Tipping point: Agile procurement is in a precarious position as waterfall looms large

Written by Dominic Campbell on 22 May 2017 in Features

Government procurement has reached a fork in the road, says FutureGov founder Dominic Campbell. Will it continue on the path to agile, and start buying services based on trust and fairness, or pivot back to waterfall?

Waterfall - on the way back? - Photo credit: Lino Mirgeler/DPA/PA Images

Back when I worked in government, I ran a number of classic waterfall projects, including one where the budget for a £10m technology system cost the taxpayer £1m just for the procurement process alone.

We’ve come a long way since then - the five years since the Government Digital Service was launched have seen a real shift in attitude towards IT in government - but we’re still only just moving beyond our baby steps.

One of the biggest problems for government is finding the right skills for their ambitious digital agenda: while there are plenty of talented people in government, there remains a mismatch between those skills and the skills needed to make digital happen - and at scale.

This isn’t unique to government. Digital and design skills remain rare while demand for those skills is growing exponentially. To fill in the gaps, government relies on suppliers - but as the skillset in central government has increased, we are experiencing a ‘missing middle’.

This is where departments are less willing to take on whole supplier teams to deliver; instead choosing to body-shop individuals from suppliers, an approach that ignores the benefits of bringing in the wider ways of working and thinking that comes from those suppliers.

It’s also a model that can do more harm to SMEs than it does to help them to grow, reducing them more to recruitment agencies than delivery partners.

And it’s really challenging for the truly creative, modern digital agencies to fit into that model. Those companies want some autonomy, some authority; to sit alongside the client and jointly own responsibility for delivering the project. They want to make the most of the skills and perspectives that both insiders and outsiders can bring to designing solutions.

However, in much of government there remains a culture of ‘fixed price, fixed outcome’ when it comes to procurement and - perversely - Agile has exacerbated the problem, because we now are facing a proliferation of ‘fixed price, flexible scope’ contracts: the worst of both worlds for suppliers.

In an Agile setting, there isn’t the same structure as the old enterprise approach: there tends not to be a piece of paper (‘change control’) that authorises changes to an upfront (best guess) scope. This is because Agile works best when a detailed scope is not pre-determined, but developed through on-going insight and agreement, with decisions on project priorities made in small increments on a regular (often bi-weekly) basis.

Yes this requires a level of involvement from the client that is traditionally not required, but the benefits are huge - everything from transparency and flexibility to de-risking the approach through regular user testing.

But time and again, suppliers - like FutureGov and others - are finding themselves in a position where Agile remains embedded in a hostile environment.

One where even if during a project the delivery team has a truly Agile approach, procurement and project governance teams still hold the power. Upfront specification is still sought (if not entirely achieved) and by the end of a project there is conflict on definitions of delivery.

We might have agreed on an amount to achieve an outcome - but the supplier’s outcome might be very different to what the client had in their mind’s eye. The supplier hasn’t contravened a contract, but the client may be left wondering how the project reached the conclusion it did. And the governance suddenly becomes very complicated.

The problem is less acute for internal projects: an organisation recognises that it’s going to spend, say, £2m on its staff, to do £2m worth of work, but in this case the money is treated as fluid; their own resources are not costed in the same way as those of suppliers’, where departments still want to know what the day rate is as well as seeking a fixed price fixed outcome.

‘Don’t let waterfall dig its heels in’

The fundamental disconnect between those two cultures is putting the Agile agenda at risk for government - the partnership between internal and external teams remains crucial to the success of digital projects in government. It is vital we work fast to develop a shared way of working that brings the most benefit from both.

Some suppliers are developing ways to get around this: I know one that is creating contractual pledges for each two-week sprint, because that’s a controllable amount of time, and comes with a promise you can make, with a known price you can account for.

It might be incredibly onerous, but if we could find a lightweight way to make this work, it could help us move closer to Agile, at least while people become comfortable with this relatively new way of working.

Because if we don’t keep working towards a time when we procure for trust - where clients and suppliers work together as honest partners on a project, where procurement judges past record, culture and fit as much as day rate and fixed outcomes - we risk opening the door for out-dated waterfall methodologies to dig their heels in and reemerge (despite 70% of IT projects failing this way).

Trust is hard to generate from nothing, especially when clients are being encouraged to work with new, innovative suppliers - but there are ways to build it up. Procurement portals could allow government employees to rate suppliers, or organisations could ask potential suppliers to come in and run a workshop to choose what sort of Agile culture they want.

And I wouldn’t even mind having punitive clauses for non-delivery if it buys us the flexibility of trust - that’s almost worthwhile. Just as long as there is a two-way contract in place where clients also pledge the necessary time and resources needed to own decisions through an Agile process.

Of course, ‘failure’ is hard to define, and harder to prove. So at the same time, we need to develop better ways of measuring outcomes and success, and governments need to be ready to change their attitude and culture.

Some of this is already happening; but we’re coming to a point where there’s a lot of risk, and that puts Agile in waterfall world in a precarious position.

Until we’re wall-to-wall Agile, we can’t learn its true beneficial impact - but if we don’t get swing the balance soon, it will crash under the waves of waterfall.

Dominic Campbell is the founder and managing director of FutureGov

Share this page


Add new comment

Related Articles

Related Sponsored Articles

How to quantify cyber risk
15 March 2018

BT's Malcolm Stokes explains how organisations can attribute accurate figures to cyber risks in order to make a viable business case.

Cyber security is one of the greatest man-made challenges of our time
6 March 2018

BT's Ben Azvine argues that the frequency and impact of breaches is increasing and we need to continuously adapt and innovate to stay ahead of the threat environment

Who keeps your organisation secure?
19 February 2018

BT's Amy Lemberger argues that having the right security in place to protect your organisation is no longer just an option. It is a necessity.