Launching Campaign Smart Tagging: New Feature Makes It Easier to Measure Campaign Performance

The 4 ‘R’s of AI-driven (Artificial Intelligence) marketing constitute the sole success mantra for us data-driven marketers.

What are the 4 ‘R’s?

The right content delivered to the right user at the right time through the right channel. This forms the bedrock of any modern marketing strategy.

Constant measurement of analysis of each campaign’s performance is crucial as it provides deep insights into what works and what doesn’t.

Key metric analysis, goal-based tracking, device-based performance analytics, etc. are features that make your life much easier as a marketer using a marketing automation platform such as Smartech. Your Marketing Automation tools require consistent upgrades and updates.

But, we don’t want to stop just there!

Our latest advanced feature - Campaign Smart Tagging – is all set to take your campaign measurement and analytics game to a new level.

Think of an intelligent marketing automation platform that can also read through the content of your campaign, identify important keywords, and assign them as tags under the right categories! Wouldn’t that further simplify your multi-channel marketing automation efforts?

Regardless of the industry you operate in, Campaign Smart Tagging is the shot in the arm that your digital marketing strategy needs. For instance, this is how Campaign Smart Tagging would generate tags for an e-commerce platform:

Campaign-Smart-Tagging-Mobile-Example 
Subject Line Categorization:

 - Seasonal
 - Discount
 - Gents
 - Product Category

Body Level Tags:
  • Product category -
     - Shirts
     - Winterwear

    Brand -
     - LP
     - Jack & Jones

How Does Campaign Smart Tagging decode the Email Content?

Smartech’s smart Natural Language Understanding (NLU) models, which fall under AI, are trained on different types of email content and are capable of scanning through the email’s subject line and the copy itself, to generate the appropriate tags and categories.

When the NLU model scans the subject line ‘Special Discounts on Menswear’, it captures the keywords and then finds the appropriate categories in which the subject line can be classified; i.e. ‘Discount’, ‘Product Category’, and ‘Males’.

Interestingly, NLU also ‘understands’ that the subject line is related to the male users’ category, as it contains the word ‘men’. Similarly, the body copy passes through Smartech’s Optical Character Recognition (OCR) engine. It first extracts the relevant blocks of text and then identifies the different keywords in it.

Based on these keywords, which can include brand names, product names, discounts amounts, etc., the NLU engine categorises them into 3 high-level categories – ‘Product Category’, ‘Brand’, and ‘Discount’. Once these granular and high-level tags are available, a host of advanced analytics opportunities open up.

How Campaign Smart Tagging Helps You Understand Customer’s Intent & Improve Your Content:

While the number of use cases that Campaign Smart Tagging can address is extensive, here a few key analytical insights you can draw from this next-gen feature:

1. Tag-Level Performance Analysis

Post-campaign performance analysis is an extremely important tool for you as a marketer, enabling you to measure your campaigns’ successes and gaps. You need to ensure your strategic decisions are backed by significant data analysis.  

While aggregating the campaign-level metrics such as open rate, click rate, conversion statistics, etc. over a period of time does help, the data is not sufficient to allow you to deduce which content performs better or worse than others.

NLU helps you aggregate the campaign metrics at the tag level – high-level or granular tags - to properly understand their impact on the underlying metrics. The insights you gather will astonish you and help you come up with the right strategies to address the specific campaign and marketing challenges.

While working with one of our e-commerce clients, we analysed their tag level campaign performance for a quarter and found that subject lines mentioning a few brand names generally resulted in a higher open rate than the ones mentioning product categories. But, once the email gets opened, mentioning appropriate product categories with proper redirect links exhibited a higher chance of conversion.

This showed that their customer base was perhaps brand-focussed in its purchase intent – inclined to make a purchase if they find something new and relevant from their preferred brand. But, they do not have enough time to browse through a lot of choices once they land on the brand specific page, as most of them may be irrelevant to them.

It was concluded that the client’s customer base was niche with good spending potential. Targeting them with messages containing popular and highly valued brand names in the email subject line was likely to be more effective.

Also, placing relevant product picture thumbnails with the appropriate product categories   in the email body ensured higher chances of eventual conversion.

Related read: Conversion Tracking: Now Easily Measure your Multi-Channel Marketing ROI

2. Design Selection and Improvement

Marketing is no trivial art, and like with any form of art, there is often limited consensus between creative minds around what is the best campaign design.

The performance or the impact of art is often not measurable on a tangible scale. But, implementing a tag-based campaign evaluation methodology can help you understand what type of campaign designs & creatives actually work or not. This makes the creative aspect of your campaigns more measurable.

For instance, let’s assume that there are 2 marketers - X and Y – marketing for an e-commerce brand. They have different ideas regarding designs for a digital marketing campaign. They both design and execute their respective campaigns over a period of time.

Post campaign, tag level analysis results can reveal if X is more adept at designing discount campaigns and Y designed better non-discounted promotional campaigns.

Another way it can be used is to track the performance of certain categories of campaigns with respect to time. In this case, we can notify the campaign team or marketers about such downward trends and ask them to take corrective actions in design or execution.

3. User Profiling

As a marketer, you would develop a single template for any campaign, which is then sent to a large target customer base. But, embracing a data-driven approach allows you to consider the perceptions, preferences, and behaviours of individual customers.

For instance, let’s assume there are 2 customers - John and Brad - who prefer different sporting brands - Adidas and Puma – respectively. Now, in a sports shoe campaign, consider that the subject line of the email is this:

‘Exclusive discounts of up to 40% on your favourite sports shoes from Adidas and Puma!’

There is a chance that Brad may not be able to see his preferred brand mentioned if he is reading his emails on his smartphone screen since ‘Puma’ comes after ‘Adidas’ in the subject line. This may cause him not to open the email.

Similarly, there can be another case, where extremely brand-conscious customers are able to view the subject line only up till the characters mentioned in blue from the below subject line on their hand-held device:

New Mobile Launch Alert: First on XYZ Apple XR”

They would not be able to see the mention of one of the best-selling brands in it, as the subject line is long and cannot be displayed entirely.

As a marketer, you obviously want Apple patrons to see what is new and consider a purchase decision. But, as this is only a small segment of your customer base, you may lose out on this niche segment. On the contrary, whatever email opens you get against this campaign, even though high, may be coming from certain segments that are not Apple users.

In a nutshell, customer-level understanding of choices proves extremely relevant in defining a successful campaign.

Using the tags and categories identified by Smartech’s NLU engine, you can segment your target audience based on tag level affinity (body and/or subject line level) and route the relevant campaigns accordingly.

Campaign Smart Tagging has some really impactful applications when it comes to tag level intelligence. While it is always possible to manually tag the campaigns, Smartech brings this AI-powered smart tagging feature that ensures that the tagging is data-driven, relevant, and accurate. 

This feature can help you cut down on your costs spent on campaign tagging (in cases where external teams charge to undertake this aspect of campaign management), reduce the intelligence transfer, and consumption time; resulting in faster decision-making.

To know more about a host of other such useful AI applications to track and measure your multi-channel campaign’s performances, get in touch with us today!

 
Debapriya Das


I am currently working as Lead - Machine Learning at Netcore, and have a rich experience into analytics, deep learning, DWBI, solutioning, consulting, and project management. I have spent a significant amount of time in my career, building solutions and strategies for pricing, loss mitigation, risk scoring etc in the Insurance industry. Most of the other problems that I have worked on have been pretty spread out and abstract, ranging from problems which can be solved by traditional statistical/machine learning models to practical image recognition and video intelligence.

Comments

We promise that we won't SPAM you.