A Technical Guide to how Civis Identity Resolution Works
Download the Whitepaper
Software-as-a-Service (SaaS) platforms enable continuous, on-the-fly improvement without complex, lengthy installation and update processes. This means we can update Civis Platform in the background, usually without any outage or client intervention, so that users can continue their work while gaining new functionality and increasing performance.
Of course, constantly updating Platform means constantly working behind the scenes to make changes and come up with new product ideas to help clients fulfill their data management and data science needs. So how do we make that happen?
Continuous improvement as an ongoing process is best understood as a four-step cycle: Plan, Do, Check, and Act (PDCA). Let’s look at how these steps are applied to Civis Platform.
Civis Analytics understands that data tools are often difficult to use, and we aim to provide the ability to manage and process your data to get the answers you seek as quickly and easily as possible. To that end, we constantly strive to help clients find the correct tools for the job and learn these tools swiftly, in addition to removing complexity and difficulty so that users never feel lost or stuck.
This is where Civis’s product and design teams focus our efforts: it’s how we measure success, and we’re always coming up with new ideas to make Civis Platform better. We pull our ideas from many sources, but the most important is user feedback. Our account managers and client success specialists stay in constant contact with users to provide assistance and gather feedback on what is, or is not, working well.
As a product manager, I review this feedback with my fellow PMs and designers to look for patterns or gaps posing trouble for multiple clients. In some cases, this might be a relatively simple change or bug fix, so we work to make those changes available as soon as possible. For more complex changes, PMs and designers consider the problem and start brainstorming potential solutions.
Once a general solution is agreed upon and an initial design is worked up, we reach out to users to get additional thoughts. Depending on what we’re working on, this can be users here at Civis, such as our solution architects or data engineers, and/or users from our client base. In all cases, we want to make sure that we get as much information as we can, since these individuals are the ones on the ground using Platform, and they know best what they need. We then iterate and refine our original design until we are reasonably confident that we have a strong direction and a clear idea of where to begin.
The product manager(s) then works with our engineering team to implement the design. To ensure that we make changes quickly and efficiently, Civis embraces an agile development methodology, meaning we don’t necessarily build a complete plan or define a very large project; the goal is instead building, releasing, and getting feedback as fast as possible.
Our developers, PMs, and designers work hand in hand to break down the initial plan into smaller chunks of work. As we release these updated pieces to end users, we continue to refine the overall plan and adjust our goals.
Once a new feature or piece of code is released, the next step is confirming that the changes we made meet their intended goals. We gather both quantitative and qualitative information from our users: for the former, we use a variety of software tools (both our own and third-party) to get hard numbers, a process that can be as simple as determining “How many users are using the new feature?” and “Is this more or less than last month?” We also use Platform for our own data science needs, breaking these numbers down and answering questions about more specific use cases and personas to make sure we know who is using a particular feature and why.
For qualitative information, we again engage in user interviews and gather assessments from our client base to see what our users think and to hear their ideas for further refinement.
Based on the data we collect during the Check phase, we determine the new release’s success. In the best case, all the feedback is positive, and we continue with our existing plan.
In most instances, there is something that still needs improvement: perhaps we did not consider a certain use case, or a new feature doesn’t solve the problem the way we anticipated. But that’s okay. The important step is recognizing and correcting the error, and that’s what this phase is for. Depending on the nature of the problem, we can simply update our plan to fill in the gap, or we can go back to the drawing board.
Civis will continue to regularly update Platform to make sure our clients have the best and easiest-to-use set of data analytics tools to solve their needs.