Help: Which Item do I Triage First?

When your team gets more tickets per day than you can keep track of, how do you sort through them and know which ones to prioritize and look at first? Of course you will try and get to all of them eventually as you don't want to let anyone down. What data points can you capture to try and sort out priority?

The dilemma: there will always be a subset of customer's whose tickets you will need to address first, but can you quantify which ones?

Define Prioritization Criteria

---

Start off by defining these concepts to ensure your priorities are well defined and aligned with other teams.

What defines a priority customer to your organization? Every environment will use different criteria to determine who gets first in line based on who you deem to be the most important customers. Some examples to get you started on this concept:

• "Top 25": who are the customers which make you the most revenue per year?

• Repeat / loyal customers.

• Customers threatening to cancel.

Issues common to many customers. Are multiple customers all calling in about the same thing? Even if they are smaller it may be smart to prioritize those issues. The cost to support teams responding to and fixing these tickets may end up being more than the cost of just fixing the root cause. Some factors to consider:

• What is the threshold to meet a common customer issue?

• How many contacts per week, day, or hour?

• How can people adequately correlate issues to be on common issue?

Critical issue that demands immediate resolution. There are always going to be some issues for which no matter how few people contact, they are too egregious not to fix immediately. To define this parameter answer these questions:

• What are critical issues for your company: data breach, data inaccuracy, or quality issues?

• Is it easy for your support staff to recognize?

• Can they be clearly defined so there are no false positives?

Collect and Process the Data

---

Now that you have defined what it is important, what is the most effective way to collect this data when the ticket / issue is entered?

Have a bug tracking system which is easy to modify. It is important to be able to be fluid with the definitions of how you want to track prioritization data. You should be able to add or update criteria on the entry forms easily as needed. See if you are getting some effective and useful data from these. If not, then change up the formula, ask for feedback to try and correct.

Collect simple data points. Start this process by collecting some some simple data points which require minimal training effort for your teams. Examples of this can be: top tier customer, customer size, threatening customer, customer unable to use product.

Use the data you have for automatic prioritization. You know your customer sizes and the amount of revenue each one brings in. Is there a way to tie account data on a ticket into your bug tracker? If so you might be able to glean info as tickets are entered about which ones to prioritize.

Try and score the data points you have, sort based on that score. Many of the data points you capture are yes / no or have a number. Turn that around and factor in each case and assign it a numerical value. Add all of your data points up and voila you have a rough score which can be sorted for your teams to use.

Correlation of common tickets via word cloud. Need to see trends and themes of which tickets are more common within a timeframe? Try putting a word cloud on your teams dashboard. Have it tie to the summary or descriptions of entered tickets within the last day or week. Every time your team goes to their dashboard they will see trending keywords along with their queue. Might help them know which ones to look at first.

Learn from this Data

---

Now that you are seeing trends from customers calling in with priority issues it might be time for some deeper analysis into that data so you can start to predict and better respond to customers complaints.

Large sized customers who contact frequently with problems might need additional assistance. There will always be a subset of customers who reach out more often than others. Learn who these accounts are. See if you can have account managers perform additional training with those who often need support. Have product owners meet with end users to better explain the design and reasoning behind the product. They can even solicit from feedback to make future iterations easier to understand.

Look for Tier 1 / 2 buzzwords which have led to major issues in the past. Review the incidents you have had in the past. Look at those tickets for some keywords which stood out. Generate some alerts based off of those keywords to notify the team proactively of a potential major issue.

Eliminate false positives. Ever had someone cry wolf or try and bypass the system by entering in data which they know will trigger you to look at their issue first? It may be time to perform some team training or change your data points over to empirical data to eliminate these kinds of issues.