Case Study - 600+ Emails Collected: How Facebook Chatbot Automation Drove Lead Generation

Overview: Creating an Automated Facebook Chatbot to Streamline B2C Lead Generation for PizzaForno

While working with PizzaForno, we had to come up with a way to connect with their customers at each of their locations to run local marketing campaigns. They wanted to increase their brand awareness, help drive promotions at the local level, and collect more data on their customer base in order to efficiently run future promotions to them. By creating an automated Facebook chatbot while utilizing geo-fenced ads, we we're able to efficiently capture user information (emails) and run local promotions that would automatically distribute targeted coupon codes.

Identifying the Key Challenges

PizzaForno operates under a unique business model, as each of their locations are automated pizza vending machines with no onsite staff.

To implement this digital strategy, we needed to iron out some minor details.

This involved developing a scalable model that would enable the team to run promotions effectively at the local level across their entire fleet.

It also involved targeting users within a specific proximity to any of the locations and providing them with one-time use promotional codes. The one-time use coupon code was necessary to track the campaigns effectiveness.

The main challenge we would face is that each coupon code could only be used once and was valid exclusively at specific locations. Therefore, if a customer received a code for "Toronto" and tried to use it in "Guelph," it would not work. Due to this, we required a system capable of distributing unique codes to individual customers at each machine.

Therefore, each machine location required a unique set of codes. It was essential to ensure that end users received the correct code for the machine they selected; otherwise, the code wouldn't work, potentially harming our customer engagement.

So, when approaching this, we needed to find a way to collect the customers information, then segment each user based on their specific location. Additionally, we had to target them based on their proximity to the machine and then distribute the appropriate code automatically.

The Solution: Tailored Digital Marketing Strategy Leveraging a Facebook Chatbot

Due to PizzaForno's unique business model, we had to come up with an extremely specialized solution to make our campaign work.

The first step involved coming up with a strategy to target the right consumers and provide them with a one-time use coupon code. This required careful planning in our ad creatives, as displaying a "one-time use code" directly would make it unusable after the first redemption. Moreover, creating an unlimited use code was not an option, as it would prevent us from tracking which customers actually used it.

Next, we needed a method to track the specific location for which each customer wanted a coupon code, as the code would only be valid at that location. When running a campaign across dozens of locations, it was essential to integrate custom data layers while collecting customer details. This allowed us to accurately segment customers and ensure that the coupon code they received was valid for the machine they selected.

Based on the success of previous campaigns, we knew Meta platforms performed the best for us, so we chose to run geo-targeted ads on Facebook for this strategy.

In order for all this to work, we needed to meet the following criteria:

• Run geo-targeted ads to reach customers within specific distances from our machines to facilitate the local promotions

• Collect the customers information

• Tag them with specific data layers to ensure the code we sent matched the location they selected

• Distribute a valid one-time use coupon code to them

At this point, we had already developed a system for a different campaign that was capable of effectively tagging customers with the correct data layers.

However, to leverage the systems we had just built, we needed to direct users to a specific landing page on our site where they could submit a form.

While this approach might have worked fine, we recognized that it could make the campaigns less effective since it required users to leave the Meta platform.

This is because Facebook prefers to keep users inside their platform, allowing them to see more ads, which generates more revenue for Facebook. Redirecting users off the platform tends to increase costs, as Facebook charges more for ads that lead users away.

By keeping users on the platform and using a specific campaign objective, we could lower our CPC and streamline the customer experience, enabling consumers to easily retrieve the coupon code without leaving Facebook.

Even though this approach created more hurdles and additional work for us, we knew it would ultimately improve the success of the campaign.

But, how would we do this?

Using an automated Facebook chatbot.

Implementing the Messenger API

When creating our plan for this campaign, we decided to use Facebook ads. Through our research on integrating various strategies, we came up with the idea of creating a custom Facebook chatbot.

This chatbot allows us to gather the information needed to send the customer a coupon code for the location they select, all directly within Facebook.

To build a Facebook chatbot, we created a Facebook Developer account. After submitting the required documentation and verifying our account, we set up the Messenger API within the Facebook App Dashboard.

Due to the permissions required for our chatbot to function, we had to create a prototype and submit it to Meta for app review. After completing this second verification process, everything was ready to go.

This configuration solved our first problem, which was enabling the PizzaForno Facebook page to automatically send and receive messages through the API we planned to create.

This setup allows users to message the Facebook page directly from our ad, triggering the chatbot to start the interaction. Once the chatbot collects the user's data, it sends this information to our API. We can then use this data to distribute the coupon code to the consumer.

Building the Facebook Chatbot

The next step was to create the "message flow" users would follow when interacting with the Facebook chatbot.

For this to work, we needed to collect the specific data required for the audience fields and tags used in our email campaigns. At the same time, we had to ensure the process was easy for customers to follow.

The message flow needed to guide customers smoothly through the steps to receive their code, ensuring they didn’t drop off. If the process was too complicated, they would simply exit the messenger and continue scrolling on Facebook.

This required some planning and testing, but we eventually got the entire process dialed in.

We first needed to create a trigger that would initiate the chatbot sequence.

When the customer enters the chat via an ad, a message appears saying, "Hi! Tap the 'Yes' button to get your $5 coupon code!". This was specifically made for the campaign we we're running at the time.

When the customer taps the 'Yes' button, it triggers the chatbot sequence to begin.

After the chatbot sequence starts, it asks the customer for their email, which we needed to send them their coupon code.

The cool thing about this feature was since the user was already in Messenger, a pop-up would appear with the email address linked to their account. They could simply click the pop-up Facebook displayed to autofill it, saving them time from typing it out manually.

This simplified the process and helped us guide them smoothly through the message flow.

Next, the chatbot would display an interactive pop-up, allowing them to choose their region. This was a necessary step which allowed us to collect the specific data layers we needed to implement our strategy.

After this, a second and final interactive pop-up would appear, displaying all the relevant locations closest to the customer. This allowed the customer to select the specific location they wanted the coupon code for.

On this popup, we included an image, name, and address of each location to help customers easily recognize it. We also added a "Go Back" button in case they selected the wrong location.

That was it! Thanks to the autofill feature, customers only had to click four buttons, and we had all the necessary data to distribute a coupon code that would work at the location they selected. The entire message flow would take the user less than a minute.

With the message flow complete, it was time to implement it and begin testing!

Integrating the Facebook Chatbot

It was now time to integrate the Facebook chatbot with our other systems to facilitate code distribution to customers and collect their contact details for future marketing campaigns.

We developed a custom API endpoint to relay data from our Facebook chatbot. Once data is received, the endpoint makes a request to our database to fetch a coupon code specific to each customer's location.

It retrieves the appropriate coupon code by matching the location identifier (which was selected by the customer during their chatbot interaction) with the corresponding set of coupon codes stored in our database.

This system not only ensures that the correct code is allocated but also includes functionality to track whether a code has already been sent to a customer. This prevents customers from receiving a code that has already been "used".

Following this, this endpoint initiates a request to the Mailchimp API, our chosen platform for email marketing.

This request imports the data collected from the chatbot, along with pre-configured custom audience fields and tags (including their one-time use code) into Mailchimp.

By incorporating specific audience fields and tags like location and coupon codes into our email campaigns, we can segment our audience more accurately. Which in turn helps us create personalized emails that are much more effective.

Delivering Personalized Coupon Codes via Automated Email Campaigns

Once the contact details were imported into Mailchimp via the API, consumers were automatically enrolled in a marketing journey.

As part of this campaign, they would receive an email containing the unique coupon code for the machine they selected through the Messenger app.

Thankfully, we were able to leverage an automated code distribution and data collection system that we had recently built for another campaign, making this part of the strategy relatively straightforward.

The code was then dynamically inserted using merge tags, along with another merge tag specifying the location where the code was valid. As shown in the picture above.

Again, all this data was made possible by the custom data layers and tagging system we developed:

This streamlined process ensured that promotional codes were automatically distributed to consumers as soon as they completed the chatbot message flow.

Creating Geo-Targeted Facebook Messenger Ads

With the Facebook chatbot built and integrated into our systems, it was finally time to put it to use.

We integrated our app, "pizzaforno_chatbot," with our Facebook Ads Manager account. Once set up, we were able to utilize it in our ads.

We selected several locations to target and created separate ad sets for each one under a single campaign, with three ads in each ad set.

The conversion location for each ad set was set to "messaging apps," ensuring that the call-to-actions (CTA) in the ad directed users to Facebook Messenger.

Using geo-targeting, we adjusted the distance for each location based on factors like population size. However, for most locations, we targeted a 1-2 mile radius around the machine.

For each ad, we integrated our "pizzaforno_chatbot" app under the "Message Template" section, which meant those ads were now directly connected to the chatbot!

When a customer clicked "Send Message," they were directed to the Messenger chat, where they would go through the message flow with the chatbot.

How Our Facebook Chatbot Worked

Geo-targeted ads prompted nearby users to initiate a chat to receive a $5 coupon.

Upon engaging with the ad, users were guided through a simple process:

• Initiating the sequence

• Entering their email

• Selecting their location

• Choosing their preferred machine

Once the user completed the final action in the message flow by selecting "Choose", the API processed the information, and within minutes, the user received their unique $5 off coupon code via email.

Step 1: Engaging Customers via Geo-Targeted Facebook Ads

The customer is shown a geo-targeted ad on Facebook, prompting them to click "Send Message," which connects them directly to our Facebook page via messenger.

Once the chat opens, both a message and a button appear, saying, "Hi! Tap the 'Yes' button to get your $5 coupon code!"

When the customer taps the "Yes" button, it triggers the chatbot sequence.

Step 2: Chatbot Collects Customer Data for Coupon Distribution

After the chatbot sequence starts, it collects the customer's email and then displays an interactive pop-up that allows them to choose their region.

Following that, a second interactive pop-up appears, showing all the relevant locations nearest to the customer. This allows them to select the specific location they wanted the coupon code for.

This data is sent via an API to our system, where it is matched with a discount code linked to the machine identifier selected by the customer. It is then imported into Mailchimp through another API, with the necessary tags applied.

Step 3: Customer Receives Personalized Coupon via Email Automation

Finally, once the customer's data is imported into Mailchimp, they are enrolled in an automated email sequence.

The coupon code is dynamically inserted into the email as a merge tag, along with another tag specifying the location where the coupon is valid, before being sent to the customer.

This automated funnel now gives us the ability to run local promotions tailored to each of the 70+ PizzaForno locations across North America.

Client Testimonial

The Results: How Our Facebook Chatbot Campaign Boosted Engagement

The goal of this campaign was to create a digital marketing strategy that could target consumers within a specific proximity to the machines, collect their information, and then distributed a coupon code to them. This approach would help drive customer engagement, increase foot traffic, and gather valuable data for future remarketing efforts.

Through a combination of strategies, we developed a system to address the challenges we faced.

By building a custom Facebook chatbot app and connecting it to our Facebook page and ads, along with utilizing a data processing and code distribution system from our previous email signup campaign, we were able to achieve some great results.

Now, we have this tool in our arsenal for future use as the company continues to grow and scale. Additionally, we now have segmented customer data that can be used for future marketing campaigns, enabling us to target specific locations more effectively.

After implemented this tailored digital marketing strategy, we gathered some impressive metrics to help measure the success of our approach.

Here are the results:

The Facebook chatbot ads initiated 1,100+ messaging conversations and collected over 600 emails, meaning more than 50% of users completed the message flow!

These ads achieved an average click-through rate (CTR) of 4.12%, and maintained an average cost-per-click (CPC) of just $0.29.

Thanks to the automated system we built, over 600+ unique one-time-use coupon codes were distributed to customers.