Social Feed Analyzer

Inspiration

In 2020, there was a report that came out that stated Kiwis lost $17m to cybercrimes. Our inspiration was to build an app that can help Kiwis avoid scams. It is important to note that our organisation is a financial institution. Our social media account pages have customers asking if a certain text or email is a scam or not. We believe an app that can address these types of queries as soon as possible can help avoid cybercrimes. It also has the added benefit of significantly decreasing average response time by automating the workflow. That app is Social Feed Analyzer. It automates the manual process of replying to social media posts.

What it does

Social Feed Analyzer listens to one’s social media account’s feed for new mentions or tweets. With its ability to understand human language, it analyses each message and based on the outcome, an appropriate response is instantaneously posted on behalf of the account owner. Messages with positive sentiments receive an automatic thank you reply using an STP. On the other hand, messages with negative sentiments create Feedback cases, which are routed to appropriate work baskets. This workflow decreases response lead time significantly.

How we built it

Social Feed Analyzer is built using OOTB functionalities of Pega Platform, most notably Pega NLP Text Analyzer. It is the single most important part of our application. It acts as the brain for our decision system. Also, we built it by strictly following the Pega Express Methodology as we wanted this experience to mirror the real-world development cycle.

Challenges we ran into

As we are incredibly new to Pega, we ran into many challenges, admittedly some of them became a challenge due to our inexperience. Simple tasks, such as work routing and how to properly configure an activity, became complex tasks to us. The main challenge we ran into though was the integration between Social Feed Analyzer and Twitter API. It took us a few days to configure it properly, but once we got that one sorted out the rest of our tasks fell nicely into place.

Accomplishments that we're proud of

Our team is composed of four recent grads. Our experience with Pega is just under seven months. We view this experience as our first major milestone in our Pega career. We are incredibly proud that we were able to utilize different functionalities of Pega, especially the NLP Text Analyzer, to develop this app.

What we learned

We learned the different phases of Pega Express Methodology. It helped us deliver Social Feed Analyzer on time. Also, we learned how important to properly define your Minimum Lovable Product; it is far too easy to introduce scope creep. As for the technical side of learning, we learned Pega NLP and how to configure Agents and Rest connection. We hope to apply this learning to future projects.

What's next for Social Feed Analyzer

We plan to continue developing Social Feed Analyzer because of its potential benefit to our organisation and also for our own learning experience. There is so much more we can add to it. Future iterations will include Instagram and Facebook integrations. Ultimately, our target is to turn Social Feed Analyzer into an omnichannel.