A spend picture you can actually rely on

The three-way view most spend analysis produces: spend by category, by supplier and by business unit or cost centre. It is called a cube because you can look at any one dimension while holding the others, which is how opportunities become visible.

We will not give you a percentage before seeing your data, and we would be sceptical of anyone who does. Independent benchmarking in this category is remarkably thin, and the confident savings figures on most vendor pages are self-reported. What we will tell you after the first pass is how much of your spend is addressable and where it sits.

Not in the way most people assume. Ardent Partners' benchmarking, which dates from 2017 and is the most recent independent work we could trace, found Best-in-Class organisations saving 6.3 percent against 6.1 percent for others. Almost identical rates. The real gap was reach: 56 percent of spend under management against 39 percent. Visibility does not make you a better negotiator, it gives you more spend to negotiate on.

A standard hierarchical code set for products and services, widely used because it is common ground between systems. It works well for standard categories and less well for anything specific to your industry, which is why many organisations use a hybrid of UNSPSC and their own structure.

Use your own if your categories map to how you actually organise sourcing, which is usually the case. Use UNSPSC where you need to compare against external data or share a structure with a partner. Most engagements end up hybrid, and we agree the structure with you before classifying anything.

We show you before you commit. The free sample uses your real data and reports what proportion we classified confidently, what needed a judgement call and what we could not place. There is no published independent benchmark for classification accuracy in this category, so a vendor accuracy claim is not something you can verify. A sample is.

It gets reported as unclassified, with the reasons. It does not get quietly swept into an 'other' bucket to make the percentage look better. Unclassifiable spend is usually a data quality problem worth knowing about in itself.

Typically a few weeks, depending on how many source systems are involved and how much supplier name normalisation the data needs. The free sample gives you a realistic timeline before you commit, because data quality varies enormously.

Quarterly suits most mid-market organisations. Annual is common and usually too slow, because the picture is already six months old by the time anyone acts on it. There is no independent benchmark on refresh frequency, so treat any specific claim about it with caution.

Only if you want a permanent live dashboard and have someone to keep feeding it. A tool automates classification but leaves you the cleansing, the tail of messy supplier names and the exceptions. If you just need a reliable picture on a cycle, the analyst work matters more than the software.

SAP, Oracle, NetSuite, Microsoft Dynamics, Coupa, Sage, Xero, QuickBooks and Tally, plus corporate card statements and AP extracts. Format is rarely the obstacle. Data quality is.

Twelve months of transaction data, a supplier master extract, and your contract list if you have one. That is enough for the free sample, and it usually takes a finance analyst under an hour to produce.

Access is scoped by you, granted to named people, and revoked through your own leaver process. We contract on standard clauses, name every sub-processor, and work to GDPR, UK GDPR, PIPEDA, the Australian Privacy Principles and Saudi PDPL. ISO 27001 alignment is in progress and SOC 2 Type II is on the roadmap. We state status honestly rather than showing badges we do not hold.

Usually one of three things: consolidating suppliers where the same category is bought in several places, bringing off-contract spend onto contract, or running a sourcing exercise on a category the analysis surfaced. The analysis is the map, not the journey.

Yes, and it is often the fastest way to prove value. One entity, one region or one category, classified properly, tells you whether the wider exercise is worth doing.