Service-Level Targets: One Size Doesn’t Fit All

Hi, from my experience I found that service level on its own is not a good measure of performance or customer satisfaction. The typical service level of 80/20 can be use as an indicator but depending on the nature of the customer query that figure can be right or off. In my last role working for a financial institution wait time had a direct impact on talk time as people were venting if they waited too long before they even started talking about their query. This had a nock on effect on wait time itself as avg talk time increased which in turn further increased wait time. I found they didn't mind waiting up to 20-25 secs max, but anything above that and comments about wait time started to flow in. Cultural aspect is also important as Irish people will react differently to French, German or Italian. In a technical support environment however customers didn't mind waiting up to 60 secs for their call to be answered. So the type of call, the reason for the call and cultural aspect play a role in customer tolerance to wait and consequently to Service Levels. Not to mention the weather, if it is a free or premium number, whether they can call someone else to get their query answered will also play part in their tolerance to wait. But having calls answered within the expected time (varying as discussed above), by a knowledgeable person with professional manners who will answer the query first time in a language understood by the customer is certainly going to impact positively on their satisfaction. It is this combination that will have the biggest impact in my experience but I’m open to other views... Regards, Gilbert
Replies (0)

Hi, Andrea Your key call center observations hit the nail on the head. My experience coincides with yours. I spent ten years optimizing call center operations as a consultant and another decade leading two call centers taking inbound sales, retention, account management and technical support calls. Meeting customer needs has the greatest impact on satisfaction and loyalty, although the amount of wait time that is tolerated does vary by call type. Customers who want to buy have the least patience. Their calls need to be answered fast before they decide to call a competitor. Retention calls need to be similarly handled. Customers who are downgrading or disconnecting their service because of nagging unresolved problems are not happy with your company already. Why add insult to injury by making them wait? From the customer perspective, that wait is reflective of overall company performance. Controlling for other factors, I have done studies that show the odds of keeping customers are 45% higher if retention calls are answered in 30 seconds or less. On the other hand, one to two minute waits for account management and technical support do not adversely impact customer satisfaction, so long as there is no history of chronic problems. The way in which the call is answered and problem resolution are far more important. I have deployed systems that put customers with problem histories ahead in the queue and inform agents how long customers have waited. A genuine and heart-felt apology for the inconvenience of waiting improves customer retention by up to 32%. Service levels are often used to measure performance when in fact they are the least useful metrics. Most telling are employee and customer retention rates. A lack of loyalty severely impacts the bottom line and results in word-of-mouth that keeps potential customers away. An emphasis on speed of answer puts undue pressure on agents and only serves to place quantity ahead of quality. Armando Rancano http://www.armandorancano.me
Replies (0)

Two points - first, TARP has found that with the exception of emergency services and flight operations, 80/20 is almost never cost justified - more like 80/45 or 90/60 makes sense if when you answer the hone you fully can answer the question. Secondly, rather than investing in incremental CSRs an investment in one preventive anlyst asking "Which of all these calls can be prevented via better custoemr education and welcome packages?" often allows you to prevent 5-10% of total inbound calls - a much better investment.
Replies (0)

Andrea: I have to dissent on targets . . . in general. We have found them to be part of a broader thinking problem about the design and management of work. The problem is that we look at work from a cost and productivity viewpoint (in contact centers) and from the view three ?s are asked (1) how much work do we have? (2) How long will it take to do it? (AHT) and (3) what service level do we want to provide? These questions help us plan, but do not help us improve the work. We have found that these measures look at the wrong problem. By not looking at all demand as work, we can look to failure demand (demand caused by a failure to do something or do something right for a customer). This demand makes up 25-75% of most contact center demands . . . it is an opportunity for improvement and is systemic in nature. The use of targets is always cause for alarm and we have found creates waste. Targets become the defacto purpose of the CSR, when the real purpose is to serve the customer. Regards, Tripp Babbitt
Replies (0)