|
    
|
INFOTECH The Imperatives of ERP or RIP? Compare stringently. Plan meticulously. Check extensively. Test rigorously. That's how Gestetner is getting its ERP implementation right. By S Chandrashekar
It isn't ERP that is to blame. The technology is good; no, it's great. The fault lies in its implementation. With the mismanagement of the process of converting the networked corporation into an ERP-driven organisation. Few transitions are as massive as the one that the traditional, paper-powered company makes when it switches to the ERP paradigm, where data is entered but once, where information is available for real-time decision-making, where the shopfloor can respond to a change in the customer's order in the twinkling of an eye.
Get this transition wrong--and you are sentencing your company to chaos compounded by computers. There are many issues involved in getting it right, but one crucial factor comprises two activities before the ERP package even enters your organisation: the choice of the package, and the entry-strategy for the software into your company. Warns Vivek Marla, 38, Senior Director (Practice), Oracle Consulting: "There is no such thing as the best package. You choose what suits your business best." How do you manage your ERP? One benchmark in pre-implementation ERP management has been created by the Rs 65-crore Gestetner, the Delhi-based manufacturer of configurable digital copiers (a.k.a. mopiers). Led by CEO Paul Wilkinson, 46, and the Chief Accountant-cum-Manager (Information Systems), Manorama Nagarajan, 37, Gestetner offers some first principles of managing the transition to the ERP era. The Requirement Gestetner's need for a real-time information management system flowed from its business design. In the mopiers market, understanding every customer's individual requirement, and translating it into the appropriate specs is crucial for success. So is on-time delivery. Both need instantaneous access to a mountain of data as well as databases of consumer preferences. After all, a typical product consists of as many as 45 permutations of features, settings, and add-on modules. And buyers' needs had to be met immediately if fleet-of-foot competitors like Modi Xerox or Canon weren't to run away with Gestetner's customers. Adding to Wilkinson's woes were the back-end demands from Gestetner's English parent. Monthly accounts had to be presented by the first week of the following month to the global office, which kept the accounts department in every branch and the CEO fully occupied. "We weren't able to update our databases regularly, which slowed our forecasting. When I called up my sales factsheet on the 7th of the month, it still showed me one-month-old figures," recalls Wilkinson. Fatal in a business where weekly forecasting-cycles are the standard.
Executed effectively, ERP would, obviously, solve both these critical problems. First, all information about customer preferences, delivery-schedules, production planning, and other backward linkages would be integrated, enabling different functions to operate in sync, and cut out delays. All data would be warehoused in one place, where it would be easier to sort out errors. And Gestetner would then have the information critical for competitive success when and wherever it was needed. And second, it would enable databases across the company to be updated with just the one-time entry of data, without the process having to be repeated in different departments of the company. In fact, a study had shown Wilkinson that his people spent 70 per cent of their time on data-inputting and 30 per cent on its analysis and usage. His target: 50 per cent each. To these benefits, he added a third: links with the supply chain so that it too could move to the same tune as Gestetner when executing orders. Acknowledges V. Balakrishnan, 38, Regional Manager (Rajasthan & Gujarat), Gestetner: "We believe that our focus should be firmly on the customer. So, all our information networks should be directed towards that." With this basket of requirements drawn up, Gestetner went scouting for its ERP system. THE ERP ALGORITHMS
The Search At the outset, Gestetner side-stepped one pitfall in the selection-process: before choosing an ERP package, a company has to decide whether it will modify its business model to suit the package. Or whether the model is inviolate, and it is the package that has to be adapted. Ab initio, Wilkinson decided that he was willing to change his processes, if necessary, since that would lower the customisation of the package and, critically, its cost. In fact, the extent of customisation is always a tricky issue. Says Shekar Dasgupta, 42, Country Manager, Oracle Software India: "We discourage the customisation of the package beyond a limit since the implementation becomes inflexible, no matter what the company's future business plans are." The dilemma settled, Gestetner went shopping.
No sooner had he let it be known that his company was in the market for an ERP system than Wilkinson was besieged by pitches from vendors: sap, for its R/3; BaaN, for its eponymous package; Oracle, for its applications; QAD, for MfgPro; Ramco, for Marshal; Visesh, for Movex; and Eastern Software, for Makes. Along with Nagarajan, Wilkinson decided that he wanted to see each of the packages at work before making a choice. However, requests to the competing sellers for demos of their products at other companies led to only boardroom presentations there. Reckons Wilkinson: "We must have visited at least 8 companies, but it was the same story everywhere." Instead, he asked to be taken to shopfloors and accounting departments, to middle managers' desks and sales-offices to judge for himself how effective each package was. But the field-trips provided concern--not decisions--with most of them providing little instances of genuine benefits. At Gaga Garments & Co. (apocryphal), the duo discovered that senior managers didn't use the ERP system for their reports. At Rightfit Valves (not the real name), the CEO did not even have a PC on his desk. Recounts Wilkinson: "The vendors promised a lot, but we got to see very little in terms of implementation." In fact, Wilkinson and Nagara 3 E M% EEM9 l EM9 Z 3M+ )EEEEEEM M E E 3 _^[U SVWE M% EEM9 EM9 3M+ )EEEEE}. EM ( M MU ( . EM ( M#MU ( _^[U SVWE E M% EEM9 l EM9 Z 3M+ )EEEEEEM ( M E E 3 _^[U SVWE M% EEM9 EM9 3M+ )EEEEE}. EM H M MU H . EM H M#MU H _^[U SVWE E M% EEM9 l EM9 Z 3M+ )EEEEEEM H M E E 3 _^[USVWh j E ( P2 _^[USVWE M+ EE M+ EE EEE EEEE EEE;Ec EM ( = . } E+E;E E+EEE } E EE} lowed him to leapfrog to a new level--and not just make incremental improvements. Realising that on-the-job demos of successful usage were difficult to come by, he had to apply other parameters. It was those, he says, that threw up QAD's MfgPro as the best fit. Cost, for instance, was a crucial factor. To keep its investments at a level where the returns would be justified, Gestetner used the licence-to-service ratio to help choose its vendor. This is the proportion of the licence cost--which scales up with the number of computers that are hooked onto the application--to the implementation cost. Comparing it across the competing ERP packages made it clear just how high the total bill would be. Since Gestetner had a budget--Rs 1.50 crore, with the leeway to climb to Rs 2 crore--this proved to be a differentiator. In fact, reasons Wilkinson, the cost of an ERP package is not just the ticket-price or implementation-tab; it also includes the money that has to be poured into making the changes in different parts of the organisation that the software demands. Since Gestetner would effectively be locking itself into the software it chose, the selection had to be made carefully. One of the most critical factors turned out to be the communication channels that the package would need for optimum performance. Would Gestetner have to augment its information infrastructure? That is where the sap, BaaN, and Oracle systems proved cost-unfriendly, says Wilkinson, since they needed dedicated leased lines to link different centres. Complains Nagarajan: "We have offices in 28 locations. Dedicated lines to each of these sites would have meant a huge bill every month just to keep the ERP running." That automatically ruled out products like sap's r/3, despite the undeniable attraction of having 1,000 best business practices from around the world packaged into it. Explains Partha Iyengar, 39, Country Manager, Gartner Group (India): "Vendors that are adept at selling their products to large corporations aren't proficient at meeting the needs of small enterprises. Even a Rs 100-crore company would be classified by most ERP vendors as a small company." However, MfgPro came up trumps since it could be run on Gestetner's existing hardware at all the 28 branches, with the cheapest form of networking: dial-up modems. Analyses Anand Veerkar, 35, General Manager (Country Sales & Marketing), Thirdware Solutions, which markets MfgPro: "MfgPro has a distributed computing architecture, and is optimised to work efficiently over a dial-up network. It also requires comparatively less investment in central server capacity." Adds Anil Bakht, 44, CEO, Eastern Software Systems: "We lost out in the case of Gestetner only due to its consideration for telecom and server costs." Gestetner almost opted for Oracle, though, because of the innovative approach taken by the vendor. Says Wilkinson: "Their people spent hours analysing our critical results areas, understanding our existing processes, and working out what processes we would like to have in the future, and how ERP could help us." Ultimately, Oracle turned out to have only one deficiency, flowing from the third factor that Gestetner looked at closely as a differentiator: the extent to which the globally-developed package had been pre-adapted to the needs of the local business framework. Covering issues, for instance, like the unique taxation
system. Says Nagarajan: "This was most important fo q Eg PE E/ Ef@ D Ef@ 6 Ef@ ( # }}} Ef@JE@J! Ef@J E@H! Ef@H E@$MILPEf@&Pu7
fMIHUfDJEf@HEf@$E _^[U SVWE MA
THE ERP ALGORITHMS
The Implementation That the CEO's commitment is vital for ERP to succeed is, by now, virtually a tautology. But, besides orchestrating the mandatory communication sessions with the rest of his people to explain the whys and wherefores of ERP, what Wilkinson also did was to prove that he would switch to the new system--always the highest hurdle to vault--personally. He achieved the demonstration effect by turning his own room in the office into an ERP Learning & Training Centre--and moving into his secretary's office. "I wanted to drive home my commitment to my colleagues," he says. "Besides, I don't need 6 tonnes of air-conditioning power." More important, Wilkinson decided that he would be the first to transit to the new system, and completely abandon the old one. So, he banned the entry of files of paper into his office, insisting that all information and reports reach him only through the workstation on his desk. "If ERP means being on-line, I should truly be able to call up any file from anywhere on this machine," he says. His people were, obviously, slower to change. After a week-long business trip, Wilkinson found a heap of paper files that had snuck back onto his desk! Back in his original office, Gestetner was going through the first phase of its ERP-implementation process. The starting-point was a training-session for 15 cherry-picked people, selected, after tests and feedback, on the basis of their expertise in their functional areas: inventory, sales, marketing, accounts et al. Why pick the best people? For starters, they would, obviously, be the quickest to understand how the system could be used to improve the data-management for their processes. Equally important, they would also be able to disseminate their new-found knowledge to their functional colleagues. Since financial accounting was the hub of Gestetner's ERP, it was Nagarajan--and not Wilkinson--who was chosen as the leader of the team. The 2 of them then selected the rest. The learning strategy? First, master the application. Next, review the business processes, with the ERP perspective added to the worldview. And, finally, learn to apply the software to the processes. Explains Nagarajan: "In the first week, the idea was to forget that we were employees of Gestetner, and to learn to use the ERP software as well as understand the business practices built into it." In the second, the team from Thirdware joined the Gestetner people, learning from the latter how the company's processes actually operated. Then began the crucial 16-week period, during which Gestetner's Group Of 15 implemented ERP as a pilot project, both for the software people and for the people learning to master it. The responsibilities were clear: while Thirdware would manage the technical aspects of customising the software and creating prototypes for testing, the Gestetner team would test-run them to examine their efficacy, and suggest changes. Initially, the exchanges were over the extent of customisation, and the areas where the original software needed the greatest level of change. Gestetner took great care to plan, improving the chances of a good fit between the application and the uses to which it would be put. Adds Nagarajan: "The thinking at this stage eliminated the need for a lot of rework." For instance, Gestetner's payroll- system could not be serviced by the module that MfgPro offered. Having realised this early, the module was customised right away instead of being caught short afterwards. Time to jump in? No. Still operating on the principle that every day spent testing upstream can save upto 10 days of wastage downstream, the 2 teams went on to replicate Gestetner's entire business model in silicon. Essentially, the company's processes and data-flows were re-created entirely on a laboratory computer. And MfgPro, customised over the previous weeks, was run on it. While the proportion of fixes needed was not high, some of them were critical--such as up-country customer-care. Nor will the dry-runs, which are still taking place, end now. Gestetner will follow up the fine-tuning with a real-life implementation in a small pocket: the National Capital Region (NCR) as the sales and service area, and the Head Office as the accounting centre. As a pilot, this was ideal as all aspects of the ERP network could be tested between the ground floor (NCR Sales) and the first floor (Central Accounts) at Gestetner's 4-floor headquarters. And this was the regional process chosen because it did not interfere with the rest of the country's sales and distribution--these months are the peak sales period for the company--so that a glitch wouldn't cause Gestetner's operations to break down. Crucially, real-world goals were set for this mini-implementation so that the outcome could be compared to the actual needs of the company--and not merely admired for their own sake. Only after these rigorous iterations will Gestetner finally apply the software on its actual business processes. As mandated by its training, the implementation will be done on a Do-It-Yourself basis. That's because an application implemented by those who will use it is, obviously, easier than something that has been set up by non-users. Once the results start flowing in, the documentation required for formalising the knowledge and embedding it in the corporate memory will be prepared. After this phase (estimated cost: Rs 1 crore) is completed, Wilkinson will spread the strands of his ERP network to other parts of the organisation. In the second phase, the idea is to cover the company's branches, going through the same process followed during the pilot implementation so that the learning is garnered first-hand by the people who will use the software. And then, in the third phase, the company's manufacturing facilities at Nashik (Maharashtra) and Calcutta will also be brought into the loop, with the total bill standing at Rs 1.50 crore and 30 months. THE ERP ALGORITHMS
The extent and focus of the planning that Gestetner has put into its ERP-usage should, by rights, be rewarded by quick results and relatively trouble-free implementation. Sure, there are almost certainly more road-blocks in its path, but they will, equally certainly, be products of highly unpredictable factors. Having not only prepared a roadmap, but also test-travelled on it as far along the way as possible, Gestetner can expect fewer problems than others who are exploring unknown territory. To build the ERP enterprise, therefore, always put the cart before the horse: Planning before Enterprise Resource Planning. |
|
Issue Contents Write to us Subscriptions Syndication INDIA TODAY | INDIA TODAY PLUS | COMPUTERS TODAY Living Media India Ltd |
 |
 |
It is burning hopes to cinders. Reducing the dream of data-on-tap into a
nightmare of non-performance. Turning grand visions of operational super-efficiency into
terrorising images of investments pouring down the drain. Company after company that has
tried to cash in on the wonderpower of ERP--Enterprise Resource Planning to the
orthodox--is gnashing its teeth in frustration and despair at the inability of its
multi-crore software to inject speed, accuracy, and synchronisation into its business
processes. 
