Flawed logic

Look carefully at the typical factory floor, and you'll see something odd. And it's this: production people diligently using computer systems that they know to be sub-optimal – while knowing at the same time that there are specialist manufacturing systems out there in the marketplace that would do the job, but which they are prevented from using. So take a look at what they're using instead of a specialist manufacturing solution, and sometimes you'll see a system laboriously constructed in Microsoft's Access database, or its Excel spreadsheet tool. Sometimes you'll see them using the generic capabilities of whatever ERP system the business relies on as its transaction backbone. And sometimes, you will see them using a specialist best-of-breed manufacturing system – but a legacy, out-of-support version, rendered semi-obsolete by an inability to upgrade to something newer. In each case, the outcome is the same: what you'll see is a manufacturing function fighting with one hand behind its back, forced to use tools far from adequate for the task to which they are being put. So what's going on? How has this come to pass, and how exactly is this influence exerted? In fact, there's no single, causal factor at work, explains Kevin Luxton, founder and chief executive of QiSOFT, a specialist provider of a suite of factory-floor productivity and quality solutions. Partly, he says, what has happened is down to hard-pressed IT departments: under pressure to deliver ERP implementations, they're resisting production requests for specialist manufacturing systems. In short, they haven't got the budgets, they've haven't got the integration manpower – and they do know that what they have got is an ERP system which seemingly offers very similar functionality. "'You want a quality system? Use the one that's built into the ERP system.' That's the message coming out of IT," says Luxton. "It's a very understandable message – particularly when you factor in ERP vendors' equally understandable tendency to over-sell the capabilities of their products." Granted, this is a message that helps to stretch IT budgets further. The typical manufacturing business running an ERP system from companies such as Microsoft, SAP, or Oracle, will have made a significant investment in that ERP system. Expensive software will have been licensed, users will have been trained in how to operate it, and the system's tentacles will have been woven into the heart of the business. So superficially, the logic is clear. Why license yet more software in the form of a factory-floor solution such as a quality system, or manufacturing execution system? Just insist that the manufacturing function use whatever capabilities that Microsoft, SAP, or Oracle have built into their respective ERP systems. The trouble is, it's a mistaken message. Simply put, says Luxton, it is seriously flawed – both factually and logically. From a factual perspective, he asserts, the argument misses the point that the manufacturing-specific functionality built into most ERP systems is fairly generic, and lacks the depth and richness required to make a real difference to manufacturing performance. "Sure, there's a quality module – but do a feature-by-feature comparison with a specialist best-of-breed system, and the gaps are enormous," says Luxton. "What is in the ERP system is a quality module in name, but not in functionality. And it's the same with other factory-floor solutions: ERP on its own just doesn't deliver." Which takes us to the logical flaw in the IT function's argument that ERP-based manufacturing functionality is better. Because the core competency of a manufacturer is manufacturing – which means that the business should be doing all that it can to further its manufacturing prowess, and building its manufacturing-based competitive edge. Which won't be achieved by settling for anything less than the best when it comes to manufacturing-centric or factory-floor software. Put at its starkest, then, the final decision on the choice of manufacturing software should lie with the manufacturing director – and not the IT function. Because from a strategic perspective, what matters most to a manufacturing business is raw manufacturing functionality. "It's the tail wagging the dog," says Luxton. "To have the IT function take the final decision – and veto compelling best-of-breed offerings purely on the grounds of ERP standardisation – is madness. And strategic madness, at that." But it's a madness that's far from uncommon, he stresses. Time and again, reports Luxton, he meets companies which are making software selection choices on precisely those grounds. "They'll come and talk to us, be impressed with the functionality we offer, and then be told by their IT functions that they can't buy it," he says. " And typically, it's a battle with one of three outcomes: either the manufacturing people come out on top, or they have to settle for whatever functionality is in their ERP system, or the manufacturing people continue to struggle on, achieving what they can with their Access databases and Excel spreadsheets. However, many of those eventually realise that they cannot just 'make do' with what they have, and years later return to buy the system that they were so impressed with in the first place." And as with ERP-based functionality, such homebrew solutions are also flawed – and not just because hand-crafted spreadsheets can be bug-ridden, and consume scarce resource to create and maintain. Instead, says Luxton, the danger is that they are generally backwards-looking, providing information about what happened yesterday or last week, but very rarely providing useful insights into what is happening right now. "This matters because, for manufacturers, real-time information is actionable information," he points out. "The short and faster the feedback loop, the better. If things aren't going right, you want to be taking corrective action, and not planning overtime or re-work." So is there any good news at all in this stark assessment? According to Luxton, yes. Because two separate developments are once again tipping the scales in favour of niche applications. First, he points out, the design and engineering teams of more and more ERP vendors are recognising that their inbuilt ERP capabilities are not best-suited to complex real-time manufacturing, and its predominantly non-transactional data. "They've looked at the investments required to develop and maintain solutions that would deliver critical functionality in some of these areas, and are backing away," he says. "Beyond a certain level of basic capability, they're asking themselves: 'Is this where we want to go?'." But this unease isn't reaching the ERP vendors' salesforces – nor the manufacturing industry customers to whom they're speaking. "When you are selling systems for millions of dollars, pounds, or euros, you're certainly going to want to give the impression the system does everything," notes Luxton. "And equally, when you are buying a system that costs millions of dollars, pounds or euros, then you are certainly going to expect it does everything." The second development again offers hope. Simply put, ERP vendors are building better and better connectivity and integration features into their ERP offerings, using technologies such as XML and web services. In other words, integrating a best-of-breed solution to an ERP system has never been easier – and is likely to become easier still, over time. Nevertheless, points out Luxton, the battle won't be over any time soon. But at least the battle lines have now been realistically drawn up, he says. "It's no longer a debate about functionality, or integration issues," he concludes. "It's about cost: does a manufacturer save money by settling for an inferior solution, or does he do the right thing and go with a built-for-purpose specialist system? And that's a debate we're very, very happy to have."