We often come across retail clients with years of brand awareness work, with more than 70% of their turnover coming from Brand campaigns traffic and with the same challenge: How to increase the revenue share of Non Brand campaigns on the Search Network in a profitable way.
The reality is that the difference in profitability in Brand and Non Brand traffic (traffic generated from generic, category or product searches that do not include the term of our brand), in general but especially in this type of companies, is very large. You may ask yourselves: "if we could support all the traffic acquisition in Brand campaigns, being the cheapest and most profitable, why not do it? Some of the reasons why it is important to diversify the revenue share between Brand and Non-Brand campaigns are:
1. Reduce the dependence of upper funnel investment on the acquisition and generation of sales in the lower funnel. If the vast majority of your revenue share is based on Brand campaigns, the success of your sales campaign will be highly dependent on what has been done in the upper funnel, be it media, online or offline. A possible loss of market share will be directly proportional to a drop in impressions of your performance campaign.
2. Provide more inputs to the new user strategy vs. the retention strategy. It is very important to have two differentiated strategies at the business level:
new user acquisition, one of the fastest ways to grow revenue and achieve short-term objectives, and retention, which will allow us, among other things, to increase the average ticket and the all-important Lifetime value.
retention, which will allow us, among other things, to increase the average ticket and the all-important Lifetime value.
These strategies should also be reflected in the acquisition strategy. If we focus the acquisition of new users only on users who know our brand (whether or not they have purchased), we will be excluding ourselves from the market with all that this entails.
3. Opportunity to monitor your potential audience. According to Google, 20% of daily searches are new. That means that it is highly likely that the user will change the way they search for your products in a slightly different way over time, especially those users who are in a knowledge phase. This is where Non Brand campaigns are a good opportunity to connect with our potential audience and for example discover new search terms that we can use.
The truth is that with every acquisition strategy we devise (whether in Paid Search or Paid Social), what we seek is to find and enhance those niche audiences that we know have a higher probability of conversion - not only we refer to clusters of users by interests, but the combination of these with devices, times of day / week, demographic and geographic data, behaviors - ultimately crossing all sources of information and data that we have at our disposal. Once these niches have been found, we will be able to:
Balance the available budget towards them.
Help balance the CPC we pay according to their respective conversion rates, establishing differentiated optimization strategies.
The level of aggressiveness in the use and application of audience strategies will vary depending on the strategic need of each client. It is at this point where we want to share an aggressive audience strategy in response to the challenge posed above:
Increase the revenue share of Non Brand campaigns in Search Network in a profitable way.
With this audience strategy we will develop the Non Brand acquisition strategy targeting only users that are part of audience lists created by sales funnel level.
This strategy as such is designed for companies with a very high dependency on Brand campaigns and with very high levels of profitability requirements in Non Brand. In the same way, it can be flexible and adapted according to the situation and needs of each company.
We share the main keys:
First key: Audiences
We will start with the definition and organization of the audiences that will be part of the strategy. The objective is to be able to find well-differentiated groups of audiences according to levels or positions that the user occupies in a sales funnel. The variables that we can use to create audiences are:
Number of transactions
Recency (30d, 60d, 240d, etc).
Source/medium.
We can sort these audience groups by levels of relevance or qualification, with the lowest values being the closest to sales and the highest being the closest to awareness (here, for example, we can use similar audiences from audience groups of lower levels). In this example we will work with 10 audience levels with one group per level. Within each group we can create one or several audiences per recency, for example:
Second key: Segmentation
This is where we control the reach and potential of the campaigns through different segmentation factors:
Audiences: We will set up audiences in segmentation mode. We will only impact users who are searching for Non Brand keywords and who are also within our audience clusters.
Product: do we want to work the entire catalog or do we want to select categories/subcategories/genre with better margins, stock depth or seasonality?
Geographic locations: do we want to segment an entire national territory or do we want to target geographic areas with better conversion rates? We could talk about segmentation by regions, populations or even coordinates by location of own physical stores and/or competitors.
Keywords: The degree of granularity from more generic to more long tail has to be the result of analyzing the volume of potential users and the degree of segmentation that we have applied in the previous points. These can be worked in two ways:
Dynamically (DSA campaigns/adgroups).
Combined: main keywords on the one hand (as a rule we would not work with keywords of more than 2-3 terms), and on the other hand DSA adgroups to collect the longest tail searches.
Our recommendation is to start with a high level of segmentation - always according to the volume of total users that we will discuss in the following points - since we can always make this segmentation more flexible later on once we have started to drive sales in a profitable way and we need volume.
Third key: scalable campaign structure
We are going to create a powerful Non Brand campaign and keyword structure that contains these audience groups in targeting mode.
To achieve a scalable structure we will have to create this structure of campaigns and Non Brand ad groups in a way that allows us to activate and deactivate audience levels depending on the volume of traffic we generate. Ideally, we should start with the audience levels closest to conversion. The objective will be to activate the structure of higher levels as the traffic in lower levels decreases (remember that we are crossing audiences with keywords, so the reduction of potential audience is important).
In the following example, we set up a campaign by level (following the previous example): within each campaign we add the audiences we have created for each level in segmentation mode, and we set up ad groups for the main term groups (either with DSA or with keywords).
We will start with all the campaigns paused and we will activate the lower level campaigns - the opening of campaigns will be progressive according to volume needs.
Fourth key: message adaptation
Another strong point of this strategy is the opportunity to improve the degree of engagement with users. This improvement translates into higher CTRs and therefore an improvement in Quality Score and a reduction in the real CPC that we end up paying. We will achieve this through:
The personalization of messages in ads depending on the level of depth of the sales funnel in which each user is, as an example:
"Get inspired" for a user who is at an early stage of consideration.
"Complete now" for a user who has reached the product card but has not initiated the checkout process.
"Come back and finish" for a user who has reached the cart but has not completed the order.
Add promotions and personalized product cross-selling: Like the message, we can use this strategy to offer promotions and related products through extensions to clusters of users.
Fifth key: volume of users
The volume of available users will determine the level of depth and number of audiences we can work with. In the higher levels of brand awareness we can use similar audiences from the lower levels. The same happens with the long tail level of the keywords: the higher the volume of available users, the more long tail keywords we can work with, while with lower volumes of users it will not make sense to go for the long tail, as we will reduce the potential audience too much.
Sixth key: measurement
As always, the way we analyze campaign results can lead us to make incorrect or hasty decisions. We always recommend working with data-driven attribution models, as in this case it is accentuated if the comparison is against Brand campaigns. This means that we have to avoid the last interaction attribution model in Google/Bing Ads platform.
It is quite common behavior, especially in models with medium/high time to conversion, that the first interaction of the user is through Non Brand campaigns and ends up finalizing the purchase by entering through a Brand ad.
If we do not work with our own attribution models, in Google Analytics, when working with the last interaction model, we will have to take into account the assistance in conversions when analyzing and drawing conclusions about the performance of Non Brand campaigns.
In short: this is just one more way to take advantage of one of the most powerful resources available to digital companies today - proprietary data - by applying a layer of personalization that allows us, on the one hand, to improve the user experience with our brand by connecting with them in a more relevant way, and on the other, to be more efficient in the way we manage our advertising investment.
