Source: https://frntir.ai/guides/why-ai-projects-fail

Around 80% of AI projects fail to deliver value, and the causes are people, process and data, not technology. How mid-sized businesses get AI right.

---

Practical Guide

# Why Most AI Projects Fail, and How to Be in the 20%

By **Aidan Dunphy** & **Cyril Le Roux**, Co-founders of Frntir · Last updated: **July 2026** · 10 min read

Most AI projects don't fail because the technology doesn't work. They fail because the organisation around the technology wasn't ready: the data, the processes, and the people. The good news: the failure modes are well understood, and avoiding them is mostly unglamorous, achievable work. Here's what the research says, and how to be in the minority that gets real value.

The short version

- **The failure rate is real.** By some estimates ~80% of AI projects fail to deliver value, and abandonment is rising fast.
- **It's not the algorithms.** Roughly 70% of what makes AI succeed is people and process, only 10% is the model itself.
- **The root causes are boring.** Data quality, unfixed processes, unclear ownership, skills gaps, cost misestimation, and neglected change management.
- **The 20% do the groundwork.** They pick one workflow that matters, fix it before automating, measure a baseline, and scale only what works.

## How often do AI projects actually fail?

By some estimates, more than 80% of AI projects fail to deliver value, roughly twice the failure rate of conventional IT projects (RAND, 2024). And the trend is worsening: the share of companies abandoning most of their AI initiatives before production jumped from 17% to 42% in a single year (S&P Global, 2025).

~80%

of AI projects fail to deliver value, by some estimates, about double the rate of conventional IT projects

RAND, 2024

17% → 42%

jump in one year in the share of companies abandoning most AI initiatives before production

S&P Global, 2025

~95%

of generative-AI pilots deliver no measurable impact on the P&L; only ~5% drive real value

MIT NANDA, 2025

70%

of what makes AI succeed is people and process, only 10% is the algorithms (the 10-20-70 rule)

BCG, 2024

The generative-AI numbers are starker still. MIT's 2025 NANDA study, *The GenAI Divide*, found that around 95% of enterprise generative-AI pilots deliver no measurable impact on the P&L, only about 5% drive real value. Adoption is near-universal (78% of organisations now use AI, per McKinsey, 2025), yet 74% of companies struggle to achieve and scale value from it (BCG, 2024). The gap between using AI and getting value from it has never been wider.

## Why AI projects fail: people and process, not technology

The single most useful finding in the research is BCG's **10-20-70 rule**: success with AI is roughly 10% algorithms, 20% technology and data, and 70% people and process (BCG, 2024). Most organisations invert it (months evaluating models, weeks on change management) and then wonder why adoption stalls.

This reframes the whole question. It isn't "what can we automate?" It's "what outcome matters, and what's actually blocking it?" The companies that get this right pull ahead measurably: BCG's "future-built" firms achieve about 1.7× the revenue growth and 1.6× the margin of laggards (BCG, 2025). They aren't playing a different game: they're playing the same game with better foundations.

The amplification effect

AI doesn't transform your business. It amplifies what's already there. If your processes are clean, AI makes them faster. If your data is consistent, AI makes it useful. But if your processes are tangled, AI tangles them at machine speed; if your data lives in silos, AI hallucinates to fill the gaps. This is why automating a broken process makes it worse, not better.

## The real root causes

Failures cluster around six causes, and only one of them is really about technology. The rest are about data, process, people and money.

| Root cause | What it looks like | Source |
| --- | --- | --- |
| Poor data quality & readiness | Different systems define "customer" differently; reconciliation happens in spreadsheets; nobody owns cross-functional definitions. | Informatica, 2025 |
| Processes never fixed before automating | Steps that exist "because we’ve always done it that way"; the same data re-entered in three places; decisions made inconsistently. | Frntir analysis |
| Unclear ownership & operating model | No one is sure who makes the final call on cross-functional issues; departments guard their own metrics; accountability is diffuse. | Frntir analysis |
| Technical skills gap | The team can run a pilot but can’t integrate, maintain, or govern it in production. | UK Gov, 2025 |
| Cost misestimation | Hidden spend on integration, data prep and infrastructure blows past the plan. | Mavvrik/Benchmarkit, 2025 |
| Neglected change management | The tool ships; people don’t trust it, don’t adopt it, and quietly route around it. | Kyndryl, 2025 |

Two of these deserve a number. Poor data quality is the most-cited blocker: 43% of organisations name data quality or readiness as their single biggest barrier to AI success (Informatica, 2025). And the money rarely behaves: around 85% of organisations misestimate their AI costs by more than 10%, with nearly a quarter off by 50% or more (Mavvrik/Benchmarkit, 2025). Gartner has predicted that roughly 30% of generative-AI projects will be abandoned after proof of concept (Gartner, 2024).

## The pitfalls that sink projects

Beyond the root causes, five recurring traps derail otherwise promising initiatives. Each has a simple antidote, none of which is "buy more technology".

- **AI theatre.** Chasing visible, impressive-looking AI instead of the boring back-office work that actually pays back. (Amazon's "Just Walk Out" checkout, marketed as fully autonomous, reportedly leaned on around 1,000 workers in India to label and validate transactions, and was pulled from Amazon Fresh stores in 2024. Amazon disputes the "watching" framing.) **Antidote:** solve dull problems that cost real money.
- **Technology-first thinking.** Starting from "we need AI" and hunting for a use case. **Antidote:** before any initiative, answer three questions: can we measure success? do we have clean, owned data for this? have we fixed the underlying process? If not all three are "yes", you're not ready.
- **Underinvesting in data.** Spending on models while the data foundation rots. **Antidote:** expect data readiness to take the majority of the effort, not a footnote.
- **Neglecting change management.** 45% of CEOs say most employees are resistant or hostile to AI (Kyndryl, 2025), and around 75% of workers worry AI could make their jobs obsolete (EY). A tool nobody trusts doesn't get used. **Antidote:** involve people early, communicate honestly, train, and embed AI in the tools they already use.
- **Unrealistic timelines.** Expecting quarterly returns from a multi-year change. **Antidote:** set board expectations for one-to-three-year value, with faster wins on narrow use cases.

## How to be in the 20%

The organisations that succeed don't have better algorithms: they do the groundwork first. This is the sequence that separates the 20% from the rest.

1. **Start with a workflow that matters.** Not the most complex or most political one. Pick something contained that costs real time or money and whose outcome you can measure: invoice processing, lead qualification, document review.
2. **Map it end-to-end as it actually works.** Not the official version. Who touches it, which systems are involved, where it gets stuck, where people email spreadsheets because the systems don't talk.
3. **Fix before you automate.** Clean up the workflow first. Once it's clear, the AI part becomes almost obvious: you can see where automation helps and where human judgement still matters.
4. **Define success metrics and a baseline.** What does success look like in numbers, what's the current baseline, and what timeline is realistic? If the answer is "efficiency" with no number, you're not ready.
5. **Run a controlled pilot.** Small team, short window, measure everything, gather feedback, iterate. Expand only when results warrant it.
6. **Scale systematically.** Let operational reality (not a vendor timeline) set the pace.

## How AI colleagues change the odds

If 70% of the problem is people and process, the tool you choose should be one that fits how people already work, not another system to adopt, learn and babysit. That's the design idea behind Frntir's [AI Synths](https://frntir.ai/guides/ai-agents-vs-copilots-vs-chatbots).

A Synth is a named AI colleague with persistent memory that works inside the tools your team already uses, operates with bounded, explainable autonomy under a human boss, and keeps a full audit trail. That directly targets the failure modes above: it lowers the change-management barrier (it lives in your existing email, chat and docs rather than being a new destination), it accumulates institutional knowledge rather than losing it, and its actions stay visible and correctable. It doesn't remove the need to fix your data and processes (nothing does), but it attacks the 70% instead of adding to it. If you want to see what that looks like for a specific job in your business, [book a call](https://frntir.ai/contact).

See it in practice

Vision Meditech, a 25-year manufacturer, hired a Synth named Wallace, and started with one contained, high-value workflow: complex quote replies. Those went from an hour to minutes, and the hardest feasibility answers from three weeks to the same day. [Read the Vision Meditech case study](https://frntir.ai/case-studies/vision-meditech).

## Frequently asked questions

What percentage of AI projects fail? By some estimates more than 80% of AI projects fail to deliver value, roughly double the failure rate of conventional IT projects (RAND, 2024). The share of firms abandoning most of their AI initiatives before production rose from 17% to 42% in a single year (S&P Global Market Intelligence, 2025), and MIT NANDA found about 95% of generative-AI pilots produce no measurable P&L impact (2025).  Why do most AI projects fail? Because the hard part is not the technology. BCG’s 10-20-70 rule holds that AI success is roughly 10% algorithms, 20% technology and data, and 70% people and process (2024). The common root causes are poor data quality, processes that were never fixed before automating, unclear ownership, skills gaps, cost misestimation, and neglected change management.  What is the 10-20-70 rule for AI? A framing from BCG: in a successful AI initiative roughly 10% of the effort is the algorithms, 20% is the technology and data plumbing, and 70% is people and process change. Most organisations invert it (months on model selection, weeks on adoption) and then wonder why nothing sticks.  How do we stop our AI project from failing? Start with one contained, high-value workflow; map how it actually works (not how it is supposed to); fix the broken process before you automate it; define success metrics and a baseline; run a small controlled pilot; and scale only when results warrant. Treat it as an operating-model change, not a software installation.  How long does AI take to deliver ROI? Longer than most boards expect. Narrow use cases such as document processing or triage can pay back in months, but organisation-wide value typically takes one to three years of sustained effort. Be wary of vendors promising quarterly returns.

## Sources

1. RAND Corporation, [The Root Causes of Failure for Artificial Intelligence Projects and How They Can Succeed](https://www.rand.org/pubs/research_reports/RRA2680-1.html) (2024).
2. S&P Global Market Intelligence, *Voice of the Enterprise: AI & Machine Learning* (2025).
3. MIT NANDA, *The GenAI Divide: State of AI in Business* (2025).
4. McKinsey, *The State of AI* (2025).
5. BCG, [Where's the Value in AI?](https://www.bcg.com/publications/2024/wheres-value-in-ai) (2024) and *The Widening AI Value Gap* (2025).
6. Informatica, *CDO Insights* (2025).
7. UK Government (DSIT), *AI Labour Market Survey* (2025).
8. Mavvrik / Benchmarkit, AI cost study (2025).
9. Gartner, generative-AI forecast (2024).
10. Kyndryl, *People Readiness Report* (2025); EY workforce survey.

![Aidan Dunphy](https://frntir.ai/images/Aidan-Dunphy-Profile.webp) ![Cyril Le Roux](https://frntir.ai/images/Cyril-Le-Roux-headshot.webp)

**Aidan Dunphy & Cyril Le Roux** are the co-founders of Frntir.

Aidan has 25+ years in product strategy and technology leadership (B.Sc. Mathematics, Executive MBA). Cyril has 20+ years scaling product organisations, including as VP Product at TransferGo (MBA, The Open University). Frntir builds AI Synths for mid-sized businesses.

## Want to be in the 20%?

Tell us the recurring job you'd start with, and we'll show you how a Synth could own it, inside the tools your team already uses, with the groundwork done properly.

[Book a call](https://frntir.ai/contact)