Is Skype a viable option for large call centers?

First let me apologize for the long answer here but this is a slightly loaded question.There are a number of issues around Skype and its use in a contact center.

First, some consideration should be given to the type of center you're working with. Is it an inbound center, an outbound center or both? Is it primarily servicing a B2C or a B2B market? Is it a single purpose (e.g. an "in-house" center) or an outsourcing service provider acting as a center for other businesses? All of these weigh into whether a center should consider Skype as a primary telephony/chat/collaboration platform.

I think the least important issue is audio quality (QoS), although that is certainly something you should consider and weigh against your target calling demographic. The user experience (for the caller, the agent and the management team) is about more than simply the audio quality of the call; a truly integrated solution can, and should, minimize the pain for the all of these parties and so the ability to provide clean integration interfaces, consolidated reporting and good workforce management solutions really should carry the most weight in this decision.

I encourage contact centers to consider the “golden rule” of telephone customer service: never ask for information twice or transfer the caller more than once. A prime example of a bad caller experience is calling to a company being answered by an IVR providing the requested information, and after reaching a live agent having to provide that information a second time. That experience is made worse when the agent then needs to transfer the caller because they cannot address their needs. Even more frustrating is when the needed resource is on a completely separate platform and the agent can’t make the transfer. Integrated solutions can minimize these issues: the IVR cleanly passes the information to the agent where integrated “screen-pop” solutions provide the agent with all information needed to handle the call based upon the IVR input. As you start to “cobble together” multi-vendor solutions, particularly where industry standards are not followed (Skype uses a proprietary call control and a proprietary peer-to-peer VoIP codec) this becomes much harder to accomplish.

Further, as staffing levels and call traffic increase within a contact center the need for consolidate reporting of all interactions whether handled completely within the IVR or handled by an inbound/outbound live agent become critical to success. Adding outbound campaigns and multi-media interaction channels (email, chat, video, etc.) into the mix makes getting a complete picture of the contact center even more difficult. A contact center solution built around a single integrated unified communications infrastructure makes this much easier.

Lastly, large centers need to think about whom their agent’s are what skill sets they have and with what technologies they work best in order to properly address their staffing levels. And they must ensure that once those schedules are built they are adhered to. Think about how hard that becomes when you cannot get a true picture of the caller experience end-to-end or, get information from one of your platforms. An integrated solution from a minimal number of providers and which provides a rich integration environment ensures that your workforce management software can take all of these issues into account and properly built a schedule.

These issues are less important for small call centers and in fact are often ignored. But, as the staffing levels and call traffic increase they become a requirement. This need really starts in a "mid-sized" contact center (75-250 agents) for a "large" contact center (200+ agents and or distributed locations) it is not a "nice-to-have" it is a “must-have”.

Integrating a technology like Skype into your center as a proof-of-concept or to address a specific demographic makes a great deal of sense but I would caution against building an entire center around it until you have though through how you will address the issues I’ve outlined.

! thing to add - distortion, jitter, echo, and other call-degrading factors are at least ADDITIVE, and sometimes MULTIPLIERS of each other. Start with a cell phone + noise around the caller in her car. Now add unmanaged (non-QOS) trunking- also with heavy compression. Ugh. This is a recipe for frustration for caller and agent alike. Jim Graham's eloquent analysis above, and golden rule example are right on. If you're setting up a call center for the complaint department at the IRS or Cash4Clunkers - go right ahead with Skype, or Vonage, etc. But if you care about intelligibility, fidelity, and ease-of-flow of the call, and the customer's perception of your company ... stick with cleaner trunking methods.

PS - In an RFP presentation we made recently, we highlighted times when with ESL or a hearing or mentally-impaired population, VOIP could be of concern.

A woman from the evaluation panel stood up and said "I was involved in selecting a state-wide call center system, and Cisco (who had equipped a similar center in Texas) was a prime contender. We had lots of comments from disability rights advocates in Texas wanting to be sure that we didn't use VOIP in the California call center - because they had such a terrible time with intelligiblity in Texas." And this was with "real" trunking, not Skype. There was still enough aliasing, packet loss, smoothing, etc. that those with hearing problems (or hearing aids) had very challenging times understanding the call center reps.

I think all of the above have merit certainly from a technical and quality perspective, however the economical view has been ignored.

Skype is not cheap compared to other solutions. Forgetting the rates which in some cases are very expensive, fundamentally they charge in 60 second increments.

This alone multiplies the operators expenses by 4-10 fold.

Call centres are about driving costs out of the system (which is why we use them) and therefore on the call costs alone Skype is not a viable service not considering the above mentioned quality and performance issues.

Using an outsourced IP Hosted solution would provide a huge cost saving, while ensuring SLAs (try calling Skype for support) are in place to keep the majority of the above issues under control (in the hands of the technicians whose job it is not just for you but many other companies). This obviates the need to own your own infrastructure and is expandable as your business grows.

There are many organisations that provide these services that are by far, more flexible, less expensive with higher quality than the offering provided by Skype.

Skype is great for small business and personal use, however the costs are still high compared to other solutions AND you are tied to your PC or computer network.

To put is simply... PSTN-There is no substitute.

Having said that, I tend to look at this from a different angle. For the immediate future copper is king and as long as you base your Telephony from that starting point I see no reason not to integrate Skype on top of PSTN connectivity. The biggest hurdle is account control... as in whom maintains the account as employee's come and go. I just concluded the Skype for Asterisk beta and the emerging solution of Skype for Asterisk combined with the Skype Business Portal allows for that kind of control and merged connectivity. To be fair, Skype and FreeSWITCH can be paired the same way in that the agent will be answering the same handset regardless of where the call came from.

Run any Telephony purely on Skype? Not on my watch. Add Skype connectivity and provide customers with an additional ( potentially very convenient ) way of connecting? You bet. As long as it is tiered on top of current best practice.

Incidentally, there is a reason "Open Source" is gaining ground rapidly in Telephony. It is more advanced, offers greater flexibility and can grow and adapt over time.

That and even the "Best" of so called "Traditional" Telephony systems have down time and they are absolutely horrible to manage.

Call centers aren't cheap. The reason they aren't cheap is because customers expect a high level of service and that level of service comes at a price. With large call centers spending millions on software and hardware to ensure callers have the best experience possible, why would they leave the call path to chance?

Skype will not ensure QoS, so the calls are more likely to suffer from jitter, delay, and echo. From a technology standpoint, with many call centers using voice recognition for self-service, that delay, jitter, or echo could re-route them to a different department or further frustrate the caller if they have to repeat themselves several times.

Skype is a viable option in several instances, but call centers, especially large ones, are not one of them. When a caller calls in to a call center, they are either looking to spend money or solve a problem. In either instance, the company they're calling wants to look as competent and professional as possible.

If they're looking to spend money, they don't want repeat themselves due to poor line quality or worse yet, be disconnected because of latency. The caller will feel as if the company doesn't want their money or isn't reliable enough to do the job.

If the caller is calling to solve a problem (delayed shipment, wrong product, broken widget, etc.) and experiences ANY problem on the call, the customer is going to be even more upset which means more calls get escalated.

My initial knee-jerk reaction to this question is to say, "Not yet." However, there are solutions in development that may ease the transition. For example, FreeSWITCH (www.freeswitch.org) has a free, open-source software solution that allows inbound and outbound Skype calls, and it can handle transcoding to/from SIP or TDM for interfacing with existing PBX equipment. Companies wanting to experiment can always set up a simple system and notify potential callers that they have a Skype interface. Naturally, the IT team will need to be ready to monitor Internet traffic levels to make sure that Skype calls don't overload the system. In the end, though, I'd have to say that it is at least possible to do very low cost proof-of-concept testing.

For those who want to know more about the way FreeSWITCH handles Skype channels feel free to visit us on IRC: #freeswitch at irc.freenode.net. Community member Giovanni Maruzzelli is the main developer of the Skype interface.

-Michael

That's an easy one, no. For Company's call centres that already use Microsoft Exchange Server, or even those that don't, look at a Microsoft Office Communicator product instead, for example Interoute's "Interoute One" (www.interouteone.com).

@Bill Moore

Your points are well taken. However, those points apply to the quality of PSTN calls in general, and mobile phone calls in particular. In fact, VoIP in general suffers from relatively poor call quality because everyone seems to want to save bandwidth, so they use low-quality codecs like G.729. (By low quality I mean the quality of the voice, not the quality of the effort that went into creating the codec.) Any customer who calls into a CC on a cell phone is already expecting a low-quality connection. Similarly, any customer savvy enough to use Skype is going to be familiar with its ups and downs.

I say give it a shot and see what happens. A POC test is not that difficult. Having empirical evidence is certainly preferable to having the opinions of us "experts." :)

-MC

P.S. - I have a low tolerance for PSTN, GSM, G.729, etc. because I work on a conference call all day long and most of us connect with 32k Polycom Siren codec. When someone calls in on a narrow band connection the difference is quality is obvious.

All the previous answers have merit. However, to best answer this loaded question, then one has to ask the question: "Why would a Large Call Center want to gamble on losing an account because of their desire to cheapen their VoIP technology by using a platform that is clearly not equipped to handle the level of inbound or outbound traffic of their operation?"

Let's face it, Skype is not and nor was it ever intended to for use in a contact center. In fact, without some measurable level of QoS Skype is barely tolerable for individual usage as a communication medium.

All the previous answers have merit. However, to best answer this loaded question, then one has to ask the question: "Why would a Large Call Center want to gamble on losing an account because of their desire to cheapen their VoIP technology by using a platform that is clearly not equipped to handle the level of inbound or outbound traffic of their operation?"

Let's face it, Skype is not and nor was it ever intended to for use in a contact center. In fact, without some measurable level of QoS Skype is barely tolerable for individual usage as a communication medium.

Technologically, there are various solutions for interfacing Skype to other telecom equipment, so the integration is certainly possible. You might even make the argument that offering a Skype option to certain demographics of customers could be a significant value-add. However, as several others have already mentioned, Skype--by its very design--is a "best efforts" technology. And while that may be tolerable or even OK for two private individuals making a free call "over the pond", it is not a suitable technology for a business providing call center services to their customer.

I look at it this way: The cost of Tier-1 telephony (whether it be traditional TDM or SIP/VoIP is probably one of the smallest cost components affecting a call center's total operating cost today (agent labor alone is many times the fully-loaded cost of the telephony), and just replacing their Tier-1 carrier with Skype for transport doesn't eliminate the cost of the ACD, IVR, call recording, etc. In the end, even for that "special" demographic that might find Skype appealing, the risk is too high for lower customer satisfaction or loosing customers altogether just to save a few dollars.

PS - Someone mentioned that maybe if the callers were employees (who would presumably be a captive audience and be more forgiving, although maybe not by choice), then Skype would be a viable option. But here too, I think it may be the proverbial "penny wise, pound foolish" mistake. After all, if you have a call center providing services to your employees, you are paying the salaries of both parties on the line--making dropped calls, misunderstood communications, and other call quality issues doubly frustrating and expensive.

Skype is a piece of software which can be used with certain hardware to enhance it's functionality. If millions of users around the world use skype everyday to communicate with each other which involves placing millions of calls, why not one large contact center process all those millions of calls instead. Skype is certainly an option that needs to be looked into in the contact center but it's certainly not the only player.

Certain factors that Skype would need to take into consideration before it can be deployed in a Large Contact Center.
1. How Large is Large? - Agent/Customer base
2. Does the Contact Center use their own customized softphone to perform telephony functions.
3. How reliable is skype versus what the contact center uses today?.
4. Would Skype be ready to enter into a vendor certification process?.(with say Avaya, Cisco, Genesys,Aspect,Nortel...depending on what the contact center uses to be able to integrate with their APIs)
5. How secure is Skype in a Contact Center?.
6. How much would Skype save a contact center versus using their own infrastructure to perform telephony functions?. Need a business case.

Skype can be a component of your call center that will likely help you reduce costs and enable better customer collaboration. Skype alone, though, likely is not the answer. There are examples of companies (see Gerry Blackwell's, "The Call Center of the Future?" http://www.voipplanet.com/solutions/article.php/3822166) that use Skype successfully, but they have other software like OnState's to enhance its capabilities into an enterprise solution.

Shawn

Many good answers and discussion.

Aside from functionality and quality of service. The single most important requirement in a large call center is REPORTING. Without a breadth of reporting tools a call center business is flying blind and heading in a direction that will put them out of business.

Possibly an angle has been missed that should be considered-
Is not the whole idea of a call center to be an accessible and easily accessed solution to the caller for dial in as well as dial out. As VoIP awareness grows more and more callers are switching to different options for their initiated calls as a cost saving excersize, moving away from the traditional PSTN CPE. For these users whom originate their calls over Skype and other VoIP carriers would it not be a wise choice to offer them both a PSTN number and a VoIP number?

If the user has already made the choice to use the low cost telephony technologies then as a rule they will already be aware of the potential pitfalls of call quality.

Surely adopting the approach of enabling the traditional PSTN interface in parallel with the VoIP / Skype interface as a choice to the dialler would enable the call center to manage a migration to the low cost technologies as the quality improves whilst retaining the ability as a fall back to the traditional solution. Whilst the initial hardware cost may higher the cost saving of the Skype solution when in use should negate this after a period of usage, coupled with the ability to fail over to generic PSTN.

It is also a known phenomena that some countries are best called over SIP communications to gain a higher quality call ( depending on the site of the call center ) and some regions actively ban SIP communications ( some parts of the middle East etc ) so this would not only advocate by require the retention of PSTN.

A last consideration that I would put forward would be the operators in the call center. Currently the trend is to have them sited in the call center its self. Using Skype and associated SIP technologies these operators could be working from home relatively cheaply and from a wide geographical area – Removing the need for travel which could introduce the ‘green’ argument!
The codecs used as discussed earlier can be G726 and even G723 and if the hardware used is of a high enough quality from experience you should be able to attain the same quality as that achieved with the G711 comms, how ever these solutions tend to be point to point excluding the use of Skype ETC

Working for a VOIP provider might make me bias, but I will try to keep from going to the dark side. If I was going to open a call center, I would never put all of my dialtone eggs in one basket. We are all familiar with the concept of least cost routing, using a combination of VOIP and TDM you can actually create a Highest QoS Routing Plan. Voip issues, when they happen, come in waves or have a single causality. Having a TDM alternate can allow you to ramp up or fail over as needed, it will be a combination of IP based and traditional Dial tone. Obviously you get different inbound vs outbound calling ability and the need to select a PBX/CallManager that can work with both. But this solution can cover, quality, reliability, disaster recovery and future upgrade needs.

Since call quality is the life blood of most, if not all call centers, any VoIP/SIP solution that does not provide Quality of Service is best avoided. There are a number of true business-grade solutions that can do the job, but generally speaking the "free" and residential focused services are not among them.

Simple answer..."NO".

Look for a solution with a proven track record in Call Center technology and with tons of available, expert support - not Skype. Your business is counting on you.

I feel Skype does have its place in a large call center. Not as the primary service but definitely as a back up to POTS, T1 and other Voice connections. I have also seen companies that force internal traffic between branch offices to the Skype service to save dollars. Is the connection perfect every time, no, it is an internet service. Is the quality the absolute best? No, it is an internet service. Is it good enough to pass internal questions and messages? The railroads initially provided "Hoot and Holler" service across lines that ran the length of their tracks. Poor connection, poor sound levels, sometimes blown down but 90% of the time they did the job. Customers were never sent over the facilities. That was back when Long Distance was $.50 cents a minute.

At the risk of beating this dead horse- Is quality of service important? Do you need to integrate the service with the existing PSTN? Is it for internal or external use? Comapnies have been using things like instant messaging for a long time either internally or externally. The common thread was expectation. Nobody expected much and they got what they expected. Skype could be used the same way but is very limited and does not provide the robust flexibility of traditional and new call center platforms. Proceed with extreme caution.

So here finally a question in my specialty area. The ANSWER IS NO! You cannot transmit enough of voice sessions via skype, there are limitations to QOS. If you want cheap wholesale LD get a PRI, no matter if its DS1-OC3 PRi Services. PRI Stands for Primary Rate Interface. You can deliver it over an open source telephony platform like Asterisk, I personally design contact centers for the fortune 100 and I can tell you that skype will never be ready. There is too much middle ware crap involved on there suite, they are insecure and can ultimately result in high jacked CC numbers, SS Numbers and that is only the beginning of your night mare. For home sure it's cool. But for a contact center it isn't worth it. XO probably the worst carrier on the planet has wholesale LD In huge buckets for next to nothing. Look that way, or TW Telecom, AT&T ,QWest or find a local lec.

Best of luck but to quote office space "That is a horrible idea" when asked about the jump to conclusions board.

NO. Skype is a cosumer propriatary vehicle software that allowstallking head video conferencing point to point for free. There is no capability of doing multicast, or content in the H.320 or H.323 platforms. The QoS and limitations have been well documented with the numerous responses.

I'll give you by un-scientific answer: No.

I am basing that on two simple facts:
1. I spent 8 years with Avaya (the #1 market share leader in Contact Centers). Contact Centers are the life-blood of any business. Mission critical nature requires solid, trusted and highly available systems. Sky, as much as I loe and use it every day, is nowhere near ready for primetime.

2. I sue Skype daily. I tolerate a bit of delay and a couple of munched up words here and there. An irrate customer who is has just about had it with something...is NOT going to tolerate that!

Stick VoIP-enabled Call Center solutions from industry leaders.

QoS, Features, Value, Type of CallCenter, etc. All are important questions to be considered and have been mentioned in the previous comments.

Is VoIP ready for the CallCenter ---- Yes. Is Skype ready for the CallCenter ---- No. However, there is a VoIP technology that is ready, such as SIP. Remember, Call Centers are REALLY CONTACT centers. (not just calls). It goes beyond Voice. You need other user agents that can easily morph their communication device between presence, voice, chat/IM, email, video, etc. And, this needs to happen within your own corporate network (not world-wide internet). Skype doesn't play there yet.

It’s not a question of whether IP telephony will overtake traditional calls, but when it will do so. VoIP telephony is now cheaper, its sound quality matches that of ordinary telephony, or is even better, and a VoIP user ID is as easy to use as a standard phone number. SIP and Skype are the two technologies fighting the battle for VoIP supremacy. They achieve the same goal: to route multimedia conversations over IP-based networks. But their approaches to do so are radically different.

To put it simply: Skype should be used for the Consumer, and SIP for Corporate usage.

SIP has standards, many developers write to it for additional functionality, QoS and Quality Monitoring and Control are possible, Service providers use it, provides same value as Skype with more value in the corporate world.

Skype is a proprietary system; we don’t know exactly how it works. The evolution of SIP is in the public domain. Skype alone controls Skype’s evolution.The Skype Network is great for Consumers. The technology allows the peer-to-peer network to grow automatically by jumping on nodes and supernodes.

Skype has one login server that supports all registered Skype users worldwide whereas each SIP VoIP provider gives a registration server for its own VoIP subscribers. Because SIP is an open standard, SIP users can communicate with other SIP users who subscribe to a different VoIP provider. An AT&T user can for instance connect to a Redband AB user seamlessly. Skype users communicate with other members of the Skype community.

SIP user agents connect to the registration server login to retrieve the information necessary to contact other SIP users. Call control is conducted between user agents. Skype uses a peer-to-peer overlay network.

Therefore, I would think that SIP can be more secure than Skype? (I may be speaking out of turn here, I do not know the encryption/compression or security measures)

But, based on this, SIP clearly positions itself for playing better in the corporate world. Especially with IP V6 gets here.

My vote: Contact Center = SIP; Consumer Applications = SKYPE

Skype is consumer solution with no support at all and various quality over time. You would not want to use that as 'making telephone calls' is your business and which your call center makes money with.....

So choose a solid VoIP solution with Call Center solutions brought by professionals. And that can also mean using Open Source like Vicidial or BasiCall both based on Asterisk (we both have experience with this). It all depends on the scalability.

Small call centers with limited budget could use open source (as long it is being installed + maintained by a company that knows how to do this) while larger organizations with different scopes might use Microsoft Office Communications Server R2 (which we can also interconnect by the way to our VoIP service by a SIP trunk).

But do not use Skype I suggest.

More information? Take a look at http://www.xeloq.com

Regards and have a nice weekend!

Milko van den Wollenberg
XeloQ Communications Internet Telephony

--