to improve the CPL data with this alternative. For the specific case, a lead is a record within the application.