All Collections
Custom Audience Filters
Custom Audience Filters

Everything you need to know about building Custom Audience Filters using Tinyclues

Diana Sains avatar
Written by Diana Sains
Updated over a week ago

What is Tinyclues' Custom Audience Filter Feature?

While Tinyclues' predictive engine by design identifies only those customers most likely to engage with your offer, there may be instances in which the nature of your communication requires certain audience restrictions. Perhaps, for example, you need to restrict a given offer to females, or to those who engaged with a specific campaign in the past.

Tinyclues' Custom Audience Filter feature allows you to create a wide range of tailor-made audience filters in just a few clicks, based on a variety of customer attributes, purchase history, marketing reactions, and predictive metrics.

Let's take a closer look!

How to Create Custom Audience Filters Using Tinyclues

Start by navigating to the Custom Audience Filter section of the UI by clicking the small Document-Magnifier icon to the left of your screen.

Next, click the blue New Audience Filter button.

By default, the toggle Display this audience on Campaign Plan is activated. And if you use several channels, the Channels field allows you to assign one or more channels to the filter.

If you do not need to use the audience in your campaigns, you can switch it off. In this case, the “Channels” field disappears.

Now it's time to consider the conditions that need to be fulfilled in order for customers to be eligible to receive the communication you're looking to push - that is, what types of customers are you aiming to include in your audience? Alternatively, what types of customers do you need to exclude from your audience? Perhaps you need to target males exclusively, or those of a particular age range; or customers within a particular geography, and so forth.

Begin by selecting an initial framework for your filter by clicking the green Include in my audience button or the red Exclude from my audience button.

Next, you'll be presented with a dropdown menu that will allow you to explore different types of criteria for your filter across several categories.

Note that these options will vary based on available data (stemming from your customer tables, purchase tables, email data, etc).

You can also reuse an existing filter as a criterion.

Please note that only one existing filter can be added to a query and only filters that do not reuse an existing filter themselves are allowed. Filters that already rely on another existing filter appear grayed out in the drop-down list and cannot be selected.

Once you select the field, you'll be able to choose between the following conditions:

  • Equals: When the selected field is exactly the same as the input value.

  • Not equals: When the selected field is different from the input value.

  • Starts with: When the selected field starts with the characters of the input value.

  • Ends with: When the selected field ends with the characters of the input value.

  • Like: When the selected field contains the word in the input value.

  • Not like: When the selected field does not contain the word in the input value.

  • Is empty: When there is no existing value of the selected field.

  • Is not empty: When the value is not empty.

  • Any in: When the selected field is equal to any of the input values (This allows you to select multiple values).

  • Not in: When the selected field is different from all the input values. (As with Any In, this allows you to select multiple values).

You will also be able to select between different timeframe types:

  • Moving Window: Timeframes selected from the Moving Window category will update each time you refresh your audience. For example, if you select In the last 180 days, with every refresh of your audience, the timeframe will be, Refresh Date minus 180 Days.

  • Fixed Dates: This option will require you to select exact dates from the calendar. Thus, no matter the refresh date, the selected timeframe will always be the same.

Ultimately, you'll be able to link separate conditions as needed, tailoring audiences to your desired level of specificity.

For example:

  • Females Living in Paris (Customers Who Are Female AND Customers Who Live Paris).

  • Purchasers of Brands A or B within a certain timeframe (Customers Who Purchased Brand A OR Customers Who Purchased Brand B).

Let's explore a few examples.

Deterministic Filter: Females Living in Paris

  • In order to fulfill the conditions of female and living in Paris, you'll need to create two distinct segments and link them via AND operator.

  • As we're restricting our audience based on customer attributes, we'll find the gender and geography categories within the Customer dropdown. In this example, the necessary gender and geography criteria are located under the Gender and Advanced Data > Zipcode sections of the Customer category, respectively (see Image A below).

Please note

  • If your customer data includes customers in multiple geographies, you would need to apply a Country restriction to this example as well (in this case, France).

  • The City category might be available to you based on your customer tables, in which case you could select a City rather than restrict by zipcode.

  • Use the tooltip next to the input field for each category to view a selection of possible inputs (see Image B below).

  • Once you've selected your filter criteria, click the blue Compute audience button to view the audience size, and the blue Create button to finalize the creation of your filter (see Image A below).

Females in Paris Image A

Females in Paris Image B

Deterministic Filter: Purchasers of Brand A or Brand B (Past 365 Days)

Now let's take a look at a sample filter involving an OR operator.

In this case, you want to select customers who have purchased Nike brand products OR Adidas brand products within the past year.

  • As before, you'll begin by selecting Include in my audience.

  • As you're looking to restrict by Brand, you'll select the Purchase category, followed by Brand; you'll then enter Brand A as your input, along with the timeframe within which the purchase should have taken place (in this case, 365 days).

  • Next, hit the OR button, and do the same thing for Brand B.

  • Once you've selected your filter criteria, click the blue Compute audience button to view the audience size, and the blue Create button to finalize the creation of your filter.

Deterministic Filter: Email Campaign Clickers

It's useful to keep track of your customers’ campaign activity. You might need to create an audience of active customers (e.g., Clients who have clicked more than 6 email campaigns in the past 180 days, as seen below).

Alternatively, you may create an audience targeting clients with a low click-through rate (e.g., Clients who have clicked fewer than 2 email campaigns in the past 180 days).

  • In order to create these filters, you'll need to navigate to Marketing Reactions > Email Clicks > Any campaign.

  • Select the most relevant operator according to your needs, the number of clicks, and the timeframe as well.

Predictive Filter: Men who Most Likely to Buy

In order to fulfill the conditions, you'll need to use two criteria and link them via AND operator.

  • As before, you'll begin by selecting Include in my audience.

  • You will find the gender criterion within the Customer dropdown

  • You will use the Propensity to Buy criterion within the Predictive Metrics dropdown. As you're looking to restrict the audience to those most likely to make a purchase in the coming weeks, you will set a range from 70 to 100%.

  • Once you've selected your filter criteria, click the blue Compute audience button to view the audience size, and the blue Create button to finalize the creation of your filter.

In the same way, as for the propensity to buy, you can create filters based on the Customer Lifetime Value (i.e. the total revenue you can reasonably expect from a single customer in the next 12 months).

You can define the range of users you want to use, with users ranked from lowest CLV to highest. The CLV value is displayed below the slider.

You define the value range: an interval between 20 and 50% allows to select the individuals whose value that fall between the 20th and 50th percentile in terms of CLV, i.e. between $287 and $535,4 in the example above. The maximum CLV value (the very last percentile) is shown on the right ($1500 here).

Additional Filter Examples

While the above examples represent three types of filters that might commonly be applied to campaigns, there are numerous filters combining various criteria that can be created using Tinyclues' Custom Audience Filter feature; but the fundamental workflow for building different types of filters is largely the same.

Let's look at a couple of simple variations on our original samples.

To start, perhaps you want to exclude purchasers of a particular item from your campaign.

  • Here you simply begin by selecting Exclude from my audience.

  • As with the Brand designation, you'll find Product within the Purchase menu, and can then designate a precise product and purchase window (e.g., Exclude customers who purchased a Wii console, as seen below).

Or perhaps you're looking to target customers who didn't simply purchase a given item in the past, or, more specifically, customers whose last purchase was a specific item.

In this case, all steps remain exactly as you would expect, with one key addition - the selection of the Last Purchase check box, as seen below.

In addition to filtering based on specific purchases, you may also find it useful to build filters around generic purchase behaviors. For this, you can use the Generic Purchase category, and use one of the available conditions:

  • Any Item: Customers who have purchased any item in the date range selected - e.g., Customers who have carried out more than X purchases in the past 365 days

  • Total Revenue: The total revenue for each customer in the date range selected - e.g., Customers who have purchased products with an overall value of more than 500 USD from June 20 to July 20 (a 4-week sale period)

  • Single Order Value: Customers who spent a specific amount in one single order during the date range selected - e.g., Customers who spent 100+ USD on a single order in the past 180 days

  • Single Item Value: Customers who have purchased an item of a specific value during the date range selected - e.g., Customers who have purchased a single item of more than 50 USD in the past 90 days

  • Number of Orders: Filter customers based on their total number of orders during the date range provided - e.g., Customers who have carried out more than 5 orders in the past 90 days

Getting to Know Your Audience

While creating a Custom Audience Filter, you'll be able to check some real-time insights that will help you to better understand that audience.

Once you feel satisfied with the audience you have selected, you can click on Show KPIs to find thorough information about the customers included in your filter: Revenue in the last year, Average Order Value, age, etc.

Audience KPIs provide you with a lot of insights to better understand your audiences and identify ways to optimize!

In the KPI panel you'll find the following indicators:

  • Last Year's Revenue: Aggregated revenue from the past 365 days, along with a sparkline displaying the revenue generated by your audience every week throughout the year.

  • Purchase Patterns: Share of purchases of the top 5 offers (expandable to 10 values) of a given audience (in blue), along with the share of purchases of the whole customer base (in red) for that same offer.

  • Average Order Value (AOV): The average revenue per order for your selected audience, based on the purchases of the past 365 days. We also provide the AOV of your whole database for comparison.

  • Active Users: The share of customers that have purchased any of your products in the past 365 days in your selected segment, to be compared with the share of active users in your whole database.

  • Custom Segmentation: Gain insights into your audience at a customer attribute level, represented as bar charts, using the same dimensions as in our Audience Insights feature.

  • Demographics: Information about the distribution of the Gender and Age of your selected audience, complete with a comparison with the gender and age distribution of your whole database.

How to Manage Audience Filters?

You can view any filters you've created within the Custom Audience Filter main page (the page you land on when you initially navigate to the Custom Audience Filter section of UI).

Each filter is easy to identify by the name and description provided at the time of creation.

Remember: Each Audience Filter should have its own unique name!

You'll be able to :

  • Check the status of your audience size; a simple graph will indicate the growth trajectory of the selected filter from the day it was created

  • Update your Audience Filter and modify conditions as needed.

  • View audience KPIs by clicking the Graph icon.

  • Duplicate an existing audience filter.

  • Delete filters you no longer use (NB: Once you delete a filter you can't get it back!).

  • Refresh your audience. Click on the Refresh button to have a look at your audience at that specific moment. Once you have refreshed it, the changes will be reflected in the "Users" column.

  • Export your audience as a CSV file. Once you have the audience you need, you might want to take further action on those clients. Export the CSV file with the customer IDs.

  • Switch ON the Campaign Plan toggle to make the filter available and selectable when creating a campaign. They will appear with the Custom tag, as shown below.

Don't hesitate to reach out to us with any questions you may have about the feature or its potential uses! Questions and feedback are always welcome, and help us to provide the best possible user experience 💪🏻

Did this answer your question?