Category Archives: Procurement Innovation

Best Practice Technology Vendor Selection for True Multi-Nationals Part II: RFX – You’re Not Asking for the Right Information!

Today we continue our series on best-practice vendor selection for your enterprise e-Procurement and e-Sourcing solution. As per yesterday’s post, this series specifically relates to the selection of technology(-based) vendors for your enterprise software needs, and e-Procurement and e-Sourcing solutions in particular.

In yesterday’s post we reviewed the traditional RFX process at a high-level, which is more-or-less correct, and then reviewed how this process is typically interpreted, which is where the problems begin, especially where multi-nationals are concerned. We noted that there are three big problems with the standard interpretation of the process, which occur in the first part of the RFI process. In particular, we noted that most supply management project managers ask for the wrong information when they:

  • ask stakeholders for product/service requirements
  • ask stakeholders for preferred vendor recommendations
  • ask vendors for capability information and self-assessment

This last request in particular is especially futile as the last thing a vendor who wants your business is going to do is say that they can’t meet your request, even if their chances of success are dismal. You need to start by verifying that the vendor has the potential to serve you, independent of the vendor. To do this, you need to start the process where an average organization ends it. In particular,

Start with the Reference Interviews

Too many organizations do this:

  • Collateral-Driven Vendor Identification
  • Request for intent to bid
  • Request for proposal and quote
  • Shortlist
  • Negotiations
  • Final List
  • Reference Interviews

By the time a typical large organization gets to the reference interviews, it’s too late. Months have been spent on the project, which needs to be wrapped up shortly. As a result, the buyer gets stuck with one of the finalists, even if none of the finalists are good solutions for the organization.

If you’re a multi-national organization, you have to start with the reference interviews. If you’re a small company, only do business in one or two countries, and only conduct official business in English (and all your international suppliers are Chinese with english-speaking reps), then it doesn’t matter because just about every vendor can serve you. But if you’re a multi-national that:

  • has offices in over twenty countries,
  • conducts official business in seven languages (English, French, Spanish, Italian, German, Russian, Portuguese, for e.g.),
  • has suppliers that speak six more languages (Mandarin, Cantonese, Vietnamese, Thai, Korean, and Japanese, for e.g.),
  • and has to support customers in over forty countries

then not every vendor in the space is going to be able to support you. In fact, despite the plethora of companies in this space even after the recent M&A frenzy, the number of pure-play best-of-breed companies that will be able to support your global e-Procurement or e-Sourcing initiative is likely countable on your fingers, thumbs optional.

And the only way you can have any assurance that a vendor is going to be able to support such an initiative is to start with reference interviews with customers who are similar in size, scope, and needs.
While this doesn’t mean that the vendor has to have offices in each country that you are in, support every language that you need supported, and have a success story for each of the forty countries you are selling into, it does mean that the vendor needs to have a global presence, should support at least a dozen languages (that meet a majority of your language requirements) with a track record of being able to add new languages quickly, and should be in dozens of countries with a history of successful roll-out initiatives to new countries.

In other words, unless you have been convinced beyond a reasonable doubt that the vendor can support your organizational needs, they shouldn’t even get a detailed RFI. Because it’s not about who can survive the funnel, it’s about who deserves to even be in the funnel. And that’s a very simple determination, as we’ll discuss in the next post.

Best Practice Technology Vendor Selection for True Multi-Nationals Part I: RFX – You’re Asking for the Wrong Information!

Before we begin, it must be clearly stated that this series relates to the selection of technology and technology-based vendors to provide enterprise software platforms, and/or implementation services, back-office (processing) functions, or technology-driven consulting services for your multi-national organization. While some of the best practices contained herein may also apply to the selection of (strategic) suppliers for high-value and/or complex products and/or services, this series particularly relates to the selection of a vendor to provide an enterprise software backbone, and, in particular, a backbone for e-Procurement and/or e-Sourcing technology for your Supply Management organization. Furthermore, no claims, express or implied, are made with respect to any other vendor selection process and, in fact, if you’re only buying paper and pencils, some of the best practices contained herein will, in all likelihood, be overkill.

Now that the preamble is out of the way, let us begin by noting that the traditional RFX processed is well understood, and well documented in many places, including in the e-RFx for Total Value Management wiki-paper, co-authored by the doctor over on the e-Sourcing Wiki. And, in the wiki-paper in particular, the high-level process is more-or-less correct.

As per the wiki-paper, you start with a three-stage RFI before an RFP, which is solution focussed (and not cost or contract focussed), which is issued before a final RFQ, which is when you collect quotes and start the actual selection / negotiation process. Specifically, the high-level process is:

  1. RFI #1: Stakeholder Requirements
  2. RFI #2: Vendor Interest
  3. RFI #3: Vendor Pre-Qualification
  4.    RFP: Solution Inquiry
  5.    RFQ: Clearly-Defined Specifications

So what are you doing wrong, especially if you’re a Multi-National? To answer that, let’s look at how this is typically translated:

  1. Product Needs, Service Needs, Preferred Vendors
  2. Vendor Info. Request, Vendor Interest, NDA
  3. Product & Service Capability Profiles
  4. Solution Design Request
  5. Explicit requirements, process definition, and bid request

See the problems?

  1. Stakeholders typically don’t know what they need in a solution. They aren’t technology experts. They aren’t supply management experts. They are domain experts. It doesn’t matter what they think they need in a product or a service, it matters what problems they are having today. You need to ask them what problems they need to solve, so that you can ultimately select a vendor with the solution that solves as many of your stakeholder’s pain points as possible.
  2. A preferred vendor is one that can offer you the best product or service from an organizational perspective, not a single stakeholder’s perspective. For example, a stakeholder might rate a vendor A+ because the representatives always responds quickly. But this is not necessarily indicative of great service. If the answer is always “we’ll send someone to fix that in a week”, and you need the machine up 80% of the time, that’s poor service.
  3. Asking a vendor if they can provide you with the necessary functionality or service levels after you have shortlisted them as a possibility based upon a review of their collateral is not likely to get you anything other than a “yes we can”, especially if the vendor also offers consulting or “value added services”. One has to remember that most (big) consulting (and value-add) organizations are driven by a greed for dollars and the reps are told to always say yes and take on as much work as possible, leaving the question of how to get it done (if the organization is already stretched or weak in that area) until after the ink is dry.

Which brings us to the biggest problems with the current selection process, which we will discuss in Part II.

Worried about P2P Fraud? – Here’s How To Prevent Most Of It!

Accounts Payable News recently ran a good article on “the top six ways to carry out P2P fraud” that every Supply Management professional should read BEFORE implementing any P2P system. While the sheer presence of a P2P system will discourage fraud, as fraud will be much harder to hide and/or require collusion if the system is properly integrated, it also enables fraud to be conducted faster and at a much larger scale if there are holes in the implementation. But first, let’s look at the frauds identified:

    • Social Engineering
      A user who doesn’t need admin access gets it by convincing IT that it will be quicker if they can create accounts for authorized individuals, or that they need it for testing after hours. If such admin access can be used to create new, fictitious, suppliers with banking information that don’t require payment approvals …
    • Fictitious Invoices for non-PO spend below the bar
      If invoices below a certain threshold, like $1,000, automatically get paid (without a purchase order or, worse, goods receipt match) from preferred suppliers if the line items are on an approved list, then all it takes is collusion between a buyer and supplier to generate and approve a few (dozen) false invoices and both get a free vacation on the Riviera.
    • Reassignment and Undermining of Control
      If a fraudster can convince others to reassign approvals or part of the payment process to himself, then he can approve invoices from fictitious invoices from fake suppliers, which are actually companies, and bank accounts, he controls.
    • Receipt of goods not actually delivered
      If the buyer, who never steps foot in the warehouse or on the construction site, receipts goods never delivered, the buyer can arrange for a supplier to be paid twice if the supplier sends an invoice before the goods, gets paid right away, and then drops off a second invoice with the goods, which is then matched against a PO and receipted. And, of course, the buyer would get a kickback.
    • Approval of unapproved handling costs
      Which were never in the contract, but of which a portion will be kick-backed to the colluding buyer.
    • Supplier Payment Diversion
      A smart buyer will open a bank account in a name that sounds like it is the suppliers name, like MJ Consulting if the supplier is M&J Consulting, provide finance with new banking instructions from a spoofed e-mail account, and collect the payments until AP discovers they have incorrect account information.

If you analyze these types of fraud, you see a couple of commonalities:

      • Fake Data
      • Lack of Control
      • Collusion

It’s very easy with modern technology to prevent the first two and make the third harder, in that more people will have to be in on the fraud for it to succeed. Specifically, if you take the following steps:

      • Lock down access to finance and admin functionality to only those who need it
        and, using fine-grained roles-based security, restrict admin functionality to only those functions admin rights are truly required by the person
      • Require 2nd party verification of all regulatory and financial data associated with a supplier
        as no one should be able to enter and confirm the same data element
      • Only a person performing a function can enter data relating to that function
        as only a warehouse or site worker will know when the goods are/are not delivered
      • Also require 2nd party verification of any data element that can trigger a payment
        So, a goods receipt, as a whole, should be verified by a foreman
      • Absolutely no automatic payments unless ( a) the supplier is verified, ( b) the supplier’s accounts are verified, ( c) the goods were verified as received
      • Absolutely no payment for an invoice above the minimum threshold for non-automatic payment without a PO
        even if verified supplier, account, and receipt of goods
      • Absolutely no payment for an invoice above the threshold for which approval is specified
        without a manager approval, even if there is a PO, verified supplier, account, and receipt of goods
      • Absolutely no P2P/e-Procurement systems that don’t encrypt user access information, account information, and approvals. Otherwise, all an enterprising fraudster has to do is either (a) get onto the server and (a) query the database for an admin login, (b) overwrite the account record with his own bank record or, and this is way too easy in some systems, (c) set the approved for payment flag next to the invoice to true. The approval field should be a system encrypted value that only the system can decrypt to a valid “pay on” date using salts, hashes, and ciphers.

This will solve the fake data issue, as there can be no fake data unless there is collusion, and the lack of control issue, as there will be no way around the workflow unless there is collusion. You can’t solve the collusion issue, but you can certainly discourage it. Criminals tend not to trust each other, and when three or more parties are required to pull off a heist, the odds are much more in your favour.

Robbie and the Coupa Factory

Oompa Loompa Doom-pa-dee-do
We’re still building great products for you!
Oompa Loompa Doom-pa-dah-dee
If you are wise you’ll try it for free.

What do you get when you get lots of cash?
Filling coffers and enlarging the stash?
Teams of developers coding like mad!
Making the app work on your iPad.

You’ll like the look of that!

It’s been a long time since Davie ran the Coupa Factory, and while there may have been a number of notable changes in management, one thing hasn’t changed at Coupa — and that’s the original product direction (and the heart of the development team*). Coupa’s goal is still to make the best P2P platform out there that’s so easy to use that even your grandmother and three-year old can use it and get all your spend under management (SUM).

In terms of progress since our last major review of the platform in 2009, which focussed on QuickStart, there has been a flurry of development — of a very interesting sorts. Having built one of the richest web-based P2P platforms on the market in three short years (with everything capable of being custom configured by an administrative user), Coupa made a very important realization — 90% of P2P is simple, straightforward, and limited in terms of required functionality. Find what you need, put in a requisition, get approval, cut the PO to the vendor, accept delivery, accept the invoice, queue it for verification and approval, issue the goods receipt, and make payment within the agreed upon timeframe when the invoice has been verified and the goods accepted. There’s not a lot of inventory management, logistics, payment structuring, etc. in an average corporate purchase made by an end-user outside of the Supply Management organization.

What there is, in fact, is a lot of bypassing of e-Procurement systems that make the process of getting your printer, paper, or widget for your production line more troublesome than going to the Best Buy website, the Office Depot website, or just calling up the supplier and asking for another shipment. So, if you want widespread adoption, which is the key to maximizing your SUM (Spend Under Management), you have to make it at least as easy to use as Amazon. Whether you’re using vendor catalog, punch-out, cXML, or an in-house catalog, searching, shopping, and requisitioning is seamlessly integrated. And it’s all accessible through their new Super Search Bar that seamlessly integrates Google search for what you need, Amazon browse by category, and free-form item/service search (which also allows product/service retrieval through internal or vendor product/service numbers) which can be global, by commodity, category, or vendor — depending on your configuration. Plus, their new dynamic shopping cart, which integrates accounting and budgeting data (so you know against what budget item a requisition will be charged, how much is left on the budget item, and how much will be left when the requisition is approved the minute you add an item to the cat), allows for split-billing and overrides in the cart in a process that is as easy as Amazon’s “one-click” checkout. And, policies can be linked to each item and service that allow for additional information within a search to be “popped-up” so the user can get full information (which can include budgetary and billing information) before an item is even added to the cart. Supplier taxes can be imported and validated against your own tax tables, fending off future nightmares for finance down the road. And when all is said and done, payments can be recorded and integrated with you ERP (and Coupa now supports out-of-the-box integrations with a couple of dozen major ERP and e-Procurement platforms).

Upon checkout, the system will automatically create the necessary requisitions (which will be sent to each individual who needs to approve an item in your master requisition), and, upon approval, the necessary POs for each vendor will be automatically generated and delivered — and all are easily accessible from your order history page with a single click.

The other big developments since our last major review are their Expense Management Solution, which was still in its infancy, and their new Spend Optimizer Solution, which could more accurately called a 360° Spend Visibility solution (which you can use as a starting point in spend analysis and spend optimization). Their Expense Management application is a great way to get your T&E under control (as iPhone integration allows expense reports to be automatically generated by users who simply have to take a picture of their receipt and import it into the application). They’ve taken usability to a new level with this one.

Spend Optimizer is their mega-dashboard reporting solution that is completely configurable and allows for the creation of just about any spend report you can dream up. Done right, you can have it display off-contract spend; late payments; budget overruns; high-spend categories, commodities, and vendors; and other spend hot-spots that could get you or your organization into hot-water down the road if not proactively managed by exception. It’s still a dashboard (which can be dangerous and dysfunctional), but you have the option to see green, see red, or see where the black holes are. It’s this last capability that makes the real difference between useless reporting tool and powerful spend miner, because, generally speaking, off-contract is where the trouble starts.

And brand-spanking new functionality is lined-up for fourth quarter. Their recent announcement that 9.635 Billion has passed through their platform is impressive, but when they pull off their next round of application development, that will be more so. Coupa is still a company to watch, so don’t take your eyes off them for too long.

Oompa Loompa Doom-pa-dee-do
They’re still building great products for you!
Oompa Loompa Doom-pa-dee-dar
They have the goal to take your spend far.

* Dave and Noah may be gone, but David Williams is still there as VP Technology, overseeing day to day development.

e-Procurement on an ERP – Harder than Fetching Groceries in a Battle Tank

Procurement Insight has been tearing up a storm recently. First, they make it clear that “e” does not change the fundamental nature of anything, then they rip the covers off of Supply Chain Leakage, and now, for the triple play, they tell us that you “don’t take your tank to the mall, Mrs. Worthington”.

In the article, the author, Ian Burdon, described a particularly dispiriting conference where he went to a session on “Re-engineering Government Procurement” where “experts” claimed that the thing to do was to sort out procurement processes then hand it all over to SAP and Oracle. Really? In this situation, I would not have thought about throwing myself out of a high window but, instead, of throwing the experts out of the high window, because anyone who would say such a thing can’t be much of a Procurement expert.

As Ian said, the reason things were often going so badly in government (procurement) despite their having invested in ERPs was precisely because they were using ERPs. ERP stands for Enterprise Resource Planning, not Enterprise Resource Procurement. Failing to even ask what the “P” stood for was the government agency’s first mistake. The next mistake was failing to understand that a “planning” system that told you there would be a need for 5,000 toner cartridges, 10,000 reams of paper, and 500 litres of industrial solvent does not have to tell you how you would go about procuring these items in order to accomplish resource planning. Nor does it have to give you RFX, Auction, or analytics capabilities of any kind as it is not sourcing, nor purchase order management, invoice management, and payment management functionality, as it is not procurement. Trying to use ERP for e-Procurement is like trying to fit a dodecagonal peg in a hexagonal hole. (Good luck with that.)

Just because someone claims to be the Procurement oracle, doesn’t mean that they are. They just might be trying to make a sap out of you. Get an unaffiliated third party to figure out what you need, and get it. Not what a big corporation, or a misguided expert, tells you.