|
How to evaluate and select ERP
The
greater the care taken during evaluation and selection of the ERP
product, lesser the time for implementation, and greater the chances
of success, says B K Khaitan. Here’s an overview of how RPG Cables
went about its ERP implementation
Investment required in ERP is
very high and an inappropriate selection of ERP could prove to be
a nightmare for the company. Meticulous planning is therefore required
in the selection process. If we take adequate care in the evaluation
and selection of an ERP product, lesser will be the time taken in
implementation and greater the chances of success. My experience
is that selecting an enterprise-wide ERP should typically take anywhere
between five to six months from the word go, to finally signing
the order.
ERP selection is very different
from purchasing off-the-shelf software like accounting or payroll
packages. We don’t need much skill to evaluate an accounting software
since it does not impact the business as a whole, and moreover,
accounting practices are fairly standardised everywhere. Therefore,
if an accounting package works for company A, there is a good chance
that it will work for company B too. But the same logic does not
work with ERP. There are a number of cases where the same ERP product
has been successfully implemented in company A but failed miserably
in company B. We need to analyse carefully as to why ERP fails and
take appropriate preventive measures.
In the evaluation process, to
start with, at RPG we performed the health check diagnostics in
the organisation to gauge how prepared it was for an ERP implementation.
On a scale of one to ten, our score was six. We had fixed the minimum
score as five, below which we would have postponed the decision
to implement ERP and concentrated on improving the business process,
systems and procedures. Many ERP implementations fail not because
of any technical defect in the software but because the organisation
is not ready for it yet.
ERP vis-à-vis business objectives
No IT initiative can be effective
if it is not aligned with the company’s business objectives. It
is therefore necessary to formulate a business strategy to achieve
business objectives and then align IT initiatives to achieve them.
ERP implementation may be one among such initiatives. We conducted
a workshop in the company to identify objectives and critical success
factors of each SBU. The objectives were defined keeping in mind
the organisation’s strengths and weaknesses, opportunities and threats
prevailing in the market. We then studied existing business processes
of each SBU and identified pain areas. The entire exercise was documented.
A detailed RFP was prepared based on this exercise. The RFP contained
functionalities required by us under each module—it was used as
a Bible to test the functionality of ERP products.
Our next step was to call ERP
product vendors and submit a complete business document along with
RFP with a request to fill in the blank columns so as to understand
the product’s suitability. We asked vendors to sign a non-disclosure
agreement before handing over sensitive business documents.
After two weeks of submitting
business documents and RFP to vendors, we arranged to meet with
vendors, where they came prepared with their presentations. We met
with our SBU heads to give briefings of their business and to sort
out any clarifications from the vendor’s side.
Having understood our business,
vendors were asked to submit the RFP duly filled in. On the basis
of their response, we gave ratings against each functionality, depending
on suitability. Some weightages were given to core functionalities
that were specific to our business. The total score was arrived
at for each product and ranking was done accordingly.
Product demo
Product demo is a very important
step in the evaluation process. Our experience has been that the
look-and-feel of the product, the level of flexibility, user friendliness,
etc, can be seen only in a product demo. Besides, product demo helps
determine end-user comfort level. It is at this stage that the key
users must be involved and their opinion taken. For the demo to
be effective, we prepared business data covering the entire business
cycle, right from enquiry to shipment stage and passed on this data
to the vendors. During the demo, we concentrated on looking at the
core functionality in the product. For example, we are in the manufacturing
sector and our products are mostly made-to-order, we were looking
for an ERP solution with a strong product configurator. We wanted
our customer to be able to configure the product on the Web before
sending an enquiry to us. We wanted to reduce our time in responding
to a customer enquiry. Similarly, we were looking for functionalities
like drum selection and cut length management in ERP, which are
specific to the cable industry. We did not waste time looking into
standard functionalities of sales, purchase, stores, accounting,
etc, since these functionalities could be found in all top-of-the-line
ERP products we were evaluating, such as SAP, BaaN, Oracle, Intentia,
JDE.
Since ERP cannot be implemented
without strong IT infrastructure at the back-end, we looked in to
our IT infrastructure needs along with the ERP evaluation. We wanted
to connect our factories and branches through leased lines, VSATs
or VPN. Each of these options had advantages and disadvantages.
The advantage of VSATs, for example, is that it can be installed
anywhere with a very high uptime and consistent performance. Opting
for a leased line would mean spending three months to first avail
approval from MTNL for a leased circuit. VPN is a cheaper option
than VSATs and leased lines; but for VPN to work, Internet service
providers would have to have a point of presence (POP) in locations
where we have business units. In case of locations in remote places,
we would have to go in for VSATs. We realised that the initial investment
in VSATs is not as high as leased lines but the operating expenses
of VSATs are higher than leased lines or VPN. We opted for a hybrid
solution using a mix of VPN, leased lines and VSATs depending on
the physical location of the unit and the type of application we
wanted to run in each unit. Bandwidth requirement is an issue that
must be addressed to be able to provide necessary bandwidth for
enterprise-wide ERP to work.
Computing total cost of ownership
(TCO) for each of the ERP products was a tricky issue. TCO comprises
licensing cost, implementation cost, hardware cost, AMC, networking
cost, etc. Licensing cost may be determined on the basis of named
users or concurrent users or may be a combination of both. There
are hidden costs, which if not understood will come out in the implementation
stage, which will raise the TCO substantially. We defined our own
standard template and submitted it to the vendors for their feedback.
In this way we standardised on the bundling of various software
modules in ERP.
Implementation partner
Implementation of ERP in most
cases is not done by the vendor but a third party. Therefore, evaluating
the implementation partner along with the vendor is extremely important.
We left it to the vendor to submit the cost of implementation on
behalf of implementation partners. We did not deal with implementation
partners directly in the beginning. All communication with implementation
partners was channelised through the vendor to avoid any confusion
or misunderstanding. We observed that implementation cost varied
from one implementation partner to another for the same product.
The difference in cost was as much as double and in some cases even
trebled. We ensured that the implementation cost given by the vendor
was agreeable to all implementation partners nominated by him.
We shortlisted vendors on the
basis of technical and commercial evaluation. Evaluation of implementation
partners was done on the basis of skill sets of implementation teams,
domain knowledge of the industry and their track record in implementation.
We asked each of the implementation partners to submit a document
containing details of scope, project plan, implementation methodology
used by them, etc.
Hardware cost also has several
components that need to be understood thoroughly. We asked our vendor
to do a sizing of the hardware on the basis of number of user licenses
and volume of transactions. In order to arrive at the user licenses
requirement, we did not go by the number of PCs we had. We knew
ERP was not going to replicate our existing business processes.
Roles and responsibilities of end-users were going to undergo drastic
changes. Keeping this in mind, we prepared a table in matrix form
in which we listed down names of employees with key responsibilities
according to their department. We then determined whether each employee
had to play any role as an ERP user and then defined his new role.
Correct estimate of user licenses had a direct bearing on the TCO.
Verification
A site visit is very important
to understand the implementation issues, if any, during implementation.
We were very selective in site visits. We asked our vendor to give
us the list of sites where they fully implemented ERP, preferably
in industries similar to ours. We visited at least one site of each
vendor and talked to project managers to know how they went about
implementing ERP and what according to them were the pain areas.
We had closed-door meetings with them, avoiding the presence of
a vendor representative. We were able to get balanced feedback on
products. In case of reference sites in other countries, we got
feedback through e-mail.
Whether to implement ERP in a
big bang approach or in a phased manner is something left to the
individual organisation. Decisions should be taken keeping in mind
level of preparedness, physical location of various units, availability
of resources, etc. There is a cost implication in each of these
approaches.
Finally, we did overall assessment
of ERP products based on functionality suitability and standard
features, product demos, TCO, site visits and feedback from reference
sites.
The author is the CIO, RPG Cables.
He can be contacted at bkay@rpgcables.com
|