A rigorous methodology that produced a defining layer of governance
We addressed this challenge with our Lean Six Sigma mindset – using qualitative and quantitative methods, such as statistical analyses and process mapping – to understand and improve the way work was done. The goal? To increase performance, decrease variations, and detect errors in processes, such as broken links.
We created an extra layer of governance across the company's partners and provided a consistent picture of their activities.
We found that each agent handled copy for about 160 ads a day and ran 15 quality checks for each one. That amounted to 2,400 decisions per shift. At the same time, agents had to manage country-specific nuances and websites that change frequently. Through no fault of their own, mistakes crept in.
We worked with the company's partners to set overarching performance parameters for managing ad quality. With their input, we introduced metrics designed to reduce rework. And we established standard operating procedures that, for example, handle errors.
At the same time, we analyzed ads across 15 industries. We categorized different elements, such as what business the advertiser wanted to promote, what attracted customers, relevant hyperlinks, and links to relevant subpages. We used these insights to create guidelines that help improve ad quality.
The technology we introduced to automate processes supports agents in their roles, cutting out transactional elements, providing insight and alerts when ads don't meet policies, and enhancing their work.
To keep things on track, we held regular calibration meetings and workshops to discuss policy details, process standardization, insights from trends we observed, and ways to enable continuous improvement. This effort paid off. Our transparent methods won the trust of all partners, which helped us establish the final, conclusive steps and routines that ads must go through before they go live.