The COVID-19 pandemic has impacted the lives and work of people worldwide, and the Go Fish team is no exception. The financial impact of a drastically-changed business environment led some clients to pause, downgrade, or, unfortunately, cancel their work with our agency, and as a result, there was suddenly less work to go around.
Instead of dwelling on the negatives in that situation, Go Fish Digital decided to turn it into a positive by using the extra time we had to finally work on some dream tasks, improve processes, and build out internal resources. One of the big projects for the content team included improving the databases our team uses to create content campaigns, and that meant collecting new data. Lots and lots of new data.
What We Collected
So, what kind of data am I talking about? Since it is impossible to predict all the topics and tangents our content campaigns will involve in the future, what kind of information could we gather knowing with absolute certainty it will be useful and not a waste of time?
The answer, in short, was to collect a ton of demographic data.
Between official government data releases and sites that use millions of user-submitted data points to create accurate measurements, it is possible to gather data on population, average age, income, cost of living, and more from dozens of countries and cities around the world. If you look in the right places, you can find data so specific that it is possible to accurately compare the cost of a McDonald’s value meal in over 100 different countries. But the internet is huge, and those kinds of data sets are spread out and rarely prove easy to download in full. In the past, this meant our team had to hunt the web for these scraps of information and manually collect the specific data they wanted.
We wanted to remove that hunting step and compile all of this demographic data into a single, convenient digital space that our entire team can access and manipulate as needed, which is precisely what we did.
How We Collected It
Some of the data we were after comes from the U.S. Census Bureau. The Census already has relatively useful online tools for finding and downloading specific information, but their new webpage, data.census.gov, has made that process more accessible than ever. For information on things like population size, average age, income levels, and more for cities and states across America, this tool proved invaluable. It not only made it easy to find and review various data tables relating to what we were searching for, but it also allowed us to download those files and upload the data into customizable spreadsheets.
Census data was the easy part of building up our internal collection of information. Pulling global information as well as specific data relating to the quality of life and cost of living would prove to be a much more involved task.
Sites, such as Numbeo and City-Data, have been utilized by the GFD team time and again to help us compare and rank cities against one another in campaigns. They also contain high-quality data sets that are nearly impossible to find anywhere else. This data is particularly valuable because they apply the same standards across their entire database, meaning that, when a metric is listed for two different cities, we can be certain it is a fair comparison even if those cities are on opposite sides of the earth.
In the past, our team has manually reviewed and collected the information they need from these sites and others like them, only focusing on the specific locations featured in their campaigns. As a result, our projects contain bits and pieces of data from all over the world, but nothing comprehensive.
To rectify that situation, we decided to scrape our way through those sites. In the past, GFD has created programs that pull or scrape specific pieces of information from websites such as Yelp. We do this when we need to collect lots of comparative data from a site, a task we generally do not have the time to do by hand.
Recently, we decided to ramp up this scraping work internally thanks to an easy-to-use tool called Data Miner, which allows our team to design and execute custom data collection on all kinds of websites.
We turned our scraping skills on the sites mentioned above (among others), and hundreds of pages later, we had the results we wanted.
How We Can Use It
Demographic data is crucial to multiple campaign types that we frequently use here at Go Fish Digital, and having easier access to it only helps to simplify and speed up the process of creating those kinds of campaigns. City rankings often use multiple pieces of information included in the data sets we gathered, while having access to accurate population counts makes it that much easier for our team to create per capita evaluations that more accurately compare conditions from one place to the next.
Beyond the campaign styles we know will benefit from having this data, our team is now better situated to explore new campaign styles thanks to the increase of the information at our fingertips. Now, our team can review the data we have collected and use what is available as an inspiration. Knowing that you have reliable and accessible information on the cost of a one-bedroom apartment or the cost of gas across the world suddenly opens our team up to the possibilities of creating a campaign around topics that might have previously been overlooked because of how daunting the data collection would have been.
While it would have been easy to let the negative impacts of the coronavirus on our work and workloads drag us down with them, our team decided to find the silver lining in the situation. By taking the extra time we were given and using it to strengthen ourselves, Go Fish was able to remain busy and productive during the pandemic and emerge stronger than ever, with increased skills and assets (including a whole lot of new data) we can use to create the best campaigns possible for our clients both now and in the future.