Category Archives: Procurement Innovation

A Hitchhiker’s Guide to e-Procurement: Procurement Models

Mostly Harmless, Part XXII

Previous Post

There are three major procurement models. While they all require the same basic e-Procurement functionality, each model dictates its own e-Procurement requirements and emphasizes certain features and functions above others. This post will address the different models and which features and functions are emphasized by the models.

The first major model is the classic model of decentralized procurement where each business, functional, or geographic unit is responsible for its own purchases. Individual business units are empowered with autonomy and control over their process and design decisions. It allows for quick processes and issue resolution and allows the organization to take advantage of expertise in the local market. However, it does limit an organization’s ability to leverage the corporate spend or to align the business unit objectives with the objectives of the global organization.

If an organization is using a decentralized model, the e-Procurement system will need to be quite flexible as each unit will likely have its own requisition formats, approval processes, purchase order distribution, invoicing, and payment processes. It will have to support multiple workflows, and allow for analyses against multiple budgets. And it will have to meet the different needs of the different units.

The next major model is centralized procurement. In the newer, centralized, model of procurement, all procurement goes through a single centralized model. This has its advantages as it allows the organization to fully leverage corporate spend and drive standardized sourcing processes across the organization. However, it can lead to lost knowledge of local supply markets and consumption patterns, which can result in sub-optimal buys for many regions. It can also increase the risk of maverick spend when the geographically dispersed site managers do not agree with centrally mandated decisions. And it can increase reaction times to unexpected changes in supply or demand.

If the organization is using a centralized model, the e-Procurement system will need to be scalable and high performance as the global organization will be using a centralized instance and a centralized data store. It will need very good fine-grained roles management as there will be hundreds, or thousands, of users, who will not only have different roles, but different levels of access to the system and the data contained within. (A manager might only be allowed to see activity from her team in her local unit.) And it will need fine-grained data classification capabilities to support unit-based analysis, which could be built-in or through an external tool.

The final major model is a hybrid procurement model known as center-led procurement. In a center-led model, a procurement center of excellence (COE) focusses on corporate supply chain strategies and strategic commodities, best practices, and knowledge sharing while leaving tactical buys and tactical execution to the individual business units. It was designed to give an organization the best of both the centralized and decentralized worlds with as few disadvantages as possible.

If the organization is using a center-led model, the e-Procurement solution needs to be extremely flexible, scalable, and robust as it will need to support distributed instances with different workflows for each business unit for decentralized purchases as well as a centralized instance with a single master workflow for centralized instances. Role management and security will need to be extremely fine grained as users could have different permissions depending upon whether the spend is centralized or distributed. And data management will have to be fairly intelligent as some spend will be centralized while other spend is distributed.

Next Post: Sectors

Share This on Linked In

A Hitchhiker’s Guide to e-Procurement: Costing a Solution

Mostly Harmless, Part XXI

Previous Post

Every solution costs more than the sticker price. But how much more? In this post, we’ll outline how to cost the various solutions as well as a methodology for calculating the expected value.

First of all there’s the cost of the license, which can be significant. If the system is enterprise, and especially if it’s an installed solution, this can be a very significant up-front cost in the six figure range. Then there’s the maintenance, which is required for support and mandatory for some solutions, and built into the price of on-demand/SaaS solutions. This can be as high as 22% a year for some solutions. Then there’s the installation and integration costs. Even a SaaS solution will require some setup, and the e-Procurement system will need to be integrated with accounting systems, sourcing systems, payment systems, and other enterprise (resource planning) systems in order for the organization to extract maximum value for the system.

Then there’s training costs. Even though a good system will be extremely easy to use and self-explanatory where basic functions are concerned, some training will still be required. This is especially true for the administrators, who have to maintain the system, and analysts, who have to analyze processes, performance, and spending. In addition to training costs, there will be support costs. Administrators will have to be employed to continually maintain the system (data) and train new users. If the system is installed, they will also have to do patches and upgrades in addition to maintaining system data and (business) processes.

If the organization is looking for an installed or hosted ASP solution, there will also be hardware costs, database costs, application server costs, and middleware costs. These costs can easily dwarf the system costs if the organization doesn’t already have any of these solutions. And even if the organization has some of these solutions in place, there will likely be additional license fees. Finally, there will likely be additional IT (support) costs to maintain the hardware, which will have to be upgraded on a regular basis, and the supporting software.

When all is said in done, the cost of a solution can end up being 10 times the sticker price, so it’s important to understand the total cost of ownership before choosing a solution. This is not to say that a solution with a seven figure total cost of ownership is expensive. It might be, it might not. It all depends upon how much it costs relative to other solutions being evaluated, how many users will use the system, how much it will increase organizational efficiency, and what ROI the organization expects to see.

Fortunately, the calculation of expected value is quite straightforward once the TCO is known. It’s simply a matter of computing the ROI according to the following formula:

(savings expected from increased efficiency +savings expected from maverick spend reductions +savings expected from newly identified opportunities) /total expected cost

While some of these numbers may appear hard to calculate, they are easy to estimate and what is really important is order of magnitude. For example:

  • if the organization expects to increase efficiency 200%, that’s a 65% workforce reduction against current workload; if the organization currently requires 20 people to handle tactical procurement tasks, at an average salary of 62K, that’s a reduction of 13 people or about 800K per year
  • if the organization currently has a maverick spend rate of 30% and expects, using third party benchmarks, to reduce that by 66%, that’s an 20% reduction in maverick spend; if maverick spend, on average, costs the organization 5% of spend on average, if the organization spends 100M annually, that’s a projected savings of 1% (20% of 5%), or 1 M in one-time savings
  • if the organization expects that an e-Procurement system will identify additional savings opportunities on 20% of spend annually and that the average savings that will be obtained will be 10%, then the organization would expect to save 2% of spend, or 2M annually

All told, if the organization expects to use the system for five years, it would expect to save 15M over five years (5*800K + 1M + 5*2M). If the total cost of ownership of the system was determined to be 3M for five years, then the organization would expect to see an ROI of 5X, which should be a buy decision. Of course, if the calculations worked out that the organization only expected to save 5 M, and the ROI was only 1.6, the decision should be to find a more cost effective solution.

For more details on cost calculations, and a starting spreadsheet, see Sourcing Innovation’s post on Uncovering the True Cost of On-Premise Sourcing & Procurement Software in the archives.

Next Post: Procurement Models

Share This on Linked In

A Hitchhiker’s Guide to e-Procurement: Catalogs & Contracts, Part II

Mostly Harmless, Part XX

Previous Post

The last post defined catalogs and contracts and discussed reasons why they will need to be revisited and revised on a regular basis. As promised, this post will address the associated challenges of catalog and contract maintenance, some associated best practices, and the benefits that could be expected from an appropriate e-Procurement solution.

Common Challenges

  • Unused Item/Contract Identification

    Catalogs are continuously updated and procurement constantly negotiates and renegotiates contracts. However, how many of the items are ever bought and what percentage of the contracts are used for more than a short time?

  • New Item Identification

    What items were bought this month/quarter that were never bought before? Which are not associated with a contract or an approved catalog?

  • Similar Item Identification

    For those items which are not on contract, were there similar items on contract that would have sufficed? If not, were there at least similar items in approved catalogs that would have worked?

Best Practices

  • Automatically Flag Items Not on Contract and Force Supervisory Review

    The best way to reduce maverick spend is to prevent it from happening in the first place. Forcing a supervisor to review all purchases not on contract (above a certain dollar limit or for products / categories there are contracts for) can put a significant dent in contract spend.

  • Automatically Flag Items Not in the Catalog and Force Procurement Review

    Not everything will be on contract, but there’s no reason that the majority of goods and services that the organization needs to buy on a regular basis can not be in the catalog. Unless the item is brand new, it should be in the catalog if it is needed. Forcing Procurement review will minimize the purchase of off-catalog items where price, and associated spending levels, are unmonitored and where pricing could spiral out of control.

  • Automatically Identify Items in Contracts and Catalogs that Have Not Been Purchased in the Last Month, Quarter, Year

    New items need to be tracked and monitored as any new items bought in quantity on a regular basis are prime candidates for future contracts.

Potential Benefits

  • Improved Contract Compliance / Reduced Maverick Spend

    The automatic flagging of off-contract and off-catalog purchases for manual review and approval can greatly increase contract compliance while simultaneously reducing maverick spend.

  • Easy Identification of Additional Savings Opportunities

    The automatic identification and tracking off off-contract items and associated volumes can identify some of the best opportunities for future savings opportunities.

Once the catalogs and contracts are up to date, it is time to begin the cycle anew. The next post will move on to how to cost a solution.

Next Post: Costing a Solution

Share This on Linked In

A Hitchhiker’s Guide to e-Procurement: Catalogs & Contracts, Part I

Mostly Harmless, Part XIX

Previous Post

A(n e-)catalog is an online catalog that lists products and services that a supplier is offering for sale. It can take many forms. It can be a simple flat file listing all products and services that the supplier sends to the buyer (who can then maintain it in their catalog system). It can be a database that can be queried. It can be the supplier’s website. It can be a punch-out marketplace. It can be an e-marketplace. It can be a virtual supplier network. As long as it lists item for sale and prices, it can be considered a catalog for the purposes of e-Procurement.

A contract is an agreement between two parties which, if it contains the elements of a valid legal agreement, is enforceable by law. In the procurement realm, a contract is usually between a buyer and a supplier for one or more services, under one or more conditions, at contracted rates. Contracts are generally managed by a contract management system, which may or may not be part of the e-Procurement system as contract management is a key part of the (e-)Sourcing cycle.

At the end of a procurement cycle, and before the next cycle begins, the catalogs and contracts need to be revisited to make sure they are still relevant and up to date. This is because, over the course of time, the following will happen:

  • the number of products purchased not in the catalog will increase

    over time, new needs will arise and replacements will have to be found for items no longer being manufactured

  • the number of items in the catalog no longer purchased will increase

    over time, needs will change and certain items will no longer be required

  • the number of items purchased against a contract will decrease

    as needs change and old items are no longer needed and new items not contracted for are required, the usefulness of contracts will decrease

  • contracts will expire

    and the pricing they contain will have to be renewed, revised, or removed from the system

The next post will outline some of the challenges associated with catalog and content management, some best practices, and some of the benefits that can be expected.

Next Post: Catalogs & Contracts, Part I

Share This on Linked In

Has Coupa Settled on a Coupe? Part III

In our first post, we discussed how, when Davie ran The Coupa Factory, their strategy was innovation focussed and they were constantly charging ahead in their efforts to bring Procurement Independence to the masses but that, lately, it seems that their strategy has shifted to putting customer acquisition first and building a better platform second. In our last post, we reviewed what they have accomplished over the past eighteen months, which isn’t too shabby to say the least (especially compared to some of their peers which do not appear to have innovated at all), but noted that there’s nothing to really shake your foundations … which is a shame considering that had Coupa taken benchmarking and supplier ratings to the next level, they could have knocked your Procurement socks off. This is the subject of this post.

In order for benchmarks to be useful, they have to be meaningful. In order for a comparison to be meaningful, it has to be against like items. And while you can compare apples to oranges, unless you’re comparing the spectra of dried samples in powdered form, it doesn’t make sense. The reality is that savings, request, order, and invoice metrics only make sense if the comparison is against a similar company of a similar size in a similar vertical buying similar products. Consider free-form requests … depending on company size that’s going to range from hundreds per year to tens of thousands per year. Frequency of self-approval … that’s not only going to depend on corporate policies but the types of goods being purchased. If the system is mainly used to purchase office supplies, who’s going to waste time approving every small order? But if the system is being used to buy high priced electronics, different story. PO value will not only vary widely between companies, but within a company. A purchase order for a weekly office supplies order in a small company will be a fraction of a purchase order for a new set of servers. Active suppliers will vary widely depending upon the size of the company and how many different types of products are being bought. Had Coupa defined appropriate verticals, segmented the verticals into appropriate sizes, and insured that the metrics were meaningful (even if that meant waiting until there were more customers in some verticals), this could have been extremely useful. However, right now, it’s interesting at best, and could be dangerous if misunderstood.

In order for supplier rankings to be useful, they have to be against meaningful metrics, and those rankings need to be defined by a majority of affected users. If they are random rankings defined by random users against random products, they are not very useful, especially if they are done by users who have only used the supplier’s products once and not users who have to work with the supplier and its products every day. In order to truly rank a supplier, a company needs to insure that all of the relevant users who use the supplier’s products or services regularly or who interact with the supplier as part of their role rank the supplier. This means that the buyer needs to send out mandatory surveys to these users. While a buyer can easily send out a survey through your standard SIM or e-Negotiation tool, what a buyer normally can’t do through these tools is figure out which organizational users are in the best position to rate the supplier. However, as Coupa enables all spend related to a supplier to be captured in the system, it’s a pretty easy query to figure out which buyers are the biggest user’s of a supplier’s products and which buyers should be ranking the suppliers. If Coupa had enabled the construction of supplier performance surveys which could be sent to the regular users of the supplier’s products in a single click, and then made it impossible for a buyer to requisition anything until the mandatory survey was completed, think of how useful it could be. However, right now, like benchmarking, it’s interesting, but not very useful.

Hopefully these oversights are just the result of Coupa going through the growing pains associated with a brand new management team and rapid customer acquisition. When you consider that The Coupa Sunflower was only starting to blossom, it would be nice to see Coupa return to the days when its releases were much more than coupacetic. After all, why should they settle for a coupe when they can build a dragster? It only takes a little bit of innovation in the right direction to bring back the excitement to Coupa Time.

Share This on Linked In