Connecting Quickbase to AWS for Sentiment Analysis
Quickbase is uniquely positioned to solve numerous business problems because of both the product's incredible flexibility and the different approach to how people build applications on the platform. The problem solvers who use the tool can resolve critical business challenges themselves in days and even hours on a system that IT trusts and can govern at enterprise-scale. Although powerful in business users' hands, the technically inclined can elevate Quickbase to new heights by integrating it with other systems through extension points such as Pipelines, Quickbase's native integration platform. Public cloud providers like Amazon Web Services provide many building block services that go together with Quickbase like peanut butter and jelly, and I want to introduce a sentiment analysis capability for Quickbase using Pipelines and AWS as a working example of this power. The Amazon piece is available at https://github.com/cpliakas/quickbase-sentiment-analysis, and developers can replicate the pattern for other AWS services as well. As context, Quickbase applications often capture data at the edge of the business. The more data that is captured, the more important it is to automate data analysis to extract insights from it. For example, let's say you want to crowd-source inspection of inventory, products, etc. The goal is for employees to snap pictures and submit comments when they happen to notice something in the field, such as an example of a great product display or a misplaced item. Within minutes, you can create a Quickbase application that is mobile-ready and able to be used in the field. A welcomed problem is when the application becomes widely used across the organization, and manual analysis of the data becomes impractical. One way to start categorizing the data is through sentiment analysis of the comments to sort the feedback into positive, negative, and neutral submissions to determine which ones to act on first. This situation is where our example comes into play. Amazon has a service called Comprehend that provides sentiment analysis. At first glance, the integration appears to be dead-simple. However, many hidden complexities around authentication and infrastructure configuration take time to figure out. The project at https://github.com/cpliakas/quickbase-sentiment-analysis provides a one-click install of the Amazon piece of the puzzle through theServerless Application Repository that has Quickbase assumptions which make it really easy to securely integrate with Quickbase through Pipelines. The video below walks through an end-to-end setup so that you can start experimenting with the capability now. The https://github.com/cpliakas/quickbase-sentiment-analysis project is released under the permissive open-source MIT license. You are free to use and copy the tool and apply the patterns in whatever manner you wish, including commercial applications where people pay for your service. I hope you find it useful, and I look forward to learning about where the tool and techniques are providing value to your Quickbase applications. ------------------------------ Chris Pliakas ------------------------------
Running an Excel/CSV through Salesforce
Hi all, I'm a sales manager here in Chicago and brand new to QuickBase. I'm also pretty new to my current company and our Salesforce data is atrocious. Before I can even begin to fix some of the back-end processes, I need to find a way to upload all of the disparate lead repositories that my teams are working from (word docs, excel files, google sheets, etc.) Is there an Exchange app or process you could recommend that would allow me to upload an excel or csv file and recieve a report of which of those accounts are in Salesforce already (based on account name, address, phone number, homepage url, etc.) and the current account owner? Separately, if there's an existing Exchange app that would allow me to upload new accounts and contacts into Salesforce from an Excel or CSV file and assign account owners automatically, that would be incredible as well. Thanks, ------------------------------ Matt Chinnis ------------------------------