Friday, February 26, 2010

Opening a business bank account takes a week. So what!

A survey by Finextra last year reports that:
For the majority of corporates, it takes more than one week to open a new bank account, according to a survey Finextra Research conducted over the summer. For this and other reasons, 44% of corporates say they would switch banks to get better service, standardisation and automation through electronic bank account management (eBAM) processes.
OK, so a week is a long time to perform such an important transaction, but really, do the survey's sponsors believe that the time it takes to open an account really is a driver for customers to switch banks? Once your in, your in. The time it takes for every other institution to open a new account is probably a block to you actually moving on. It seems to me that the headline is trying to push the point a little hard here.

When I opened my new business account (admittedly a small business not a mega-corporation, but I bet they are a large volume of what banks are dealing with), Bank of America had the account open in about 60 minutes. It was a painful process to watch, but it worked. From my blog about the experience:
The issue for me is that the systems that agents must use to set up accounts appear not to be at all customized to the type of account or the needs of the branch agent. It appears to be true that agents and brokers in financial institutions are limited in the types of accounts they can open less by the skills they have selling an appropriate product to a customer or being licensed to sell what is available, but in fact in the amount of training they have in the account opening systems. In all, it took me an hour to open the account. The agent was friendly and helpful, but how few customers can she help in a day if the systems force her to work at that rate?

The survey from Finextra doesn't reflect my experience, even if I do believe that many organizations could do a lot better at account opening.

A post from the Improving It blog

Let us help you improve your business today. Visit

Thursday, February 25, 2010

Lean methodologies and employee development

I've been chatting with some of the experts in the field of Lean methodologies, leadership development and change mangement, to understand the relationship between Lean & Agile methodologies and employee development. I'm trying to understand if there truly is a relationship between Lean, the development of software to improve processes with agile methodologies such as Scrum, and enabling people to perform (as individuals, managers and in teams) better through employee development, or if the the only training and experience Lean wants to give employees is how to implement Lean better.

[Feb 25, 2010 - I just added a related video at the end of this post - short and to the point. Thanks to Mike Leamon for this.]

As I look at various sources of information on Lean, especially comparing it to the Toyota Production System, it seems to me that despite all the talk about the importance of teamwork, respect for opinions, and development of specialists, the process improvement methodologies actually ignore a large part of what can make organizations better - employee development. In so many examples I have read, the training side of a Lean project is not about making people better employees, better teamworkers, better leaders; it becomes assumed that through indoctrination in the Lean way, you will do these things naturally. There must be a thousand attributes of how people could be enabled to work better, without improvement of their processes, that Lean must miss.

Please don't get me wrong - I believe Lean is a powerful methodologies for making processes operate better and Agile is a powerful methodology for implementing software to further assist in the transformation. But Lean is not intended to be a completely holistic program for change in an organization. Therefore there is a distinct chunk of business improvement that is being missed by not addressing the performance and traits of employees and leaders outside of (or maybe it is alongside) the process. Is this a perception industry? Do people care about employee development, or are they just workers in a process that we can swap in and out at will?

Your thoughts and comments are much appreciated...

A post from the Improving It blog

Let us help you improve your business today. Visit

Wednesday, February 24, 2010

Corporate IT won't run Linux desktops because of the 'Skype attitude'

Skype is not a small product or small company. Their VOIP product is used by millions of people every day (it currently tells me that 20,944,942 people are online), and carries more international phone calls than AT&T. Skype's attitude to Linux desktop users is frankly disturbing. Take a look at the chart of software issues for the Skype client:

76% of issues in the JIRA database are new. Compare this with the Windows client, where 21% of issues are new. It seems that Skype is truly uninterested in their Linux user base, expending minimal or no resources on it. Even if you take into account that the Linux client is several steps back from the Windows version and the development team is apparently working on a new version, the number of open issues is still worrying (they claim to be in a late Beta stage, but audio does not work with a large proportion of modern OS versions and there is no timeline to fix it). So, Lesson #1, DON'T INSTALL A BETA VERSION OF SKYPE IF YOU WANT TO CALL PEOPLE!

Lesson #2 ties in closely with Lesson #1, and really highlights Skype's contempt for its Linux user base, and worse, new customers. If go to install Skype from the corporate website, as a new Windows or Mac user you are taken to a page where you can download and install the latest stable version of the product. Linux customers beware: you are taken to a download page for the Beta version of the client - remember this is the one with audio that does not work with the recommended audio configuration of many common Linux distros such as Ubuntu. You have no option of installing the latest production version of the product. Even as a paying Skype customer, buying telephone call credit you do not have access to a production version of the product. So good luck if you try the Beta, find it doesn't work and want to roll back. You can't, because Skype has removed all links to the production version of the software. You are just left with software that doesn't allow you to make phone calls. As a note, I have raised issues with the development team, Skype support team and attempted to reach the Product Manager. For months. No pleasant surprises unfortunately.

So why is the Skype attitude to Linux users so lousy? Now, let's use the common assumption that Linux desktops account for at least 1% of desktop deployments. This suggests that the number of Skype users online at any time, running a Linux desktop is 200,000. No small number, though still only 1% of the overall user base. Or 1% of Skype's income of $550m in 2008, which means only $5.5m. Linux customers are not worth the effort. Skype is probably regretting ever putting a Linux version out in the wild (because now they have to support and update it), and are scared of the Linux community's attitude that would prefer they publish libraries that can be used to produce open source clients separate from the snails paced development Skype does with the proprietary client. Its hard to reconcile the open source concept into such a proprietary company.

It appears that in Skype's mind, all Linux users are potential Beta testers of their software. We all want to suffer software failures and issues. We also apparently want no way to ever recover from them, unless we follow Skype's unknown and unpublished roadmap. This is the attitude to the Linux desktop of a large commercial software provider. Maybe this helps us understand why corporate IT is so reluctant about installing Linux desktops - even if they don't want Skype on their networks, they can't trust a highly visible commercial company to write Linux software that works and manage it professionally, so what hope is there of trusting free, open source organizations from doing the right thing?

A post from the Improving It blog

Let us help you improve your business today. Visit

Tuesday, February 23, 2010

Company assets aren't just laptops

Every company that has laptop or desktop PCs puts a shiny sticky label on it - the asset tag. This is designed to help the IT team identify the piece of equipment, service history, warranty and all types of other information related to it. The tag just provides a reference number that allows a service employee to lookup the item in some type of asset management system (an EAM system, CMMS, spreadsheet or paper). It is rare to see similar identification on other equipment, and obviously even more difficult to see how this same concept can be applied to items that have no physical presence, such as software licenses.

The reality is that anything that an organization owns that has a significant value, and that may require preventative maintenance, testing, repair or eventual disposal, can be tracked. The asset can be a row in a database, a line on a spreadsheet, a record in a dedicated asset management system or an asset in a Consected CMMS solution. Tracking the item in a more structured solution helps companies manage a range of related information and documents in one place:
  • Purchase details: vendor, date, cost
  • Warranty details: warranty length, terms, certificate, contact details
  • Product or equipment details: documentation, description, modification history
  • Location: address and location, local contact details, site access details
  • Preventative maintenance plans: service schedules and history, issues
  • Inspections: schedule inspections, electrical safety inspections, schedules, regulatory and legal requirements
  • Failures, service and repairs: history, repairs, spare parts, inventory
  • Replacement plan and performance metrics: planning cost-effectiveness of the equipment
  • Work orders and work management: assigning staff to service activities and customer requests

Many companies don't need all of these items, but they do see the benefit in collecting the major information and work orders in one place. The complexity of a CMMS can mean that a company actually ends up with more assets to track in terms of servers, software, printers and so on. Sometimes a lighter-weight approach such as Consected is needed, to start making things better, faster.

By having a single system in place to access information about a company's equipment, software licenses, servers, air conditioning units, tools, telephone, laptops, you can get the information when it is needed. You can also ensure that employees that must service, repair or work with these items know where it is, when they must do the work and what tools they will need, without you guiding them every step of the way. Finally, you can ensure that you meet the growing set of regulatory requirements for the disposal of equipment, while meeting your organization's new policies for the recycling or safe removal of waste electrical and electronic equipment (WEEE), often known as e-waste.

Meeting all the requirements for managing assets through their whole life is complex on paper in an Excel spreadsheet. Sometimes a more complex solution can make your working life easier, more productive, less stressful, and more rewarding.

A post from the Improving It blog

Let us help you improve your business today. Visit

Monday, February 22, 2010

Onshoring needs supporting too

There is a big push around onshoring - the term applied to pushing out parts of your business operations to locations that are often considered more cost effective, while avoiding the difficulties of locating services overseas. The McKinsey quarterly targets onshoring in the following way:
The advantages of staying onshore are greatest for a company with products that are not labor intensive, have short life cycles and high obsolescence costs, and target very time-sensitive customers.
The reality of onshoring is that it needs the same level of technology support, if not more, as offshoring operations. Of course, there is not the complexity of dealing with the local cultures and politics of a far off nation to deal with, but as McKinsey puts it, the benefit of onshore comes with time-sensitive customers. Only if you can ensure the premium customer service reps, the expert technical support staff or the fashion designers have all the information to hand that they need, and can communicate effectively with the rest of the company, can you ensure that your onshore operations will deliver the real-time, local timezone value that you are aiming for.

One big complexity for organizations is not in sharing their customer data (CRM is normally available) or communicating in an ad-hoc manner (email and phones are pervasive). The issues come typically from assigning work, handling and tracking correspondence received in disparate locations, sharing documents, and keeping visibility of performance. Without some form of work management, business process management, document management, collaboration or case management, it is virtually impossible to ensure that the remote locations (potentially multiple onshore offices and your HQ) are sharing information and tracking work effectively.

Any one solution is not necessarily the right way to go, but it should be a relatively easy task for a business improvement specialist to look at an organization and identify how improvements can be made, both in process and technology. Attempting to use people with only an internal world-view may limit the options presented, just due to the obvious risk that comes with going against the status quo. At the same time, you shouldn't always need a team of McKinsey consultants to tell you that an office in another state needs to communicate with your HQ through a series of fairly easy to identify, and implement ways.

A post from the Improving It blog

To implement workflow and process automation in your business today, visit

Friday, February 19, 2010

Embedded processes are a temporary fix

A sponsored IDC paper, "Embedding Additional Value into Applications: What Enterprises Need Most from Application Vendors" talks about the results of a survey on why application vendors need to ensure their applications can integrate and operate effectively inside and alongside other applications. Ignoring the obviously sponsored tone of the paper, one chart showing the results of the survey jumped out at me:

Of the optional responses to the question, "Please rate how important it is to your organization that each of the following features be embedded within your application vendors' transactional applications", the lowest number of respondents placed an importance on a "Configurable workflow engine to automate and optimize processes".

Does this mean that optimizing business processes through automation is not important?

The question in fact is leading up to an explanation of why the workflows inside enterprise applications such as Oracle, SAP, PeopleSoft, Pivotal, etc, are often so half-hearted in their implementation, and largely disregarded by businesses. The survey shows that people recognize business processes are not trapped within a single application.

Of course, components of a process may be embedded or contained in a single application, but it is rare for the whole end to end flow of information and work to be handled in a single application, accessed by every person that needs to touch it or interact with it. for example has just invested heavily in a new process modeling capability to fit into its application development environment. This is likely to lead to yet more workflows trapped inside small ad-hoc applications, with frustration from users. Development effort is likely to be diverted before the tech buzz dies down, focusing on get-rich-quick attempts at leveraging this new capability rather than an effort to manage the process independently (but fully linked with) the applications that people use to get their work done.

Business processes exist outside of individual applications. Organizations that achieve process excellence (lower costs, reduced waste, faster response, better customer service, higher quality) know this. Although these companies won't be afraid to leverage small embedded workflow improvements where they are cost effective to do so (i.e. freely included in current software and they improve a chunk of work that is otherwise manual), they know that this approach comes at a cost: a need to eventually incorporate that workflow into the broader business process that everybody involved has access to.

A post from the Improving It blog

To implement workflow and process automation in your business today, visit

Coming soon... Download the podcast of this blog post

Tuesday, February 16, 2010

Keeping it simple - Fidelity claims successes stem from reducing transaction errors

Reuters, reporting on a rare speech by Fidelity Investment's Vice Chairman Abigail Johnson, included some important advice that any financial institution could benefit from. According to the report, Johnson:

... made it clear that running retirement plans and improving customer service depended on having smooth-running systems and squeezing out incremental improvements.

"Even a small reduction in errors and rework can have a significant impact," Johnson said. "When computer systems cause errors, you can end up with The Dreaded NIGO - the 'not in good order' transaction," and risk alienating customers.

Fidelity has built a reputation in the industry around automated data processing, though what is not always clear is what percentage of their overall business processes are automated. But one could expect that the company (renowned for technology build-versus-buy always falling on the 'build' side) would have built a strong workflow automation technology to underpin their lines of business operations.

As Johnson's statements make clear, getting workflows right that incorporate customers, agents and backoffice administrators is probably also central to Fidelity's low rate of ''The Dreaded NIGO". Total automation, or complex models with every permutation represented are not required. Being able to capture information cleanly, track its use and prevent unnecessary errors are functions of a system in financial services or any industry that are essential.

A post from the Improving It blog

To implement workflow and process automation in your business today, visit

Onboarding employees - flying by the seat of our pants

Introducing a new employee to your company, getting him or her enrolled in appropriate benefits plans, ensuring the legal paperwork is appropriately filled out, even making sure there is a desk, PC, phone available, often appears chaotic. For many new employees and the HR / office managers that run around trying to pull everything together, there appears to be little predefined process to the onboarding 'process'. Flying by the seat of your pants is a better description, and the benefit of this approach is motivated by not needing to think about it ahead of time rather than getting a new employee comfortable and productive faster.

Many business process management vendors talk about employee onboarding as a process that their software can address and make better, and that is not entirely untrue. It would be possible to draw a sequential flow around the activities to be performed to onboard a new employee, but the reality is the workflow is not sequential. The vendors would respond that they could also draw a set of parallel branches, all coming back together at some predetermined, and largely artificially set time. In my observations of employee onboarding, this might work, but its likely that modeling the process branches would take longer than running the actual process.

Employee onboarding, as anyone involved can see, looks more like a project plan or bunch of checklists for a lot of different people than a series of straight-line process flows. There are some dependencies on the different branches of work, but it seems that most activities can be performed independently. This becomes pretty obvious when you look at the work that needs to be done:
  • Hiring manager: define employee role, position in organization, requirements for access to IT systems, work location, training, etc; introduce employee to the department; check in at defined checkpoints to ensure things are going to plan.
  • Office manager, shop supervisor, etc: organize facilities appropriate to the role, such as desk, phone, safety equipment, timecard.
  • Benefits administrator: identify appropriate benefits and present options; ensure completion of paperwork; register 'elections' in appropriate systems and coordinate with external suppliers (for example personal health information to insurer).
  • Trainers: deliver company and role training according to predefined schedules and to order.
  • IT: setup accounts; deliver appropriate PC; ensure security privileges are correct and seek appropriate approvals for secure systems.
  • Human resources / onboarding administrator: organize and track all the requirements; coordinate the completion of employment paperwork; identify special role or employee requirements; coordinate company training.

This is by no means a complete list, and the people in HR involved in doing this routinely know what they have to do. The problems come from the people around the outside, such as the hiring manager and IT. Since introducing new employees for those individuals may be a relatively rare occurrence, the burden goes back to HR to ensure that all the points have been completed.

This is where the value of defining an onboarding checklist comes in. Rather than investing a huge amount of time sitting in a room drawing out an onboading process, try spending a few extra minutes at the time of onboarding the next employee to document all the activities and requirements, as you run across them. Putting together a very simple project plan, especially identifying the dependencies on certain information or decisions, can greatly help those people that only occasionally participate in the process get everything done more or less right first time. And the plan is based on the real activities and issues you run across in practice, rather than an artificially enforced workflow that nobody can (or will) follow.

If you are an organization recruiting and onboarding many new employees, you can consider taking the baseline you now have and working with an external process improvement consultant to further refine the points, and identify clearly the areas of collaboration, processes, internal controls and case by case management. Working together you will also be in a better position to identify if a simple technology can help deliver and coordinate the many activities that have to be performed.

Whatever level of onboarding process improvement you choose to get to, the chaos that your new employees encounter is likely to be reduced, the people involved in the process feel less burdened by it, and employees are happier and more productive sooner.

A post from the Improving It blog

To implement workflow and process automation in your business today, visit

Wednesday, February 10, 2010

Should my financial identity be a 9 digit number?

"In 2009 the average data breach cost the affected business $6.75 million". "In about one out of five cases [...] a victim of identity theft switches banks.". These are the numbers reported by Forbes from a recent Javelin Research study on the effects of identity theft and data loss. The article focuses on two sides - when an individual's identity is stolen by chance, or when a financial services company loses masses of data due to a breach.

The cost of a bank acquiring a customer is hundreds of dollars, stemming from the marketing, the expensive promotions and the actual setup fees when the customer starts opening that account. So losing potentially thousands of customers through lax security measures is going to be painful for the bank, as it should be. In really large cases, the impact on an institution's brand, and the security or lack of it that people associate with that brand is impossible to calculate. That will make future customer acquisition efforts even harder.

Interestingly on the other side, where a customer is a one off individual in identity theft, banks are apparently footing more of the bill. And really, is this a surprise? Good customers are hard to find, so punishing them for something that they probably have no control over can only go three ways: the customer is liable for it, tough luck; the overall consumer banking system accepts that it is a cost of doing business to keep them credible with customers, and they work hard to formulate ways of protecting customers better behind the scenes; we all get persuaded by fear that the only safe thing for us to do is buy some form of identity theft insurance.

Even as banks are footing the bill and swallowing many of the costs associated with identity theft, it seems that individuals are not letting their guard down. There is still a sense of personal responsibility it seems. According the Javelin article:
"We've found that individuals really guard their personal identity," says Van Dyke. "They have a high degree of fear and anger when it's stolen, even when they have zero liability."

If consumers lose faith in the financial system to help them protect an identity that they have little control over, it is possible that we could see an impetus for a system that is not driven by how well you can commit to memory a nine digit social security number that has to be shared with almost every organization you do business with. Really how relevant is the IRS advice to 'keep your SSN secure', when there is little control of who can ask for it or how well they protect it?

A post from the Improving It blog

To implement workflow and process automation in your business today, visit

Tuesday, February 09, 2010

Improving processes doesn't always improve customer service

I had to smile when I read a post by Scott Francis this morning, on how a process improvement at Starbucks had actually ruined his experience and enjoyment of buying coffee there, and slowed down the process at the same time.

I can sympathize with Scott in his tale, where the process improvement has made it less important for the staff to know their customers, since they are more fixated with tapping the order into the screen of the register than talking to the customer and remembering their name and obscure order. When you're a regular, it promotes a nice feeling to have you and your vague coffee-mocha-concoction remembered. But partial automation of the order taking process removes that human touch just a little more. Screen fixation is common enough when talking to people at any age who have grown up around PCs, so its kind of refreshing to get service that does force people back into that mode. Starbucks appears to have broken that.

Frankly, Starbucks is about volume. The kids working in the store probably don't earn a huge amount, and their benefits probably barely eat into a busy store's margin. So should Starbucks be gradually automating them out of the picture? Should they be aiming for a point where you tap your own order into a screen at the door, and pick it up from a faceless vending machine slot further down the line (because there is no reason that the register and the coffee machine can't be integrated, since the machine is mostly push button anyway)? That is not what people are paying for, but could be easily achieved if profitability per coffee is the only desire.

I wonder if Starbucks took any of the advice of the process improvement experts talking about simulation (which I blogged about a little yesterday) - they have a repeatable process that is prime for simulation, but did they test a hypothesis for a new process just on paper or in software, or did they try the new proposed approach in a range of stores, not just the poorest performing? Based on Scott's feedback, the process improvements actually destroy the performance of a well performing store. Perhaps Starbucks could have looked at a range of improvements - like getting rid of the floor blocking 'gift displays', that nobody buys stuff from, putting in an extra service spot for a real, intelligent human being earning little more than the wage of any other service job, but happy to have some more cash to buy beer in the next college semester. The ROI on that could perhaps have been instant, compared with the cost of installing a sticker printer and retraining staff to be less personal in their customer relationships.

Process improvement can be a driver for better customer service - but it seems that there is a breakpoint. Replacing poor service due to long waits and lost customer information with a good workflow helps customers no end. Replacing the human touch that people still sometimes desire with more and more automation actually takes it too far, and we can all see through the fact that better customer service is no longer the driver - cutting costs is.

A post from the Improving It blog

To implement workflow and process automation in your business today, visit

Monday, February 08, 2010

Simulating business process chaos

If the flow of work in a troubled business process is on the edge of chaos (the analogy is the butterfly flapping its wings in one part of the process upsetting something seemingly remote), more than a nice bell curve of work and workers, how do you identify how to make you process better? This is the question a business process improvement discussion (BP Group on LinkedIn) has recently arrived at.

Most people would like to believe that mathematics and statistics could model their workflows, so that changes could be made in software that predict whether the change is good, better or terrible for the real process. Some of the common software simulation tools (components of BPM suites or independent) offer the type of capability to take a known process and model how changes will improve it. You are in a good position if you already have a known process with known input patterns and statistically smooth and predictable workloads. But if you are still tying to escape process chaos, you need to work hard with experience and common sense to identify how to make the process better, repeatable and measurable before software simulation will offer value, rather than more chaos.

When asked my opinion of whether a process is suitable for simulation, I answer the question with a question (or three):

1) is there a high enough volume of work for the inputs to be fairly represented by linear functions for workload?

2) are the actions performed by individuals and teams of workers predictable and repeatable enough in the time and effort required to be represented statistically?

3) is the process complex enough to require a computed model of the workloads, or if it is simple, are you as likely to get a similar result from common-sense?

Note, I'm way out of my depth on the actual functioning of the statistics applied (OK, rusty, not completely sunk). My fear of applying software simulation to many processes is this: at what level does the fact that work items are discrete, individual items of work start to affect the process calculations, rather than the nice curves and lines that the stats functions feed into the system? To use some common analogies, its like driving on the highway - normally you could model the flow of cars with fairly simple flow functions, but when there are too few cars on the road, the flow is affected by the individual driver's own actions, and when there are too many cars, the flow stutters and stops intermittently, before bursting back up to 50 mph for 20 seconds. I don't think that the software simulation tools can account for this type of discrete modeling or over-capacity modeling, which is often where the benefit comes in (really, don't you have bigger problems in you organization to be focusing on than the process that is running well?).

The approach that seems smart, but may not be viable in reality, is to trial a change with a representative portion of the process - a group of individuals performing a black box section of the process differently from the way it is performed today, allowing the a more or less true comparison of the improvements without jumping in with both feet. You commit resources to doing this, of course, and in physically constrained processes (such as a production line) it may not be possible at all. In some processes, this investment may be the only option.

For many of the use cases outside of those with masses of people and masses of predictable work (call-centers), it seems that the models required would be so specific to the process that you would be needing a custom development exercise to model your process that would change the moment you fixed the next piece of the process.

For many businesses, what is needed now is a whiteboard, a person experienced in the current process (with an open mind), a person experienced in process changed (often a consultant external to the business unit or company), and some way of measuring where work goes and how long it takes to get completed. Understanding your processes, and making them 80% better is essential before taking the tools that will allow you a statistically correct improvement of another 7.3275%.

A post from the Improving It blog

To implement workflow and process automation in your business today, visit

Friday, February 05, 2010

Anti-money laundering or discrimination?

Anti-money laundering - touted to prevent the funding of terrorist organizations, can also affect ordinary individuals if the policies and controls at the time of account opening are not carefully applied. This recent circular from the Hong Kong Monetary Authority (HKMA) to 'authorized institutions' (AIs - in this case banks licensed by the HKMA), shows that sometimes the people applying the rules can get a little overzealous:
There have been reports in the last few days that there were cases in which some persons from certain ethnic minorities had encountered difficulties in opening bank accounts with certain AIs in Hong Kong. According to the reports, this has been linked to the anti-money laundering and countering financing of terrorism (AML/CFT) requirements of the HKMA.
As the HKMA goes on to say:
"AIs should adopt a balanced and common-sense approach with regard to customers connected with jurisdictions which do not or insufficiently apply the FATF recommendations. While extra care may well be justified in such cases, it is not a requirement that AIs should refuse to do any business with such customers or automatically classify them as high risk and subject them to enhanced CDD process. Rather, AIs should weigh all the circumstances of the particular situation and assess whether there is a higher than normal risk of money laundering."
This is all very well, but how does a bank instruct its account opening agents in applying a balanced and common-sense approach? It certainly does not mean that an automatic trigger should be applied to anyone not from Hong Kong or mainland China (and Hong Kong banking is used to international clients), but how do you help an agent faced with a client from Burma (for example) decide what a balanced approach is, or when to insist on enhanced customer due diligence (CDD)?

It is hard to provide guidance based on common sense. The question becomes whether it is the responsibility of the banks to provide automated decision management software based on business rules software, as James Taylor commonly discusses. Or are there other simpler systems that they can use to make a decision on whether to follow an enhanced due diligence process?

I certainly can't claim to be an expert in the legal issues related to what demonstrates that an institution is following best practices in the AML portions of account opening, though I would suggest that any of the following may be better than offering 'common-sense' as the proposed solution:
  • Checklists providing a simple CDD risk rating, with an outcome driven by a numeric result
  • Decision tree, allowing a series of answers to questions to drive an agent down a specific path to a decision to accept or review further
  • Refer to an expert - a central expert that can be called to make a judgement under certain trigger events
Any of these approaches could be quickly implemented as purely paper / manual solutions, or built into a portion of an existing (or new) automated business process to truly enforce its use. I'm guessing that Hong Kong is not alone with this type of issue where common-sense is being recommended for decisions that are hard to assess by non-experts. Significantly better solutions are not hard to imagine, and may actually save institutions money (and regulatory pressure) in the short term.

A post from the Improving It blog

To implement workflow and process automation in your business today, visit

Thursday, February 04, 2010

Classification of documents and processes

"Everybody focuses on document classification, but nobody ever talks about how to classify business processes. WTF?" - the considerably sanitized question I received from a client this week. Its an interesting one that maybe I will change to "How should a business align the classification of its business processes with the documents and records generated and captured within those processes?".

Back a few years, when this blog was new(er) and I had the energy to be putting some interesting thinking into writing posts with uncatchy titles, I wrote about the converging classification schemes of documents and records. This discussed how the formal taxonomies and fileplans of corporate records management might start to align in a meaningful way with the full-text indexing (Google-style search), tagging (at the time Technorati, now every blogging and bookmarking tool out there), relational databases 'indexes or titles' and a bunch of other ideas for identifying information. The conclusion was that they need to, since all these approaches have value. The cost of doing so for a business can be enormous, since adding this information can be labor-intensive.

Now add the question from my friendly client, and not only do you need to work out the meaning of corporate information, you may also need to classify the business process that produced it. The question is, if I really do go about filing documents in a structured filing system, should that be based on the content of the document, the way we need to retain and eventually dispose of the documents, the business process that produced it, the arrangement of processes documented within a COBIT structure for SOX, or some other classification that only a highly paid expert could imagine?

There are many arguments for many different approaches. We already have file plans for organizations that need to manage their formal records - the EPA for example talks about its file plan in decent depth, and how mere mortals can build on the fundamental structure within their part of the business. The focus with the EPA's guidance is to organize the records into something that allows the information to be found in the future, the security requirements and the need to archive long-term or destroy the documents according to a schedule. I don't see much business process in this.

In my opinion, a well thought out plan for storing your documents and business records automatically includes the business process that generated the information, but only if it is valuable to do so. A simplistic example based on an insurance company, could put records into this type of classification:

-> Line of business
-----> Customer
----------> Policy
--------------> Underwriting
------------------> Quote
------------------> Policy generation
------------------> Renewal
--------------> Claims
------------------> Assessment
------------------> Subrogation
--------------> Customer service

and so on...

What I see here is that at the lowest level of granularity, the business process generating or capturing the information is represented. Typically each of these items is represented by an end-to-end business process, automated or not. The details of work within the process is identifiable within these buckets (according to its own indexing scheme) and the documents must be meaningfully linked into the same work, while being available standalone. This last part is important in my opinion - the old imaging and workflow systems of years ago often held the documents within the 'work item'. If the workflow ended, the documents disappeared. Or you had to artificially tack the retention and disposition of records onto the actual business process. You could never attach more than one process to a piece of information either.

However you build out a new records structure for a business, of whatever size, the processes you run become part of it. In my opinion this can help you get to a seamless transition between what is just a written structure, and something that is actually useful for identify and finding information scattered across the business.

When it comes to technology to help simplify some of this, it makes it easier to identify if you need electronic records management, document management, report archiving, business process management, workflow or one of the many acronyms that make up a complex technology stack. Decide on the focus that helps your business work better and then ensure that fits into your information management requirements. If you try the reverse, you may not have a business worth keeping records for.

A post from the Improving It blog

To implement workflow and process automation in your business today, visit

Monday, February 01, 2010

BPM SaaS - you can't do it with the same old stuff

I ran across this 'non-news' press release for Appian on the free PRLog service this evening. Based on its own research, Appian claims to lead the business process management software as a service (BPM SaaS) market. Since I haven't done my own research, I wouldn't wish to question this fact. My question is whether hosting a traditional, complex BPM engine and design tools on the Amazon public cloud is really different enough to allow the benefits of BPM to meet the masses. Is Appian a leader in SaaS? Or do they have a bunch of customers who thought that an application service provider hosting model really made more financial sense at the time? My gripe is not with the success - more the fact that I don't believe traditional BPM is a good fit for general purpose SaaS.

Now I can't deny that the survey they cite is a call for BPM SaaS to start to make its mark, but I believe it is for the wrong reasons. The September 2009 survey "SaaS BPM: Silencing the Skeptics" report from Datamonitor is quoted as saying: "The ongoing recession has helped the cause of both SaaS and BPM. The SaaS model has endeared itself to customers that are wary of upfront capital investments, while BPM has helped businesses adapt quickly to the massive changes in the external environment... Datamonitor believes that this is the right time for BPM to take its SaaS strategy mainstream.".

Unfortunately, this quote suggests that the only reason customers would select a SaaS model is to change the payment terms for the system. It is a cold, hard fact that traditional BPM software is just too complex for most companies to just pick it up and use it. The many benefits of BPM modeling, simulation and optimization are lost on companies that really want to rapidly pave the cowpath, to improve selected processes in a few days, allowing them to move onto the next pain-point. Simulating and optimizing processes (the functionality pushed by BPM suites such as Appian's) is for those organizations that are already well down the path of BPM, not those wondering where the path starts.

Congratulations to Appian for forging ahead with this model, especially as their competitors like Lombardi and Savvion get swallowed by conglomerates that would never allow a business model like this get in the way of software license revenue. Software as a Service is a great fit for companies wanting to improve their businesses - in my opinion traditional BPM reworked for a mass hosting model may not be the way to go about it. Its still the same old stuff, just hosted in the cloud.

A post from the Improving It blog

To implement workflow and process automation in your business today, visit

Dealing with volatility of account opening

The markets are recovering. Oh, no they're not. Oh, they are again. Are we expecting new customer applications this month? Wait a minute, we're overloaded.

New customers are the life-blood of every bank, credit union, mortgage company and securities brokerage. New customers bring new cash, and immediate opportunities for up-selling to other profitable products. The volatility of new home starts (up 10.7% in November 2009, down 4% in December - American Bankers Association) demonstrates the importance of converting prospective customers effectively when time are slow, and handling spikes in volume to take advantage of a rush.

Amazingly, most financial services organizations have products that still require an account application form to be printed, filled out in ink, signed and sent. Credit cards, mortgages, and brokerages typically need paper applications. This leads to 'not in good order' (NIGO) applications accounting for between 25% and 80% of applications made on paper, with manual processing taking 45 minutes (according to Forrester). In the lean times, the time it takes is not an issue if you have retained your back-office administrators, but during a rush everyone struggles to keep up with the volume of new customers. More mistakes are probably made during these rushes than any other, leading to further backlogs.

This is not new news, so why am I harping on about it again? Well, account applications are still done on paper application forms, which signifies to me that organizations still haven't worked out how to handle their account opening online. They are still making mistakes and valuable customers are still being irritated by their details being incorrectly set up. Customer services still has to deal with the aftermath of data corrections. And the auditors and regulators are still questioning whether they really have the supervisory controls in place that demonstrate they are following best practices.

For this reason, Consected has released a live demo of an account opening application form and back-office workflow, so financial services organizations can see how they could improve by putting applications online with minimal cost and effort.

Consected provides consulting services, SaaS and complete solutions to fit the needs of a specific client. A small credit union has very different needs from a mid-sized brokerage firm. So if you work for an organization that knows it needs to improve its account opening, but doesn't know where to start, contact Consected for a free 'account opening evaluation'.

However firms feel they want to approach streamlining their account opening forms and processes, they should consider putting it in place before the next big rush.

A post from the Improving It blog

To implement workflow and process automation in your business today, visit