Your call center software might come with all the bells and whistles under the sun, but if it doesn’t work when you need it to, those features aren’t worth the cost of the proverbial package they came in. Reliability is paramount to delivering consistently excellent customer experiences, so it’s a top factor you’ll want to weigh when choosing a contact center solution.
In a recent LiveVox survey, we asked contact center leaders about the most important things they consider when they’re in the market for a new software platform. Eighty-two percent of respondents ranked system uptime as ‘important’ or ‘very important.’ It was second only to customer experience in terms of what decision-makers value.
Reliability, security, and uptime are all nice-sounding buzzwords you’ll see mentioned frequently in promotional materials for cloud contact center software. But what exactly do those terms really mean and how can you practically measure them when shopping for potential vendors? Here, we’ll dive deeper into three of the key reliability criteria for contact centers and explain how to evaluate them in your contact center software search.
How to Evaluate Cloud-Based Software Vendors
1. System uptime statistics
System downtime can cost a business thousands of dollars per minute. Those costs come in the form of lost revenue, missed service opportunities, and damage to your reputation in the eyes of customers, whose trust in you may falter if they can’t reach you when they need to. This is not to mention the post-downtime costs associated with bringing everything back online, such as labor and the cost of system repairs.
Your vendor service level agreement, or SLA, should explicitly state the uptime, packet delivery, and latency you can expect, with uptime being the most important of the three. 99.99% uptime is the industry gold standard, and the number certainly sounds reliable. But how does it break down in terms of actual minutes on the clock?
To understand, we’ll need to do some simple math. Network uptime is measured as follows:
Number of hours your network is up and running per year divided by the number of total hours in a year. (8,760) = Annual uptime percentage
So, if your network is unexpectedly down for two hours every year, your equation looks like this:
8,758 ÷ 8,760 = 99.97% uptime
- 99.99% uptime translates to roughly 30 minutes of unplanned downtime every year.
- 99.999% means the system is unexpectedly down for less than ten minutes every year.
While there’s only a 20-minute difference between the two, that extra nine can come with a significantly higher price tag, so you’ll need to weigh the added expense against the cost of a potentially longer outage window.
It’s worth noting that outages happen to nearly every business. So in addition to asking about uptime figures, it’s important to take things a step further and learn what contributes to downtime. What are the most frequent causes of crashes and what measures are in place to prevent outages?
For example, power outages can cause hardware to crash. Having a reliable backup power supply would be a feasible preventative solution. A contact center software that uses cloud architecture with geographic redundancy will also minimize the impact of location-specific incidents.
So far, we’ve been discussing unplanned downtime, but we must also consider planned outages, which are a routine part of the software business but which aren’t included in uptime calculations. System maintenance and upgrades, for example, may result in several hours of planned downtime every year.
Ask your vendor how frequently planned outages occur, when they’re typically scheduled, and what happens when they overrun the expected window (like when maintenance takes longer than anticipated).
Remember, you make promises to your customers, and any downtime you encounter might cause you to break those promises. Choosing contact center technologies with maximum uptime translates into maximum customer satisfaction and profitability.
2. Is support offered during peak call volume?
Call center volumes can change on a dime, affected by everything from service issues to seasonality. Is the help desk software you’re considering able to respond in real-time to call surges without compromising on the customer experience? This is another critical measure of reliability.
High-volume call periods are stressful enough for agents as it is. If your software lags or a disproportionate number of calls are dropped when volume peaks, it only adds to their frustration, which can come across in their tone of voice and overall agent performance. This negatively impacts customer interactions. Thus, choosing an agile contact center platform that can scale up or down in tandem with your changing needs is key. Don’t forget to investigate how those changes affect your costs.
To keep tabs on how well you’re able to manage service spikes, choose a contact center software that offers 24/7 system monitoring. You’ll want this monitoring to include metrics like drop rate, busy lines, call availability, service level, the average speed of answer, and average handle time—at a minimum. You should then be able to easily translate this tracking into comprehensive reports that make it easy to see if and when bottlenecks are occurring.
Talk with your prospective software vendor about your support options during periods of peak volume. What if call spikes occur overnight or on weekends—will support personnel be available to respond to any issues?
Anyone can get customer service right on a slow day. You want a call center solution that allows you to thrive even in your busiest, most urgent moments of need.
3. What type of call quality is offered?
Finally, let’s discuss the measure of reliability that’s at the forefront in the eyes of customers: call quality. This is a measure by which you’ll be judged on every single call, so there’s no room for less-than-stellar call conditions.
Technically speaking, the path between a customer and an agent is a complex one, with many different links in the chain and potential failure points. It only becomes more complex if agents are working from home, where things like individual ISPs and varying connection strengths come into play and make maintaining consistent quality more of a challenge.
Mean Opinion Scoring
Ask your prospective call center software vendor about the measures that are in place to track call quality, like mean opinion score (MOS). MOS is a subjective score given by human testers, and it’s the industry standard for measuring the quality of VoIP calls.
MOS accounts for:
- Packet loss
- And codecs.
An MOS score between 4.3 and 5 is the preferred standard.
What “Technical Debt” Are they Carrying?
Ask also about any known issues that can contribute to subpar call quality, like misconfigured equipment, lack of bandwidth, or issues stemming from third-party providers like the phone company.
If such issues exist, what’s being done to remediate them?
As with managing call volume, managing call quality takes consistent, diligent monitoring. So it’s crucial to choose a cloud contact center software provider that prioritizes reporting and shares your commitment to ongoing service improvement.
In the modern contact center, reliability matters. If you want to deliver best-in-class customer interactions, spend time upfront evaluating contact center software vendors that have the proper measures in place to allow your customers to reach you over voice, email, text, and social media, every time they need to.