Channel Scoring – how to configure the function in the SARE system?

The Channel Scoring module allows you to check, among other things:

What is the number of users involved in a particular communication channel (email or SMS);
How the audience base changes over a certain period of time;
How a particular user behaves;
What does seasonal user engagement look like at the base.

Scoring is intended to help us choose the right type of communication over time. If we use additional segmentation (get to know our audience’s preferences better – what they click on, what they buy and when), we can better plan all communication.

Where to start with Channel Scoring configuration?

To start calculating a new scoring we go to the tab Addresses -> Scoring

Here we can start calculating a new scoring using the simple configurator (1), the advanced configurator (2), or read the brief instructions.
There is also a tab with a list of already configured scorings.

Scoring calculation – simple configurator

To go through the simpler configurator, select the Start button. In this option, there is no need to set the weight values of the various shipping parameters yourself. The system will do it for us automatically. After clicking Get Started, a window to set the scoring name appears:

Step 1

We start the configuration by selecting the communication channel we want to analyze. We have a choice of e-mail or SMS communication. If we want to examine only emails, we select them and move on:

Step 2

Now we need to specify what type of campaigns are sent to our recipients. We have 5 different types to choose from, each of which is described in detail. Choosing the right type of communication is very important, because each of them systemically has different weights assigned to opens, clicks and deliverability, which affects the calculation of scoring.
For example, in sales and advertising campaigns, click-through rate will be given more weight, and in information campaigns, message deliverability will be given more weight.

Let’s assume that our messages are informational and advertising dispatches. We mark them and proceed to step No. 3.

Step 3

At this stage, we need to set with the sliders the percentage of the types of communication selected in the previous step in all shipments. Note that the sum of the two ranges must total 100%. The system will help us set these figures.

For example: informational messages account for 30% and advertising for 70%.

Step 4

In this step, we choose which addresses from the database we want to take part in the scoring, which ones we want to exclude and what time range we want to analyze. In selecting the groups, it is important that they are appropriately matched to the types of mailings we selected in the 2. step .

For example, we want to analyze addresses that simultaneously meet all three conditions:

They participated in the shipments.
They are located in ID Group 1.
They are not in ID Group 2.

The selection of the time range for the analysis is also very important and should be set

According to the frequency of shipments made.

If we are running multiple campaigns, the appropriate interval would be four weeks, so that the data is accurate.
However, if you send campaigns from time to time, then it’s a good idea to choose a longer range so that the data is meaningful and not dependent on just one message sent.

Suppose we send a very large number of messages and want to analyze a period of 4 weeks:

We have the ability to select all groups with one click and analyze our entire base:

NOTE: If we select all the groups and mark one of the groups to be omitted below, the indicated group will not be taken into account. If we want to remove any of the already selected groups, we can easily do so by clicking on the X button:

Step 5

In the last step we get a summary of what we managed to set up in the previous steps:

In this step we can also determine to which characteristic of the e-mail address the scoring should be assigned. This way we will always be able to view the current scoring value next to the address.

In the case of active scoring in the account, the value in the trait is updated daily if there is a change in its value after calculation. In other situations, the value from the last score remains.
After clicking on the Save to trait option, a drop-down menu will appear:

We can select a trait that already exists in the account, or we can click on Create New Trait
and add it at this stage. After selecting a feature, we click CONFIRM and then SAVE.

IMPORTANT!

If we have several scorers in the account, each of them must have a separate feature to record the results. It is worth remembering that each scoring performs operations on the base, which can affect its performance. Therefore, we recommend setting one active scoring in the account.

Finally, we click on the RECALculate SCORING button:

Once clicked, you will see that the scoring has started to recalculate, and the results will be available the next day in the Scoring List section, as well as in the Reports tab -> Scoring.

IMPORTANT!

It is worth remembering that the active scoring in the account recalculates daily, so the rating of individual addresses will change. The set time range under the selected scoring each day moves by one day on the calendar. Therefore, it is important to remember that when analyzing one day’s score for a given group of addresses, the system takes into account the marked range. When we check the same scoring, for example, 5 days later, we must remember that the result may already be different. The date range of scoring has also moved by 5 days and takes into account dispatches from this new range.

ADVANCED – Modification of scoring weights

Before recalculating the scoring, by clicking on CHANGE WEIGHT, we still have the option to modify the weights that were set by the system based on our indications in the earlier steps. This already involves the advanced configurator.

After clicking, a window will appear, in which we will see what weight the system assigns to the parameters:

Here we can change at our discretion what value to choose for opens (OR), clicks (CTR), deliverability.

By determining the scoring based on email and sms records, we look for the ideal user for each analyzed day and over the analyzed time period.

An ideal user is one who showed the most activity (had the highest OR open rate and CTR click-through rate among those analyzed).

In this assessment we use, among other things, percentile, which allows us to determine what portion of the base scores above or below a given threshold.

For example: specifying a percentile at the 90th level means that 90% of the collection is at the 90th level and only 10% score above it.

According to the set percentile value, the entire scoring will be recalculated, so it is worth checking the average opening results from your campaigns first. If the opening % is low, it is worth setting the percentile at a high level, on the other hand, if the opening % is higher – the opposite.

This procedure will ensure that the results of the scoring will be most meaningful and avoid equating the converted addresses to a single, misleading result.

Scoring calculation – advanced configurator

We have the option to set the values of individual parameters for calculating the scoring ourselves. To do this, we start the calculation by clicking on the option SET scoring weights independently.