When working with clients on web application performance optimization, I often ask for a view of operational costs and overall profitability. Did your eyes just glaze over? Don’t fret. That’s the response I usually get!
The point of my question isn’t necessarily to see the numbers, but to drive home the point that poor web application performance not only increases the incidence of support requests, it also hits in other areas.
Very few organizations take the time to calculate their Cost Per Operational Minute (CPOM) of their support center, and have their data blessed by their CFO as accurate. Note that this metric is quite distinct from Cost Per Incident (CPI).
Here’s a concrete example…
One of my clients in the web application hosting (SaaS) space advertises support for several thousand concurrent users. However, one of their customers experienced performance degradation at only 10% of that workload. This resulted in a Severity 1 incident being generated.
Because of its “Sev 1” status, the CIO was looped into the issue. And then his subordinate VP, Director, Managers…and ultimately yours truly.
It took about a week to get to the bottom of the issue, after numerous meetings, email threads, etc. Ultimately, it turned out that the customer’s database was improperly tuned – a problem I solved in 10 minutes. But the endeavor spun up the organization to the tune of $15,000 of lost productivity, by my estimates.
Given a 24x7 system has about 10,000 minutes a week of uptime, this incident added about $1.50 to the CPOM. That might not sound like a lot, but consider this was just ONE incident. Also, if there were an SLA for this application, this one issue’s costs would be multiplied by all affected customers.
With this story in mind, and armed with the data from your case management system on volume, duration and escalations, you can then put a dollar cost tag on the actual cases you have. I also suggest including estimates of opportunity costs – since your team is actively engaged in dealing with these preventable issues, they had to delay work elsewhere – such as product development, training, etc.
I want my clients to have an equally precise view of operational costs and overall profitability. It’s insufficient to simply total up all your FTEs salaries, add X% for overhead, and divide by your incident volume, arriving at a “cost per incident” estimate. Yes, I know you’ve done it.
The starting point is a profitability analysis that identifies all of your support organization’s operational costs. If your support or service management team doesn’t have the bandwidth or the expertise to lead this effort, I can help. Whether you undertake this effort on your own or retain me for the project, it is vital in order to build an effective baseline for decision-making — and to establish the support organization as more than just a necessary evil.
Life After Support
Beyond support, there are also implications for Marketing and Customer Success. Poor web application performance will create chatter in the customer community (read: bad press), and may discourage prospects. It also can have an impact on customer retention, lowering both revenues and overall company valuation. The key to getting your voice heard is to have your data at hand and your CFO’s backing on the financial implications.
Retail isn’t the Only Vertical Affected by Poor Web Application Performance
A great deal of attention is placed on retail performance because retail metrics are easy to capture. Correlating page load times with e-commerce conversion rates is trivial. But performance affects other verticals as well. Media, travel, finance, etc. If there are real people using your service online, then those people’s behavior is susceptible to changes wrought by faster or slower web pages. I have yet to encounter a business that, after gathering enough user data and identifying the right metrics, didn’t find a correlation between performance and their business.
How are you regularly testing and measuring the impact of poor web application performance on your business? Do you know how to do this?
I’m absolutely obsessed with making software faster. Just contact me if you have any questions! I read every message 😄