Thought leadership Technology ERP
The inconvenient truth about enterprise software

Why Procurement Software Implementations Fail — and What We Did Differently

By Bundle IQ Research ·8 min read ·Bundle IQ Insights

There is a statistic that nobody in enterprise software likes to advertise. Depending on which study you read, somewhere between 55% and 75% of ERP implementations are considered failures — not outright disasters necessarily, but projects that came in over budget, over time, under-delivered on the business case, or were quietly abandoned after go-live. SAP, Oracle, Coupa, Jaggaer: none of them are immune. The failure rate is industry-wide, it is well-documented, and yet organisations continue to spend millions pursuing the same approach.

I've spent my career advising organisations on procurement and supply chain transformation. I've seen what good looks like, what bad looks like, and — most importantly — why the gap between them almost never comes down to the software itself.

The problem was never the software. It was the assumption that software could fix a process problem.— Bundle IQ

What actually goes wrong

When a large-scale procurement implementation fails, the post-mortem almost always identifies the same culprits. Not in every case, and not always in the same order, but the pattern is remarkably consistent.

The data isn't ready

Every major procurement platform requires clean, structured spend data to function properly. Category trees. Supplier master data. Contract records. Budget hierarchies. The assumption at the start of most implementation projects is that this data exists and can be migrated. It almost never does. The average FTSE 250 company has supplier data scattered across three or four ERPs, a handful of legacy procurement tools, and more spreadsheets than anyone has counted. Cleaning that data — standardising supplier names, normalising categories, resolving duplicates — takes months and is almost always underestimated. Projects that plan six weeks for data migration routinely take six months. That's before a single user has logged in.

The process comes second

Good procurement software should systematise a good procurement process. What actually happens, in most implementations, is that the software is configured first and the process redesign happens around it. Teams spend months configuring approval workflows, category structures, and supplier portals — and only discover at go-live that nobody has agreed on how the business will actually use them. The technology becomes a constraint rather than an enabler. Users find workarounds. The adoption rate falls. The business case evaporates.

It solves the wrong problem

Enterprise procurement software is designed for organisations with an existing procurement function. It automates, streamlines, and provides visibility into a process that already exists. What it cannot do — and was never designed to do — is provide the expertise to run that process. You can give a company SAP Ariba and they will still need a team of procurement professionals to use it effectively. The software is the tool. The expertise is still your problem.

This is the central issue for the vast majority of UK businesses. They don't have a procurement process that needs automating. They have no procurement process at all. And no amount of software solves that.

The economics don't stack up below £50M spend

The implementation cost of an enterprise S2P platform — including licensing, implementation partner fees, internal resource, training, and the inevitable scope creep — typically runs between £150,000 and £500,000. That's before ongoing licence fees of £50,000–200,000 per year. To generate a return on that investment, you need to be managing a procurement spend large enough that a realistic percentage saving covers the cost. The maths only works above roughly £50M in annual indirect spend. For the 99.7% of UK businesses that spend less than that, the economics are simply not there.

What we did differently at Bundle IQ

When I started designing Bundle IQ, I didn't ask "how do we build a procurement platform?" I asked "why do procurement platforms fail, and what would we have to be true to avoid those reasons?" The answers shaped almost every architectural decision we made.

No implementation. You're live in hours.

Bundle IQ has no implementation phase. There is no data migration, no configuration workshop, no change management programme, no go-live date. You submit your first requirement — in plain English, from a form that takes five minutes — and you are live. The platform doesn't require your historical spend data, your supplier master, or your budget structure to exist before you can use it. It builds that knowledge as you use it, not before.

The expertise is built in, not bolted on

Every brief submitted through Bundle IQ is automatically benchmarked against real market data. The tender document is generated by AI with category-specific templates built by procurement specialists. Vendor responses are scored against the specification criteria without human intervention. The contract is generated from legally-reviewed templates appropriate to the category and value. At no point does any of this require you to have a procurement professional on your team. The expertise that large organisations pay £54,576 a year to hire is embedded in the platform itself.

You pay when it works

The fundamental misalignment in traditional procurement software is the payment model. You pay the licence fee whether or not the platform is used. You pay the implementation cost whether or not the business case is achieved. You pay the annual renewal whether or not anyone logs in. Bundle IQ inverts this. Buyers pay nothing. Vendors pay a success fee only when they win work. The platform only generates revenue when it generates value. That alignment changes everything about how the platform is designed — because our incentive is always to make it work, not to make it complex.

Process first, technology always

This is perhaps the most important difference. Bundle IQ doesn't configure to your existing process — it gives you a best-practice procurement process from the moment you start. The intake form is built on category-specific templates designed by procurement professionals. The evaluation criteria are structured against CIPS best practice. The contract templates reflect current legal standards. You don't need to have a process before you start. The process is the platform.

A word on what Bundle IQ can't do

This wouldn't be an honest piece without acknowledging the limits. If you are a large organisation with complex multi-entity procurement, global supply chains, and ERP integration requirements across SAP and Oracle — you need enterprise software. Bundle IQ is not SAP Ariba and doesn't try to be. The platforms that failed the organisations I described above were right for very large organisations used in the wrong context. Bundle IQ is right for every organisation that those platforms cannot serve — which is most of them.

The honest answer is that for businesses under £50M in annual indirect spend — which is the overwhelming majority of UK businesses — the implementation model of enterprise procurement software is fundamentally broken. Bundle IQ was built from that observation. Not as a criticism of the tools that exist, but as an acknowledgement that a different model was needed for a different market.

See how Bundle IQ works in practice

Submit your first requirement in five minutes. No implementation. No configuration. Just procurement that works.

Submit a brief →