Friday, June 28, 2013

Help Your Self

I was struck recently by a TV advertisement IBM has been running.  A group of managers are sitting around a table facing some sixties songwriters asking them to write a song for them.  When the songwriters ask why the managers reveal that when their customers call them they are faced with interminable waits and a barrage of requests to key in numbers.  In the end the managers admit their customers hate them.  The implication of the advertisement is simple these managers need IBM.  There is of course another solution to this company's problem and that is a highly trained and motivated customer service group, which answers the phone.  Why is this solution seen as so remote that the advertising agency feels you will see the inevitability of using more technology to solve this companies problem, is it simply because this is an IBM advertisement? 


Actually I think some thing else is at work and to understand it I would like to take you to the supermarket.  The basic process of purchasing some thing at the supermarket has changed little up until recently.  A customer selects their purchases then waits in line to pay at a cash register.  The cashier totals up the customer’s purchase takes the customer’s money and returns any change to the customer.  Up until now the technology that has been introduced into this process has been focused on making the cashier more efficient.  Scanners scan the customers selected purchases and total up the customer’s bill, cashiers enter into the cash register the amount of money the customer gives him and the cash register tells the casher the amount of money to give back in change.  Both technologies the scanner and the calculating cash register help the cashier do his job with greater speed and accuracy. 


Recently a new technology has been introduced which does neither of these things, the self-service lane. These self-service lanes consist of a touch screen a scanner, scale and an enclosed conveyer belt.  The idea is that customer checks him self out.  The benefit to the supermarket is that three such lanes can be attended to by one store employee as apposed to the one cashier required for a regular lane, a considerable cost savings in man power. 


The problem is that the customer is now required to do the work of the cashier.  Since ordinary customers are not cashiers the process of totaling up the cost of a purchase and paying for it has become less efficient.  For example to buy a bunch of grapes the customer must navigate through three to four screens before the produce is weighed and can be added to the total.  On the other hand a cashier would key in the code for grapes from memory weigh the grapes in half the time.  Other inefficiencies are introduced by any item or process that does not confirm to the “norm”.  The result is a much less satisfactory experience for the customer who now is required to fulfill the function of the cashier. 


Basic economics would tell you that if customers did not like playing cashier then they would not patronize stores who use these new self-service lanes and stores would either be forced to remove the self-service lanes or lose customer.  But this is not what happens.  The economics of using self-service lanes are so attractive so all supermarkets are installing them and the customer will be left with no choice. 


Both the earlier technology of the scanner and calculating cash register and the new technology of the self-service lane were implemented to reduce cost.  There is one big difference between the two however.  The first two technologies reduced costs by making the cashier more efficient, this does not change the relationship between the customer and the supermarket while it improves the efficiency of the cashier and thus improves the customers shopping experience by reducing check out times.  The new self-serve lane saves costs by shifting work previously done by the cashier on to the customer and makes the process of checking out less efficient and thus degrades the customers shopping experience.


At least one delivery service, which has introduced a self-service web site, has both given the customer a choice and compensated them of their trouble.  This company gives their customers the choice of either calling to place an order and paying one price or use their web site to place and order and paying a reduced price.  I doubt my supermarket will charge me less for checking out my own groceries! 


So what is the point of all of this?  Since the advent of the ATM we have seen technology used to shift work usually done by the company on to the customer.  The advent of the web means that every home with a computer and Internet connection has a potential ATM in it and thus the customer can be expected to do more and more.  We have become so accustom to the idea of the self-service that we have convinced our selves that customers prefer it. The point is customers only want to service them selves when doing so is easer or faster then the vendor to do it for them.  We use ATMs because the lines at the back are to long or the bank’s operating hours to short.  If we believe people are working longer hours and have more pressure on them then ever before can we realy believe they want to take on add tasks such as checking them selves out at the market if it is not going to gain them any time?


While shifting responsibilities on to the customer reduces costs it also reduces the companies control over the customers experience, remember your customers are not your employees.  This is how we end up like those managers in the IBM advertisement sitting around a table asking for a catchy song that will stop our customers from hating us, maybe if we play it in the background while they check them selves out? 


Companies need to be very careful when adopting the “self-service” model and weight the cost savings against the possible backlash from costumers who have been asked to do too much.



Tuesday, June 11, 2013

Software as a Service


The ubiquity of personal computers the Internet and web-based applications is facilitating a fundamental shift in the way business applications are being sold.  The development of web-based applications has decoupled the user interface from the business application, its software and hardware.  As a result a user working at a personal computer with a web browser can access a variety of business applications running on any number of different hardware platforms in any number of different locations for the cost of an Internet connection.  This decoupling is allowing application software companies to change their business model from sales to service.


Currently business application software companies generate the bulk of their revenue by selling their software; as a result the current software business model is cyclical.  Software application companies sell their products into a market and then sell upgrades.  Sales revenue peaks after each upgrade then lags until the next upgrade, creating a cyclical revenue stream.  Cyclical revenue streams create cyclical stock prices and other business complications.


To level out their revenue stream application software companies have been looking for a different business model.  Companies like Microsoft have been switching to a licensing business model in which customers pay a yearly licensing fee to use Microsoft products and are guaranteed upgrades during the licensing period. 


While licensing can level out a software company’s revenue stream it does little to break the upgrade cycle and the postponement of an upgrade can put an upgrade outside of a license period and change the upgrade from free to fee.  Microsoft is facing this problem with its delay of “Longhorn” the next Windows operating system.  


Licensing’s only benefit to the customer is free upgrades.  Since implementing upgrades cost customers money an upgrade must have enough added value to justify its implementation costs.  The new features introduced in an upgrade may be of little or no value to some customers causing them not to upgrade even if the upgrade is free, in which case a license has no value.


Not happy with licensing some software companies are adopting a new business model, enabled by the Internet, in which software is sold as a service.  In the software service business model the application software company hosts its software application and each customer pays a per user fee, usually monthly, to use the software application.  Access to the software application is provided over the Internet through a web browser. 


Like licensing the software service business model provides a steady stream of revenue, in the form of monthly payments.  Unlike licensing the software service business model requires the software company to take on the cost of hosting the application.  The software service business model offers several advantages to both the customer and the software company that offset the higher costs of this model.


By purchasing a software service, as apposed to licensing or purchasing software, the customer has no up front acquisition costs, no hardware or software to buy, and no support staff to hire and train.  The cost of acquisition is reduced to the cost of training employees on the application and converting existing data.  The reduction in these costs allows the software company to charge more for a software service then a license.


The low acquisition cost of software services also makes software services affordable to a broader range of customers.  This affordability gives small companies the opportunity to use the same feature rich software solutions as larger companies and allows software companies to sell their service to a greater number of customers.   


Because the software application company is hosting the application, in a software service, there is only one instance of the software, as apposed to a licensing or sales model where the software is distributed to the customer and is installed on the customers’ computers in a variety of environments out of the control of the software company. 


The existence of only one instance of the software application in the software service business model has several advantages.  With only one instance of the software there is only one hardware and software platform that the software company must support.  This greatly reduces development costs.   One instance also means that the software company can introduce software enhancements one at a time breaking the upgrade cycle and eliminating the cost that cycle generates.  These cost savings help to offset the cost of hosting the application.


The single software instance of the software service business model also makes application integration difficult.  With the sales and licensing business models the customer can modify their local instance of the software application to interface with other legacy applications.  This is not possible with the single instance software service. 


However with the development of web based applications has come a push to develop standard web-based interfaces for applications.  The development of technologies like web services and XML transactions standards are all making trans-vendor application interfacing more commonplace and software service companies will be able to develop standard application interfaces for application integration.  For now the software service business model applies best to applications that can be run in isolation or with limited interfaces.


In addition to integration concerns using a software service prompts several other customer concerns.  According to the Aberdeen Group the chief concern is data security.  This steams from an uneasiness people still have about sending confidential data across the Internet, this concern is starting to wane, as evidence, H&R block filed more tax returns through their Internet site this year then through their store fronts.


The other concerns surrounding software services are: system security, back up and recovery and disaster recovery.  Software companies are addressing these concerns in the same way that application service providers addressed them, with robust equipment, policies and procedures.


In addition to these concerns a company looking at a software service should also ask, what would I do if the software company went out of business?  Will I be able to retrieve my data and how long will it take me to get that data back up and into a new system where I can access it?  The software service is providing both software and hosting so the loss of the provider means the loss of both services.  This creates a single point of failure and as a result a considerable increases in risk.


Despite these concerns the low implementation and support costs of software services make them very attractive to customers.  The benefits of a reliable revenue stream, single instance implementation and the end of the upgrade cycle, makes the software service business model very attractive to software companies.  This win win combination has attracted at least one company to the software service business model. claims to have over 130,000 subscribers and 9,000 companies signed up to use its customer relationship management (CRM) software service.  Companies pay a monthly per user fee to access its web based CRM application. claims that it can accommodate customer specific configurations and that it interfaces to most major Enterprise Resource Planning systems. also claims to be servicing companies with as few as two users up to fortune 500 companies with hundreds of users.   The question is, is the software service business model appropriate for business applications other then CRM?


I think it is and that as issues of Internet security, and trans-vendor interfacing are addressed the number of software companies adopting the software service business model will grow.  And while I do not think software services are appropriate for all of a company’s application software needs I do think they can fill many of those needs.  The benefits to both customers and software companies are just to great for the software service business model not to be successful.


Tuesday, June 4, 2013

Information Technology Centralization Is Back


Companies are “re-centralizing” their IT organizations.  After years of decentralizing IT organizations to align them with business units companies are “re-centralizing” there IT organizations to reduce costs and re-assert corporate control.  The challenge has been to come up with an IT organization structure that while gaining cost reductions and better alignment with corporate objectives, from centralization, does not lose touch with the requirements of individual business units. 


A Functional Approach


The approach to IT centralization can be broken into two types.  The first type centralizes IT organizations by business function rather then business Unit.  Under this scenario, a company with multiple systems performing the same business function, such as accounting, or warehousing, in different business units is centralized into one system, supported by one IT group.  The newly “centralized” IT organization is divided by business functions such as accounting, warehousing, etc. 


This functional type of centralization removes costs by eliminating duplicate systems.  What a functional approach does not do is centralize IT personnel.  In fact under a functional approach the IT staff has simply gone from being divided by business unit to being divided by business function.  The functional approach, by not centralizing the IT department, leaves some redundancy and cost in the IT department.  The functional approach can also produce a myopic IT department devoid of any corporate wide view, thus failing to better align the IT organization with corporate objectives.  The functional approach does not meet the goals of centralization to reduced costs and better align the IT organization with corporate objectives.


The Consulting Model


The second type of centralization organizes the IT department around the IT employees’ skills. Under this version of centralization, IT employees are organized by skill set and assigned to project teams that exist only for the duration of the project.  When the project is completed, the employee goes back to a resource pool to await a new assignment.  This type of centralization produces personnel efficiencies as well as system efficiencies.  The exposure of IT personnel to different projects through out the organization greatly improves the IT organizations corporate wide view.  Thus the consulting model meets the goals of centralization, reducing cost and better alignment of the IT department with corporate objectives.


However the problem with the consultancy or the skills-based type of IT centralization is loss of business unit control, which can have two adverse effects.  First when centralized IT costs are allocated back to business units that do not have any control over IT projects a Boston Tea Party mentality can occur, corporate taxation (cost allocation) without representation (control).  The second effect is a frustration in business unit managers who are not seeing their IT needs being met.  These business unit managers may take matters into their own hands and start funding “covert” systems.  Covert systems are IT systems funded and staffed outside of the IT organization by the business unit.  Once these covert systems are in place and successful they are difficult to eliminate and the company can devolve back into a decentralized IT model.




A highbred approach to centralization can solve the problems of the consulting model while retaining its benefits.  For example the Dominion Corporation of Pittsburgh adopted the skills-based form of centralization.  Lyn McDermid, CIO and senior vice president of IT, placed a director-level IT account manager in each of the company’s business units to keep track of IT needs.  She then grouped employees by expertise, such as mainframe, applications development and services.  The business unit IT director gives the business unit a measure of control over IT priorities.


Similarly Brat Stanco, CIO of the Gartner Group, aligned his centralized IT organization with business unit priorities by appointing a strategic business partner (SBP) from the IT group to each business unit.  Mr. Stanco then established a project-based budgeting system in which the SBPs along with business unit management could approve projects under $100,000 and that were limited to their business unit.  Larger projects go through a three-tier advisory council with business unit representation.  This structure provides the business units with some fiscal control over IT and removes taxation with out representation.




IT centralization can bring with it significant cost reductions both in the elimination of redundant systems and in work force efficiencies as well as better aligning the IT department with corporate goals.   When centralizing an IT department it is crucial to define a role for business units.  A role that will give them some control over the setting of IT priorities and budgets. 



Monday, January 14, 2013

Open Communication

So the working theory is that Consumer IT use is now driving Business IT so what might a day in the life look like if business were to adopt the consumer IT technology. 

You start you day check your company’s social app. to look at what occurred last night.  You see that there was a production problem in the Asian Plant.  You open your e-mail to see if anyone has any questions and your boss wants to know if the production problem will affect this month’s shipment schedule.  You go to the plant site and look at the production page.  You are not sure if the data presented reflects the current outage.  You check to see if anyone from production is available, they are not.  So you post your question on the page.   An hour later you get a notification that someone has responded to your post and that production will be reduced by 10% this month.  You still need to know how that will effect customer shipments.  You receive an instant message from a friend in Customer service who wants to know what the impact of the outage is so you send her the link to your post and ask her what the order backlog for the product is and would she post it to the thread.   As people have begun to check in the thread on the plant website is now growing.  It looks like you have a month’s worth of inventory on hand so the impact to this month’s shipments will be minimal. 

So what advantage did this way of working have over e-mail and phone calls?  E-mail and phone calls generally are point to point conversations.  Web posts can be open to everyone or just a selected group.  They are not ephemeral in that they can be left on line and available for as long as you have space and any one in the group can respond.  This sort of open communication seems to be at the heart of Internet Social life and may be indicative a change in the way people will communicate in business as well. 

Next time what are the business implications of adapting an open communication model?

Tuesday, January 8, 2013

My Head is in the Cloud, but my Job is not

The Cloud I am speaking of is the Internet as in “Cloud Computing” etc.  With FaceBook and LinkedIn our social lives are starting to be lived, at least partly, in the Cloud.  Like many of you I receive most of my news through the Cloud.   I can watch TV on my tablet through the Cloud.  Then why has business been so slow to adopt the Cloud?

For the first time information technology (IT) is being driven by the consumer side of the industry rather than the other way around.  Think about it.  You got a home computer because you used one at work now you want a tablet at work because you use one at home! 

This is a major change in the way IT is evolving and provides some serious challenges for business IT staff.  Social sites are still considered by business as possible marketing tools not internal business tools.  But shouldn’t they be? 
With a Cloud home life more and more people are demanding a fully connected Cloud work life.  The tools to create that work life are coming and some are already here.  Next what a Cloud work life might look like.