Labels: agent desktop, call center, contact centre, contact centre agents, enterprise mashup, mashup

Dale Vile, from Freeform Dynamics, has written an interesting article discussing whether packaged applications are becoming less relevant:

Whether it’s SOA purists telling us that we’ll all be self-assembling solutions from components, enthusiasts of modern development environments wanting to build everything from scratch, or the SaaS evangelists saying it’s all going to go into the cloud anyway, it is trendy to dismiss application software packages as being out of touch with the needs of the 21st century….but the truth is that both vendors and their customers can only move so quickly, and while the latest incarnations of ERP, CRM and other packaged applications promise a lot, migrations and new implementations consume both time and resources.

It almost seems a response to a very frank article from ZapThink advocating the move away from packaged applications to self-implemented SOA solutions.

I agree with Dale - I don’t see companies in the current economic environment looking to rip and replace their current applications and systems. If they are, then they are likely to be looking at SAAs, but as Ray Wang emphasizes, this does not remove the integration issue:

Rapid SaaS Adoption Will Lead To A Repeat Of 1990’s Best Of Breed Integration Challenges’ and ‘Given these scenarios, CIO’s and line of business apps will need to rely on stronger enterprise architecture and integration in hybrid deployments. In fact, au contraire on the death of SOA!

So what everyone seems to agree on is the increased need for integration. However, resolving the integration challenge is far from being fully addressed today. A big gap still remains: integrating applications for people. Contact centres are a good example of this, but I would venture to say that most people involved in supporting key processes end up having to use multiple applications to do their jobs. SOA technologies do not yet address this problem: user specific solutions end up having to be custom coded, in effect ending up building new silos and legacy applications, destroying the point of establishing a flexible and re-usable architecture in the first place.

To address this issue, SOA technologies and approaches need to be extended to the user interface and user interactions. This needs to happen in ways that suit human centric processes, where change and speed of adaptations are key. Current SOA programmes are generally bottom up, complex and relatively slow to get off the ground. Much more agile, fleet of foot solutions to integration need to be provided. Solutions that can start delivering value in days but without building new legacy.

This is where enterprise mashups have a major role to play. They provide not only an approach to ‘clear out their application integration backlog more quickly and at less cost’ (Anthony Bradley), they also provide a route to delivering more effectively a modern, flexible architecture. How? By extending the SOA concepts of composition of services all the way to the User Interface. In other words they provide the capabilities to turn valuable UI (from legacy applications, modern solutions, SAAS applications) into re-usable components that can easily, quickly and cost effectively be assembled into user specific solutions. Whether your implementing new SAAS applications or need to evolve from legacy systems, enterprise mashups provide the answer to integrating applications for people .

Interesting research from Genesys has just been published and is given a good write up in MyCustomer.com by Stuart Lauchlan. First, I must say they deserve ‘la Palme d’Or’ for the highest number one can find in a customer service press release: $338.5bn! Having said that, I think it is interesting to quantify the actual costs resulting from underperforming customer service in this way. Where do the issues stem from?

"The biggest problems that customers experience and which undermines the customer satisfaction levels are familiar ones. They include being trapped in automated self-service, having to wait too long for service, having to repeat themselves and having to talk to company representative who lack the skill to deal with their problems. On the other hand, most consumers surveyed argued that they would be most satisfied when dealing with a competent, live customer service rep."

Delivering a positive customer experience, enabling customers to engage with a competent, live customer service rep is actually very difficult. It requires providing the agent with systems that not just deliver the right information at the right time but also guides him or her through the steps needed to resolve the customer query. This is the only way to allow the agent to focus on the customer interaction and ensure satisfaction. The most critical aspect is an effective user interface that prompts the agent in an intuitive way through each step required to complete the task and resolve the query. What is the main obstacle to achieve this? The multiplicity and complexity of systems that agents are forced to use.

The other interesting point here is that customers prefer to talk to a live person, to a human being, to resolve their queries beyond the most simple tasks. This reinforces our belief that contact centers are and will continue to be essential in maximizing customer value. Trying to resolve performance issues by automating away the contact centre does not seem to be the right answer. Providing agents with the right tools that make their jobs easier is.

Anthony Bradley from Gartner provided some interesting views on how he sees the enterprise mashup market developing this week in his blog posting: Enterprise Mashups in Major Transition.

The transition he has identified is chiefly concerned with a shift in the use cases he is seeing that drive demand. As Anthony says:

Just about everyone, including me was talking about enterprise mashups as a new paradigm for end user computing where business users would rapidly assemble and reassemble applications in a highly dynamic fashion.

And then…

… a shift in the need began. Instead of asking about end user empowerment, clients began inquiring about how to reduce integration costs with mashups. Some organizations were trying to use mashups to clear out their application integration backlog more quickly and at less cost. Others were facing new integration challenges due to mostly unexpected mergers and acquisitions.

This accords with what many of Corizon’s customers and partners are looking to enterprise mashups to provide in the first instance. I would add that this shift does not move the interest in enterprise mashups purely to the IT domain. Fixing integration problems does not only concern IT but also business unit leaders in that they have the potential to greatly impact the efficiency of their teams. In today’s environment, many businesses are looking for way to deliver a step up in people productivity with short and low risks projects: enterprise mashups are perfectly suited to that.

Finally, this type of pragmatic integrations will naturally create a pool of reusable mashable components if done with the right technology. Once created (and paid for) thanks to these projects, these components will make it easier for end users to create their own mashups in a safe way, with the buy in from IT. So the nirvana of end user empowerment is perhaps not as distant as Anthony fears!

Labels: application development, enterprise mashup, Lean Software, mashup, mashup patterns, process mashup

This posting is the second part of my colleague Toby Baker's observations based on his work with contact centres and their desktop integration problems.

My last blog asked why contact centre desktops are so broken, and tried to explain why. This one suggests what we could do about it.

Let’s re-cap – the main problems are:

• There are too many apps on the desktop. The average is 6, I have seen 60.
• The data is all over the place – across multiple screens and apps.
• No standard User Processes – it’s left up to the user to figure out how to make it all work.

So why is this a problem?
So many reasons! Here are the main ones:

• Complex ALT-TAB, COPY-PASTE operations means longer call times, longer hold times, higher abandon rate
• Lots of applications means lots of training
• Lots of double-entry means lots of errors
• Hard to find data means lower first call resolution
• Lack of standard process means compliance is more challenging
• System-centricity, rather than customer-centricity means lower customer satisfaction.

So what can we do about it?

1 - Look at the user processes you have, from a customer and agent point of view, and see where the bottlenecks are.

• How much ALT-TAB and COPY-PASTE do you see?
• How many screens do your agents have to go through?
• How many logins do your agents have to remember?

Imagine if all this complexity could be removed – how much time could be saved? What would be the benefit of automating and optimizing this process?

2 - Decide which user process to improve first, and figure out what the ideal process is. Create a roadmap towards that ideal process, creating early fixes that create payback fast. Involve your users, and start building a business case:

• What could you do with the call handling time that you save?
• What would it mean if your new agents became fully competent sooner?
• What would happen to your abandon rate, FCR or PCA30 if you did this?

3 - Split the problem into two halves: The user processes and the applications underneath. User processes change at a different rate from system interfaces, and you need to separate the two to deliver fit for purpose interfaces to your agents. Design the desired user processes working with analysts and business, then work with it to turn the relevant “bits” of applications into the building blocks of the solution.

• Add to your business case the benefit of being able to change screen flow and therefore user processes quickly
• Imagine how much benefit there is in re-using the everything you build in subsequent projects

4 - Start a project to build a model office, and start proving your business case. See what can be achieved quickly when you implement those quick fixes on the way to the optimized User Process.

Follow these suggestions, and the problems caused by broken contact centre desktops will start to disappear very quickly!

Labels: agent desktop, CRM application, CTI