A Category Strategy That Sticks

I’m Ben Shute and I’m joined by Paul Rogers, and we’re discussing how to develop a category strategy that sticks. How do you get from a spend analysis to actionable category deliverables? In other words, how do we perform the ‘voodoo’ in between them?

This is a 5-part series in which we discuss the tips, tricks, tools and techniques featured in our latest training course “from data to insights...

There are a number of ways to come up with a category strategy and in this first episode we discuss the right things to measure.

We’ll let you know when we publish the next in the series if you join our email list or follow us on Facebook.

So, Paul, when we perform a spend analysis, and we try to go from a data set to a number of actionable insights that drive our category strategy, what are the things that we should be measuring?

I think there’s a big seven, Ben. The big seven are the seven dimensions that will normally cover the majority of variations in an organisation’s category spend.

Okay, great. What are the seven?

Well, let’s run through them. So, the first one is the number of variants used. And by variants, I mean the number of solutions used to meet the same need. When people create a spend tree or a category taxonomy, they’re investigating the different ways they can solve the same problem.

I’ll give you an example. When we classify the spend on petrol, what might be an adjacent way of meeting the same needs? The answer is to explore the spend on diesel or LPG.

Right – they’re all different ways of meeting the same need, which is fuel to move a vehicle.

Precisely.

So, the first step is mapping all the possible pathways. What’s next?

The second dimension is the value of that spend. And the subsidiary question we’d have to ask is how the value of that spend is made up.

Let’s say our total spend is $100,000. Is that $100,000 on one variant? Is it $5,000 on twenty variants? Typically, what you'll find is the majority of spend is on a couple of different variants. And then there's some outliers that are used elsewhere in the organization. So, understanding the total spend requires us to understand the aggregate sum, and also a number of variants that make up the total spend.

We have the variants and we have the make-up of the spend itself. What’s third?

Well, the classic spend cube also looks at the number of suppliers with whom we spend money. Organizations that are mature will often have high levels of compliance to the procurement solutions. And by high that might be 80%, or sometimes 90%. But even that means that 10% of the spend is not against the contracted solutions.

So how many different active suppliers do we have? You and I have worked in organizations where there'd be lots of one-time vendors that are still on the system. So I think we need to distinguish between active vendors – which are suppliers with whom we’re currently spending money in the current financial year or accounting period, for that particular sub-category – and how that spend is distributed between active suppliers.

Would you use a Pareto Analysis for that? I've come across a client that had a panel for a particular category, and they had more than a hundred suppliers on the panel, of which more than half got no business at all. So, if you did your Pareto Analysis, you would learn something valuable, which is that the spend is concentrated in a small number of suppliers.

But, just as importantly, you’d learn that there are lots of suppliers on the panel who get no business! This means clearly the current strategy is not working. Why have them on the panel if you're not going to use them?

You're quite right! I think Pareto Analysis is a good place to start if leverage is your strategy. But understanding why the spend is distributed the way that it is requires us to understand the suppliers with whom we spend nothing, just as much as the suppliers with whom we spend 80% of the total.

Okay, so you've got variants, spend and suppliers. What's the fourth dimension?

Well, the next one would be the number of distinct buyers. And there's a distinction here between users and economic buyers. We consider the economic buyer as the person who owns the budget, or who makes choices about whether we buy the red one or the blue one, whereas users are the transactional people who raise the orders.

In many organizations there may be a power user who would be dominant, but there'll be other users as well. The reason why this is helpful is if we're doing a stakeholder analysis we want to know who the economic buyers in our organizations are, and how many different ordering points there are. If we want to capture the spend, we need to understand who is actually doing the ordering.

But often, Paul, if you are doing a spend analysis, the economic buyer may not appear in a spend diagnostic.

An interesting view, because you would know who's raising the PO and what the delivery points or the performance points are. That's all you know. But my observation is that, having identified who's raising the orders in terms of the stakeholder analysis, we also need to understand who it is that we need to engage with to discuss, for example, variety reduction or the terms of any subsequent arrangement. So that requires us to helicopter up organizationally in order that we understand who the economic buyer is.

Alright, so that gives us four dimensions. I guess now we would be a good time to talk about procurement methods?

Indeed, the next dimension is procurement methods. I once worked with an organization that raised something like 50,000 purchase orders a year, and the invoice value of each of those purchase orders was less than $100!

So, how do we buy?

Many organizations use procurement cards for lower value purchases, so it would be interesting to know how many transactions we are paying without a PO and just paying against invoice.

Are we using procurement cards? Are we using a purchase order? How are we actually buying? And, more to the point, how much of our spend is not subject to any purchasing arrangement at all?

Well, that speaks to the compliance levels in spend management in an organisation.

We just talked about 80 or 90% compliance being a good result. There's still 10% to 20% of spend which is not covered by a procurement solution, which can be informative. Why are people not using existing procurement solutions?

The default assumption is that we need to police these users and force them to use our solution, but perhaps there are good reasons why they don't use our solution. We need to understand why. Why are they not using our solution?

So our sixth dimension is the frequency and the timing of any purchases.

In supply chain analysis, there's a framework called runners, repeaters and strangers. A runner would be something like photocopying paper that you would buy regularly, is predictable and your consumption is pretty standard.

A repeater would be something that you might order every six weeks, and sometimes you order 100, some months 200, some months zero, but it's ordered regularly throughout the year.

And a stranger would be something like a replacement part for a machine or something that you acquired periodically, but not on a recurring basis. This framework helps with understanding the nature of your demand. If you spend $100,000 annually on a category, is that made up of 100 orders of $1000? Or is it made up of one order of $100,000?

That's a really good point. If we're going to look at timing and frequency, do we look at the last three months, the last 12 months, or do we have a broader view, and maybe pick 24 to 36 months?

Wow! 36 months would be a big data set, Ben! But your point is well made that there might be a trend in the usage that is not visible if you only look at three month's data.

Most organizations will have categories that fit into those three different classes of runners, repeaters, and strangers. Matching demand with supply is our basic challenge, and the solution for runners, (which might be a standing order, or just in time) will be very different to the solution for strangers. A stranger might be a spare part for a critical machine, for example. And you might choose to hold that spare in stock just in case, or have a supplier on contract with a short lead time. So quite different solutions, depending on the nature of our demand.

And I suppose if we're thinking about strategies, getting our head around methods of procurement is important as well.

Absolutely !

Great. So, we’ve made it to the final dimension, and I’m going to stun absolutely nobody when I suggest that it’s pricing.

Spot on, Ben. We wouldn't be procurement people if we weren't concerned with pricing, would we? At some stage, we might ask what pricing levels the stakeholders are paying. A good example would be professional services, Ben, an area that you and I are familiar with!

When we look at the rates per hour, we may find that stakeholders are paying different rates for the same service. And the question anybody would reasonably ask is: why are some users paying a price of 100 and some 110, and some 90? Sometimes from the same source!

Understanding the variations in pricing, and the opportunities they present, is important.

Okay, so that gives us the seven dimensions, and it’s quite clear from listening to you that they're intertwined. So, how do you actually utilise the dimensions together?

Well, more information doesn’t necessarily mean a clearer course of action. What would you say if I asked what the dominant strategy that procurement people apply in 80% of cases is?

Let's reduce the number of varieties, consolidate volume, reduce the supply base, issue an RFP to three or four suppliers, shortlist the lowest two, negotiate with them and either award a panel or give it all to one supplier.

So, if that's your solution, what's the point in engaging in this level of granular analysis if you plan to 'back fit' that solution to every circumstance?

Are you saying that, potentially, a lot of people start with the strategy and actually work backwards to back-fit the strategy to the data?

Yes, I think that's an excellent point. And it's a cookie cutter solution, where the same solution is applied to all categories, whether it's warranted or not. And I think part of what the workshop that we're running will do is explore how you recognise when it is not appropriate to back-fit that cookie-cutter solution.

I'll give an example. Let's say that many organisations have a socially responsible procurement strategy or a corporate social responsibility strategy. It could be that some of the lowest value suppliers are actually local, or indigenous-owned businesses that we're trying to nurture and support. So, reducing them and cutting them out of the supply base would actually be contrary to the organization's overarching strategy. You can't always apply a cookie cutter strategy. You need to say, well, which are the 20% of circumstances that are different and need customized, tailored strategies?

Okay, Paul. Well, I'm looking forward to the course. And this is really the first of several episodes to come. Some of the many things you said I'd like to pick up on is asking the right questions to your internal stakeholders. Specifically, the point you've made on who is actually an economic buyer. And also constructing a great hypothesis. I think that would be an awesome session to have because, on one hand, we go into the analysis with our hypothesis or our strategy, but what would be the benefit of actually looking at the spend data first before we back-fit the solution?

A neat framework is to say that a 'top down' strategy is where you start with a strategy and then apply it to the data. Whereas sometimes we need a 'bottom up' strategy, where we look at the facts and ask what strategy is most appropriate. I think lots of stakeholders struggle with that because, as you've said, there are seven questions there, each of which have a variety of answers. And what the next step should be doesn’t jump off the page at you.

What we will address in the workshop is how to use those insights to develop a better procurement strategy.

Fantastic, Paul, thanks for your time. I look forward to the next session!

And remember readers, we’ll let you know when we publish the next in the series if you join our email list or follow us on Facebook.