Discover why AI in IT project delivery is deeper than good prompts — success depends on governance, consistency, and structure, not just the tools.
My friend Gareth is a radio presenter. He has just banned AI from the show prep process after it suggested he broadcast this “fact”: “George Ezra named himself after folk legend Woody Guthrie’s full name for his stage name.”
Gareth replied: “That George Ezra fact doesn’t make sense”
AI fessed: “You’re right, sorry about that! I completely made that up — it’s nonsense! Woody Guthrie’s full name was Woodrow Wilson Guthrie, which has nothing to do with George Ezra at all. Let me fix it”
At this point Gareth decided he was more than capable of making himself sound like an idiot without AI’s help.
Most organisations are talking about AI. Some are experimenting with it. Few are set up to use it well.
A PROMPT FREEBIE
Often, it’s all about the quality of the prompt.
If you fall into this category, Stoneseed has created some AI prompts that might help your project delivery team get the most out of AI. Check them out and share your feedback – we’re all on this journey together. Find the AI Prompt Guides in Stoneseed’s library
BUT sometimes it runs deeper. It is more strategic, operationally nuanced and aligned to the robustness of your governance.
We all know GIGO (Garbage In, Garbage Out), the fundamental computing and mathematics concept that states that, regardless of how accurate a programme or model is, poor-quality input (“garbage”) inevitably produces unreliable, inaccurate output.
This element of GIGO is the quality of the prompt or the question that you ask of AI.
GIGO also emphasises that data quality, governance, and correct input are essential for valid results. What many fail to grasp is, this is especially true with AI.
SPAGHETTI NEST
As a Project Management as a Service (PMaaS) provider, we’ve spent over 15 years coming to the rescue of troubled projects, providing governance, resources, and structure.
These have always been key to project delivery, and now they are vital when embedding AI into your workflow.
The AI age has brought a new set of challenges for Stoneseed to help with.
Increasingly, we are starting PMaaS engagements by helping to unravel a spaghetti nest of AI confusion and unchecked missteps. The Project Management equivalent of thinking that George Ezra was somehow named after Woody Guthrie is rife, only in our world it compounds. It can be a real problem for teams, leading to miscommunication or teams being blown completely off course by AI’s occasional total lack of intelligence, artificial or otherwise.
My colleague points out, “The difference is not tools. It is structure, and consistency when using the tools. It is having a clear, joined-up approach that works across PMOs, project managers and business analysts. Because without that, AI does not scale. It fragments.”
THE REALITY OF AI ADOPTION
In an ideal world, organisations would already have:
- A clear AI strategy aligned to delivery outcomes
- Agreed tools and platforms
- Defined standards for prompts and usage
- Consistent ways of working across teams
Something designed properly. Rolled out properly. Used consistently. Most businesses are not there yet.
Instead, what we see is:
- Individuals experimenting in isolation
- Inconsistent use of tools
- Different standards across teams
- Outputs that vary in quality and reliability
Some of it works well.
Some of it (a lot of it) does not.
And most of it is disconnected.
Let’s be clear, though, and have some compassion. This is not failure. It is simply where many organisations are on their AI journey.
WHY STRATEGY AND GOVERNANCE MATTER
AI in project delivery is not just about saving time. I think it is about creating consistency at scale.
Without governance, AI introduces risk. We’ve seen:
- Inconsistent reporting to senior stakeholders
- Misaligned outputs across projects
- Lack of auditability in decision-making
- Over-reliance on individuals rather than repeatable processes
With the right structure, AI becomes something very different:
- A standard way to produce consistent outputs
- A support layer for governance and reporting
- A way to improve clarity across the portfolio
- A tool that enhances, rather than fragments, delivery
This is where most organisations need to focus first.
Instead of asking, “Which AI tool?”
Try:
- How should AI fit into our delivery model?
- Where does it support governance?
- How do we ensure consistency across teams?
- What role do PMOs, PMs and BAs play in making it work?
>>> Ready to put structured AI prompts to work in your project delivery? Download Stoneseed’s free AI Prompt Libraries for PMs, BAs and PMOs → <<<
WHERE THIS SHOWS UP IN PRACTICE
The gap between theory and reality is most obvious in day-to-day delivery.
Portfolio and Governance
Senior stakeholders expect clear, consistent reporting.
But when each project communicates differently, AI can amplify the problem rather than solve it.
With the right governance in place, AI can:
- Standardise reporting outputs
- Improve clarity across portfolios
- Support better decision-making
Without it, you simply get faster inconsistency.
Communication and Stakeholder Management
AI can help structure communication but tone, judgement and context still matter.
A well-governed approach ensures:
- Consistent messaging
- Appropriate levels of detail
- Alignment with stakeholder expectations
This is where PMs and BAs remain critical.
From Messy Inputs to Structured Outputs
Much of delivery is still about turning unstructured information into something usable.
AI can accelerate that.
But only if there is a clear definition of what “good” looks like.
That definition comes from governance and BA-led thinking, not from the tool itself.
A Joined-Up Delivery Model
At Stoneseed, this is how we approach project delivery.
Not as separate roles or disconnected capabilities.
But as a joined-up model where PMOs, PMs and BAs work together to create structure, clarity and control.
Our Project Management as a Service model is built around providing the right capability at the right time.
- PMOs to define governance and reporting standards
- Project managers to drive delivery and accountability
- Business analysts to create clarity from complexity
That combination is what makes AI usable at scale, because tools on their own do not create consistency. People, structure and standards do.
WHY THIS MATTERS NOW
AI is already being used across delivery teams. The question is no longer whether it will be adopted – it will – it is! The questions to focus on now are whether it will be:
- Structured or fragmented
- Governed or ad hoc
- Scalable or inconsistent
Organisations that focus on strategy and governance now will get significantly more value from AI over time.
Those that do not risk creating more noise, not more clarity.
A PRACTICAL STARTING POINT
Of course, not everything needs to be solved at once.
While wider strategies and governance models are evolving, there is still value in helping delivery teams in the moment.
That is why we have also created a set of practical, copy-and-paste AI prompts for PMOs, project managers and business analysts.
They are not a strategy.
They are not a solution to governance.
They are simply useful tools.
A way to:
- Save time
- Improve clarity
- Handle day-to-day delivery challenges more efficiently
Think of them as a short-term support. A way to reduce friction while the bigger picture is being put in place.
WHEN YOU NEED MORE THAN THAT…
Prompts can help individuals work faster, but they do not create consistency across a portfolio.
They do not define governance and they do not align teams.
If your organisation is looking to move beyond experimentation and towards structured AI adoption, we can help.
Through our PMaaS model, we support organisations in putting the right governance, delivery structure and BA capability in place to make AI work properly.
No hard sell. Let’s just a practical conversation about what you need.
>>> Find out more about our BA services and access the IT Project support you need now! Download the BAaaS brochure — free → <<<
A FINAL THOUGHT
AI is not the hard part. Using it consistently, effectively and at scale is.
So …
Start with strategy.
Build the governance.
And then use the tools.
More about Project Management as a Service from Stoneseed
Notes:
Woodrow Wilson Guthrie (July 14, 1912 – October 3, 1967) was an American singer, songwriter, and composer widely considered one of the most significant figures in American folk music.
George Ezra Barnett (born 7 June 1993) is an English musician, singer-songwriter, and podcaster.
