The 3 Ps of Backlog Management –and What to Do About them
Last week at the SX Live conference, I was lucky to present with Chris Romrell of NICE and Rob Hartwig of SupportLogic on the topic of backlog management. Here’s a synopsis of my part of the presentation.
Backlog can creep up on you or be a constant companion, but in either case backlog is bad
for customers, who experience longer resolution time, have to escalate issues, and as a result have poor support experiences
for your team, as the support agents shoulder higher mental loads, managers are in firefighting mode, and CSMs fight churn
Before you can select a proper backlog management solution, it’s essential to establish a root cause for why the backlog is high. I like to organize root causes into the 3 Ps:
People. This is everything that has to do with the particular agent who owns the case:
- Insufficient technical skills
- Weak troubleshooting skills
- Poor time management skills
- Afraid to say no to customers (so single issues often blossom into many more, with no closure possible)
- Scared/ashamed to ask for help (so they remain stuck for too long without making progress)
Process. This category is for issues that are beyond the scope of a single agent and instead are related to the organization.
- Insufficient staffing
- No/weak collaboration process: there’s no easy way for agents to request help when they need it
- Poor backlog management: there’s no organized process to keep backlog under control
Product. This category focuses on the actual product or service you are supporting.
- The product is buggy, and cases linked with bugs pile up
- The product is mis-sold: customers’ use cases don’t fit the product capabilities so they report issues that can be intractable
- The product is hard to use
So what can you do to fight backlog? Using the same 3Ps to organize solutions,
- Tackle people issues through better training whether it’s for product knowledge, troubleshooting, or soft skills; coaching; and maybe changing your hiring model to bring in differently-skilled agents.
- Improve your case-resolution processes (first light-blue column, below): assign cases more precisely, pump up nowledge management, offer case clinics, and implement a swarming process for collaboration.
- Upgrade backlog-management processes (second light-blue column): set resolution targets appropriate for your environment, as they are a powerful psychological marker of what constitutes an “old” case, perform individual case reviews, review aging cases, and employ escalation detection methodology.
- Mitigate product issues both tactically, by setting SLOs with the Engineering team, and strategically by creating a Voice of Customer program with the Product, Sales, and Engineering teams.

Many processes can be supported by AI-based technology including automating QA ; automating case assignments, whether for initial case assignment or as part of a collaboration process; and sentiment analysis to detect escalations. Are you using AI to help with backlog management? Share your experience in the comments.