Every company knows what they sell.

Sales departments know, finance know, even the CEO knows the top level numbers.

Compare that with how many companies truly know and understand what they’re buying.

It’s shockingly few. And yet, especially in many manufacturing industries, goods and services associated with the cost of production can account for over 50% of a company’s turnover.

Saving 5% on the cost of production is a lot easier than a sales force increasing their revenue by the necessary amount to account for a 5% improvement to a company’s bottom line.

Why a solid spend analysis of what you’re buying is essential

The risks of not having a sound knowledge of:

• exactly what you’re buying;
• from whom;
• and at what quantities and unit prices,

are substantial. The dangers present in such a data-starved procurement function or buying organisation are:

• Inefficient use of employees’ time, resulting in them managing too many low value, non-critical “long tail” suppliers
• No overview of single-sourced vendors, and consequently no risk management strategy in place to mitigate supply risks
• Not knowing where the potential low hanging fruit is to get some quick hits through competitive tenders on high spend categories which are easy to implement.
• No clear understanding of the extent of maverick spend – i.e. the % of total spend which is not being managed by a procurement professional.

It’s not easy for a procurement manager to do a spend analysis from scratch.

Poor data is usually the culprit.

The data often makes it difficult to get meaningful, actionable numbers.

But fear not, we have to start somewhere.

I’ll assume for the purpose of this article that you’re not using a software solution, although this inevitably makes it way easier!

The first steps to analysing your spend

The best starting point for putting together a simple spend analysis is usually a mixture between:

• POs issued to external vendors
• Data from your Finance department based on actual invoice values paid to vendors

Typically, a business will trust the latter rather than the former.

Let’s therefore start this process by obtaining a report from Controlling showing what the organisation has paid to each external vendor who has submitted at least one invoice.

This should be easy to do for any organisation using an ERP system. You don’t need a top-of-the-range software like SAP or Oracle to pull basic data.

Ask your Financial Controller to provide you with a report containing the sum of all invoices paid to external vendors over the past 12 months, or previous calendar year if you prefer.

This will give you a broad overview and enable you to filter in descending order your largest to smallest vendors by spend.

While this is a great start, this only tells you half the story.

You’re likely only going to have a list of vendors and the total amount paid to each of them at this stage.

It’s unlikely to tell you the detail you’ll need to make a meaningful evaluation of your vendor base…but it’s a start…

The 4 most basic categories of spend

Drilling down into this level of detail will usually require delving into PO data and liaising with internal stakeholders.

Usually, Procurement is the gatekeeper of PO data, in one form or another. The functions placing or requisitioning these POs may not necessarily report into a procurement organisation i.e. engineering stores, marketing assistants, IT project managers.

However, PO data in most cases will be owned or accessed by Procurement. In smaller businesses which don’t have a procurement function, there is usually a an admin assistant somewhere within the organisation who “owns” the process of issuing POs in the system.

Note: If you don’t have a formal purchase order process, your problems are more basic. This is definitely the first fire that needs extinguishing, and quickly!

The 4 key areas of data you’ll need to drill down on will include:

• How much of your spend is made up of direct materials for the production process? (typically repeatable spend, found in manufacturing businesses)
• How much spend is made up of indirect materials (typically one-time or infrequent spend, like office supplies, furniture)?
• How much of your spend is services (often under contract)?
• How much of your spend is capital expenditure (usually large sums of money on machinery, construction, real estate etc, executed once)?

This is the point in the process where you’ll probably get scuppered by poor data.

There’s no easy solution to correcting and cleaning historical PO data I’m afraid.

You’re just going to have to manage as best you can.

Your ERP systems will tell you who raised the purchase requisition or PO.

Any high spend POs which cannot be deciphered, usually due to poor PO text descriptions or a mish-mash of goods and services on the same order, will have to be clarified with the person who originally requested it.

The first spend analysis you perform will unlikely yield perfect data but that’s OK. Having something is a good place to begin to build on the data in future, and to put countermeasures in place to turn poor quality data into valuable data.

Tagging your vendors

The final step of a basic spend analysis is to tag the vendors: this breaks down the elephant into bite-sized pieces.

My suggestion to keep things simple is to use just 2 separate categories of tags:
Firstly, you should put these vendors into buckets based on whether they are:

Critical

The vendor cannot be replaced because there is no other possible source.

Typically, this will be because they are either an OEM equipment vendor, hold a trademark or intellectual property which prevents a competitive tender, or due to geographical reasons are the only authorised distributor of a certain product or are the only supplier locally who is able to provide a service critical to maintaining operations.

Preferred

The vendor could theoretically be replaced but it will be costly or risky to do so.

They don’t hold a monopoly position but extracting yourself from the existing vendor relationship will be painful, and certainly not possible without a detailed exit strategy.

Examples include electrical maintenance suppliers who know a production site better than your own maintenance technicians, an equipment vendor whose machinery is the legacy equipment installed in a production plant, or customs brokers and clearing agents – especially in frontier markets, or perhaps an outsourced facilities management provider for your corporate offices.

Transactional

The vendor can be relatively easily replaced without major issues beyond the usual teething problems associated with on-boarding a new vendor.

Some common examples include generic spare parts vendors, office supplies, security services, payroll administration, transportation vendors.

While there may be some resistance from internal business partners who don’t wish to change supplier, there is no overarching business risk why the vendor should not be replaced.

This basic analysis enables you to understand how many vendors are indispensable, as well as how many could theoretically be replaced over time with a proper strategy.

Most importantly though, it tells you how many vendors are non-essential and thus easily replaced. That’s where your lowest hanging savings are going to come from.

More interesting is how much spend does each of these categories encompass?

You will probably need to liaise with key stakeholders to understand exactly what each vendor is supplying and how these vendors are viewed by the people who requisition the goods and services. They will be able to give you accurate feedback of what tag to apply to each vendor.

Just be sure that you clearly explain the definitions of each tag.

A stakeholder will rarely voluntarily identify one of their favourite suppliers whom they have a close relationship with as being expendable.

You may need find that you need to challenge some of these definitions further down the line, but this is fine for now. The devil of the detail can come later as your refine your strategy.

How much of the spend does Procurement actually control?

 Next comes the critical initial yardstick to understand how much of this spend is actually being controlled at present:

Managed by Procurement

Procurement manages the relationship with these vendors. There are varying degrees to what extent procurement is involved in these vendor relationships. This can be split out into sub-categories later. Don’t worry about this too much for now.

For now, tag these as any vendor where procurement has been involved in a negotiation on pricing or contractual terms.

These can also include legacy vendors who were not originally selected by procurement or those where the day-to-day vendor relationship is led by engineering, marketing, HR etc.

Maverick spend

These are any vendors where the majority of the spend has not been touched by procurement during the sourcing process, or the RFQ was conducted by somebody in another function in the business.

If procurement has been involved just in the administrative process of executing the PO, this must not considered “procurement involvement”.

This is simply performing an administrative task on behalf of the stakeholder who is requisitioning the goods or services.

For all sense and purposes, this is non-compliant or non-controlled spend, depending on the terminology that your organisation uses.

Obviously, not all maverick spend is equal. Things like batteries and A4 paper are less serious concerns than having production machinery or professional services not being negotiated by a procurement professional.

What do you know, and what is still missing?

 This is sufficient to perform a basic spend analysis to have a general idea of where your priorities should lie.

After you’ve tagged suppliers based on these simple criteria, you will then be able to place your vendors into buckets which will prioritise how you tackle them.

High spend, critical suppliers who are tagged as maverick spend will be your biggest area of concern.

If the “keys to your castle” are not being managed by procurement professionals, this is likely to be low hanging fruit, or high levels of risk lurking out there, depending on which way you look at it.

What you’re most likely not able to do after this first, basic analysis is to fully understand the consequences of what this data is telling you.

You may potentially be able to identify some easy quick wins you can deliver in terms of cost reduction.

However, it is unlikely that you’ll be able to identify process improvements which can be made to remove the amount of non-value added activity your buyers are undertaking to manage low spend, transactional vendors.

We delve into this in my subsequent article.