## An introduction to Sourcing Optimization

### Strategic sourcing is slow…

Strategic sourcing is a procurement process that continuously improves the purchasing process within a company. However, despite the vast technological advances made in IT over the past 20 years, strategic sourcing itself is still slow. Most modern procurement teams still spend large amounts of their time battling with Excel, manually cleaning bid data and analyzing bids by hand in order to find the most cost-effective bidder. It is therefore of no surprise that this type of manual work is a recipe for disaster – with errors and frustration being the main ingredients. Indeed, a recent Deloitte survey showed that 62% of CPOs said that they can’t trust their teams to execute. This means that more than half of CPOs do not have sufficient faith in their procurement teams to find the most cost-effective suppliers!

With all of the 21st century’s technological advances, this begs the question: do things really need to stay this way? At Keelvar we say: NO! That’s why we bet on **sourcing optimization**** as **one of the best ways to reduce spending and increase the efficiency of your strategic sourcing process.

### …so what is sourcing optimization?

If sourcing optimization is one of our top bets for cost and time savings, then what exactly is it? Well, *optimization* is defined as *“the action of making the best use of a situation or resource”*. Therefore, within the context of procurement, this means that sourcing optimization allows you to find the most cost-effective supplier or carrier, given a set of resources, cost, quality or time restrictions.

For example, as part of an ocean freight event, we may need to find the cheapest carrier. But by doing so, you may also want to factor in not just the quoted carrier costs, but also other factors, such as transportation time, whether or not carriers are incumbents and whether or not the carriers satisfy certain qualitative conditions. You may also have a maximum upper bound on the number of carriers that a given port can support, therefore requiring you to find the *find a sweet spot between cost and the number of carriers*.

Doing all the necessary calculations for this manually is very time consuming, if not impossible for large sourcing events. Each restriction might generate thousands of possible combinations that would need to be analyzed. Mistakes might creep in and, and due to the large amounts of data and constraints, not finding an optimal supplier will lead to wasted resources and a higher spend.

This is where sourcing optimization comes into play. The idea behind the development of the Keelvar Sourcing Optimizer is simple: we wanted a programme that allows you to quickly and easily define your restrictions using a set of simple, understandable rules. It was important to us that these rules could be created by anybody – on the spot – without needing to be a programmer or Excel whiz. Once you define these rules, the optimizer will take these rules along with all the bids, discounts and possibly qualitative data that is in the system, and automatically calculate the optimal solution for you. This way, using the optimizer, you will be able to find the optimal carriers quickly and with minimum effort. Therefore, using sourcing optimization allows for:

- Increased savings potential
- Increased collaboration and transparency
- Reduced cycle times
- More strategic insight & control
- Increased adoption rates

#### Sourcing optimization in practice: Finding the cheapest carrier using the “Low Cost Baseline” scenario

Talking about how sourcing optimization will make your life easier sounds fantastic in theory. But what does this look like in practice? Let’s find out by looking at a simple, real-world, example using the Keelvar Sourcing Optimizer: Imagine that we are running an ocean freight event, and that we want to find the cheapest carrier, using merely the prices that they quoted. In other words, we will use the sourcing optimizer to find the cheapest carrier per lane, without any additional restrictions.

Our example assumes a standard ocean freight event design which has already been entered into the Keelvar Sourcing Optimizer. The event consists of the following columns:

- A lane identifier column that uniquely identifies a given lane.
- A business unit column, which identifies the business unit that will be served by the given lane.
- An origin and destination column, which indicates the country from which the shipment originates, and the country to which the shipment is destined.
- An equipment type column which indicates the type of equipment to be shipped.
- Two columns which indicate shipment requirements: Direct Required (i.e. whether a direct or indirect shipment is required) and Max Transit Time (the maximum permissible time that it can take for the cargo to be shipped).
- 7 columns whose cells are to be completed by the carriers. These are our bidder input columns.

Carriers have also already submitted their bids. So all that is left for you to do is use the sourcing optimizer to find the lowest-cost supplier. To do so, you need to use something called a “scenario”. A scenario is basically just a grouping of rules that the optimizer will use to translate and calculate the optimal results. By default, your events will have one empty scenario created automatically. This scenario is called the *“Low cost baseline”* scenario, and its sole purpose is to calculate the cheapest supplier per lane.

Let’s go ahead and “evaluate” this scenario. Pressing “evaluate” basically tells the optimizer to calculate a result. And that’s it! Once evaluated, all you need to do is click on the scenario and you will be able to see exactly which carrier won which lane (see figure 3) below.

#### Too easy? How about limiting the total number of suppliers?

Granted, the previous example of finding the cheapest carrier was fairly straight-forward. The Keelvar Sourcing Optimizer had already created the low-cost baseline scenario for you, and all you had to do was press “evaluate”. You will argue that the “real world” isn’t that simple, that you have constraints and restrictions to consider and that just finding the cheapest supplier per lane can still be done fairly easily in Microsoft Excel. And you’re right! So that’s why we should create a second scenario, in which we take your constraints into account.

In this second scenario will assume that you have certain limits as to how many carriers you can service at a given time. For example, your ports or loading bays, may have a capacity restriction that makes it impossible for you to support more than two carriers. We may therefore want to find the cheapest carrier, but under the condition that we can support at most two carriers at a given moment in time.

Returning to the Keelvar platform, we must therefore add a new scenario to our ocean freight event. After having created a new scenario by simply pressing the “Add a scenario” button, we want to add a “Limit winners” rule (figure 4). This *“limit winners rule”* presents you with three primary choices:

– which bidders to apply the rule to,

– which lots to apply the rule to, and

– the number of permitted winners.

The rule may be applied to all bidders, or you may choose specific bidders, or bidders from a pre-defined bidder group. Next you must select which lots the rule will apply to, which may be all lots, a specific subset, or a pre-defined lot-group. Finally, you may limit the award to be at least, at most, or exactly a specific value. Within the context of this example, we want to configure the rule to “limit the number of winner for all bidders on all lots to at most 2”, as shown in figure 4 above.

And once again, that was it. Once the rule has been added, your new scenario is ready for evaluation. All that you need to do next is press “evaluate” and observe the results.

#### …mitigating risks using the “Limit winners” rule

If the previous example was still too easy for you, then how about we consider mitigating risks? Using the “Limit winners” rule, you can mitigate some risk from individual suppliers by enforcing that we have at least a certain number of suppliers. For example, to ensure we have at least two winners by each origin-destination pairing we may configure a rule as follows:

Note that the “Grouped by” dropdown shown in figure 5 above allows you to group lanes according to the selected columns. In this case, for each unique pairing of origin and destination the rule will enforce that we have two suppliers across those lots.

### Better to light a candle than to curse the darkness.

Often the impression is given that you must adopt (all at once) an all encompassing suite of solutions that will do everything for you. At Keelvar we believe the opposite.

This guide is designed to show you how your sourcing process can be easily transformed to become efficient, effective and reliable.

Start small, start sustainably, learn to apply the basics. Incrementalism is not a mistake in terms of adopting a strategy for the long term.

For this reason, we offer the option of a per event licence, with the option to move to an unlimited events/ time based licence when you’re ready.

However, hopefully the short examples shown in this article were sufficient to illustrate an alternative to manual, Excel-based bid evaluation and serve to show how sourcing can be made more effective. Maybe your organization will improve their sourcing process in 2019 through sourcing optimization, or maybe they will improve the process via other means. Whatever your approach, remember that if your strategic sourcing strategy is slow and painful, then it is time for change. As the old saying goes: *“better to light a candle than to curse the darkness.”*

### Want to see the Sourcing Optimizer in action?

… schedule a demo with one of our Keelvar experts.

**About the author**

**Benjamin Jakobus** is a software engineer at Keelvar, specializing in sourcing optimization and automation.