Monthly Archives: March 2011

ERP is NOT Always the Answer

Reading this recent article on “Mitigating Risk and Exposure from Subsidiary Operations” in Industry Week, one could get the impression that the only way to mitigate risk is to deploy one or more (connected) ERP systems to manage corporate data. Nothing could be further from the truth. While you do need consistent data and compatible systems, you don’t need an ERP. But I guess I should have expected such misleading advice given that the article was written by a VP at SAP, one of the biggest ERP vendors in the world.

According to the article, in order to mitigate risks to the company’s supplier, quality, liquidity, financial reporting, and unbudgeted spending, a company must streamline and automate mostly manual systems to:

  • enable the sourcing group to automatically provide information on preferred suppliers and negotiated terms to every subsidiary
  • enable headquarters to have ongoing visibility into cash-on-hand and receivables and payables across the organization
  • streamline the financial consolidation process
  • streamline inter-company purchasing transactions
  • implement collaborate processes such as forecasting and budgeting

Furthermore, according to the article, to accomplish this automation, a company needs to either:

  • deploy the same ERP system across the company,
  • deploy a two-tier ERP with simple data integration, or
  • deploy a two-tier ERP with process integration.

First of all:

  • A (cloud-based) SaaS e-Sourcing/e-Procurement platform with contract & supplier management can maintain preferred suppliers and terms and be accessible by every subsidiary.
  • A shared (cloud-based) SaaS accounting / finance system will allow headquarters to have a view into each subsidiary’s financials …
  • … and this shared system will streamline financial consolidation.
  • A (cloud-based) SaaS e-Procurement system will streamline inter-company purchasing, and
  • a cloud-based inventory / distribution / warehousing / logistics management system will allow for collaborative forecasting and budgeting.

So you don’t even need an ERP at all to accomplish the stated goals. Furthermore, while you do need integrated data, you can maintain this data with a simple relational database and integrate it using an off-the-shelf data analysis package with good ETL (extract-transform-load) tools that can merge flat-file data dumps from each system into one file/database for analytics purposes.

This isn’t to say that an ERP at headquarters to maintain master data isn’t worthwhile, just that you don’t need one, and that you certainly don’t need ERP deployments at all of your subsidiaries to accomplish the goals, which is important because enterprise ERPs generally cost seven figures and the cost is generally not justifiable for a small subsidiary.

One Important Lesson Not Learned From Six Failed Implementations

Over on the 21st Century Supply Chain, you’ll find a post entitled “six lessons learned from six failed software implementations” which is quite scary, because it indicates that there was one very important lesson that the organization did not learn.

If you don’t understand the technology, get a 3rd party consultant who is an expert in technology to guide you. (Don’t rely on a vendor!)

One failure is understandable. Every organization will fail in a technology project at one time or another. What’s important is what happens next. If the organiztaion is able to identify what it believes are the (primary) reasons for failure and solutions to those problems, then it is understandable if the organization tries again on its own. (And if it doesn’t, see the above lesson.) If it fails again, then the organization has to admit that it needs help and get the help it needs.

Because if it doesn’t, it’s just going to fail again and again and every other lesson learned is going to be irrelevant because the likelihood of it succeeding in time to get an ROI is slim, approaching none over the long term.

Overqualified Candidates are Truly Rare …

… and rarer still are instances when you should pass on these candidates. A recent post over on the HBR blogs that asked if you “should hire an overqualified candidate” made some great points about the assumptions made by Hiring Managers when presented with “overqualified candidates”, and hinted at a few others.

Most Hiring Managers misunderstand what overqualified is
A candidate is only over-qualified if they exceed the skill requirements of the job. This means that the following candidates are not overqualified:

  • candidates with an advanced degree that exceeds stated educational requirements
    because the education might not be that relevant anyway
  • candidates with considerably more years of professional experienced than expected
    because if most of a candidate’s experience is in a different role (because they just changed career paths a few years ago), the experience with respect to specific skill requirements could still be minimal
  • candidates with a lot of experience in similar roles in the function
    because candidates with 20 years in tactical order placement and processing would not have a lot of experience in strategic negotiation, a major requirement for sourcing professionals today

Today’s job definition will not be tomorrow’s job definition.
Business is evolving as rapidly as the technology that drives it evolves, and this means that the requirements for a role are no longer static. If the job responsibilities are evolving rapidly, you will need a candidate with more education, skills, and experience than the job requires today to keep up.

There’s nothing stopping you from paying a candidate what he or she is worth.
Maybe you planned to pay 60K, but if you get a candidate who is so perfect for the role that he or she will be twice as productive, and you can get that candidate for 90K, you’re getting someone who can do the work of two people for only 75% of what it would cost you to hire two lesser skilled candidates.

Just because a candidate is overqualified doesn’t mean that he or she will be bored or move on quickly.
This particular misconception drives me nuts. Some jobs are always challenging. Like sales. You never know what the customer is going to want. Or development. Technology is always changing and you never know what new technology is going to pop up that you will have to integrate with or what new bug will appear in the next release that you will have to track down.

Not every candidate wants your job.
Not everyone wants to be the boss … and, in fact, a candidate who has been the boss and decides that she would rather spend her days getting work done instead of fighting fires, going to a never ending stream of management meetings, and micro-managing lesser qualified employees who can’t keep on track without constant guidance is less likely to try and take you job than an overly ambitious over-achieving up-and-comer. If you create the right position for the individual with the most impressive non-boss title you can give them, pay them well, and free them to do what they want to do, they will likely be more than happy to leave you to you own personal boss-hell while they build systems that work, successfully source strategic categories, and design and implement new processes for efficient operations.

Bottom line, there are very few overqualified candidates and fewer still who would not make a good hire if you pay them well and give them the opportunity to shine (because most people would rather complete a task and have a sense of accomplishment than “be the boss”). So if you get a very qualified candidate, the first thing you should do is get her in for an interview before the competition does — because she is the type of candidate you want.

Is Continuous Improvement Top-Down or Bottom-Up?

It’s neither, because it’s both.

As per this recent article in Industry Week that asks “top down or bottom up”, Continuous Improvement (CI) requires top-down executive and management support, but the improvement itself must be initiated at the ground-level by the day-to-day workers in the plants, the back-office, and the warehouse floor.

The key to improvement is to identify inefficiencies and fixes for those efficiencies. The best people to identify these inefficiencies are the people who use the systems and processes every day. Likewise, the best people to identify solutions are the people who will have to use those systems and processes every day. Although you may need to bring in a CI facilitator to help your teams unlock their creativity and identify the solutions that will improve your operational efficiency, true solutions will come from within.

And even if these solutions cause the organization take a short term hit on the balance sheet or P&L, a true improvement will deliver lasting impacts where working capital, cash flow, quality, and productivity are concerned.

The Technology Waves Are Always Washing Up On the Shore

But they don’t always make a lasting impression. So how do you know which ones precede a tsunami?

It’s hard to say, but like a tsunami, the waves are not caused by hot air (wind) but deep movements within the (end) user community (geological effects). If you keep this in mind, it is more likely that you will be able to identify the technology waves that will reshape the business landscape.

And you will be able to make sense of this recent report on “the technology waves that are reshaping the business landscape” by Accenture. In the report, Accenture identifies eight trends driving the future of information technology. Some of significant, others not so much. Let’s take them one-by-one.

  • Application Services as Utilities
    One just has to look at the massive success of Apple’s App Store to realize that the end user mindset has shifted from applications as large monolithic software packages locally installed from CDs and heavily supported by local users to to small, point-based solutions, which can be installed from, and live in, “the cloud” on-demand. Where consumers go, businesses, which employ the same consumers, must follow.
  • “Social Identities”
    Facebook has proven that “social identities” are important to our online society. And with a number of “enterprise” platforms working on integrating social technologies, it’s clear this is a significant shift that will shape application design for years to come.
  • Cloud Computing
    Amazon, Google, and other big providers have proven that a number of businesses, for better or worse, want to move to “the cloud”. This is another shift that is here to stay.
  • Analytics
    We’ve been talking about analytics and BI for years, but there’s never really been a sharp jump in demand to identify BI as the next big thing or to indicate that it will be any more important in the year to come than it was in the year before.
  • Data Security
    Security has been a constant issue for over a decade, and the need for security is no greater this year than last year. And most firms are still of the mindset where they are only going to take this seriously when they get breached.
  • Data Privacy
    Data Privacy, which was big in the 90’s, thanks to Facebook, is now big again. The widespread, public, backlash to Facebook’s initial lack of privacy, and controversial privacy agreements, has re-ignited the privacy debate across the globe and privacy is now under the microscope again.
  • Architecture
    Architecture is obviously going to shift as applications move from monolithic software packages to on-demand utilities, but since consumers don’t really care about architecture, it’s going to take a back seat to the application-on-demand movement.
  • User Experience
    Today’s users demand the Rich Internet Application experience. Thanks to Adobe, Apple, and other companies focussed on the user experience, consumers expect a constantly improving experience. This wave is also growing.

Net result, applications as utility on demand, the cloud, social identity, privacy, and user experience will play a major role in technology in the years to come, but analysis, data security, and architecture, will continue to take a back seat to these more prominent issues.