Go beyond AI adoptionSee where work happens, andhighlight the impact.

Most companies have adopted AI across multiple teams and tools. Very few can tell you what's actually working. Persist connects the dots between your AI investments and real business outcomes — giving you the visibility to make smarter decisions about where to invest next.

Get Included
The Problem

The gap between AI adoption and proving its value is only getting wider.

40%

Over 40% of agentic AI projects are projected to be canceled by 2027 — as enterprises pour resources in with no visibility into what's working.

Gartner, 2025↗
6%

Only 6% of organizations report significant EBIT impact from their AI investments. The other 94%? No measurable bottom-line return.

McKinsey, 2025↗
29%

Only 29% of executives can measure AI ROI with confidence — despite 79% perceiving productivity gains. The gap between belief and evidence continues.

IBM, 2025↗

AI is not failing. We are just not measuring it right

What Persist Does

Measure, observe, and orchestrate your AI stack

The IndexAvailable

Your AI impact, scored and benchmarked across 6 dimensions

Connects to your platforms through API connectors and quantitative surveys. The result: a scored, benchmarked report that answers "is our AI investment actually working?" with data, not opinions.
SalesforceSlackMakeHubSpotCopilotNotionmore
The ObservatorySoon

A live registry of every AI tool running across your stack

Auto-discovery and manual registration for every AI tool. Health monitoring beyond uptime: drift detection, dependency mapping, and coverage analysis.To be shaped with founding partners
The ConsoleSoon

One command bar for your entire AI operations

Propose changes as diffs, approve or reject, deploy with full audit trails. Progressive autonomy that grows to full orchestration as trust builds.To be shaped with founding partners
AI Index Score
64of 100
51
Avg
37
P25
59
P75
Security52Adoption80Throughput70Environment35Impact74Quality75
Your orgComparable avg
Percentile BenchmarksAgainst a custom comparable set — matched by size, sector & geography
P25P50P75Top
Security
52avg 58
P44
Adoption
80avg 52
P78
Throughput
70avg 48
P72
Environment
35avg 40
P45
Impact
74avg 45
P79
Quality
75avg 50
P75
4 agents2 workflows1 automation
NameModelTypeDriftDepsCoverageStatus
Salesforce Einstein
Salesforce Einsteingpt-4.1AgentStable3
92%
Active now
Intercom Fin
Intercom Finclaude-4AgentStable2
88%
2 min ago
Make.com Pipeline
Make.com PipelinecustomWorkflowDrifting5
64%
14 min ago
M365 Copilot
M365 Copilotgpt-4.1AgentStable4
95%
6 min ago
Zapier Sync
Zapier Sync—AutoCritical6
31%
Failed · 1hInspect
Error Log
Connection timeoutCritical
Rate limit exceeded (3x)Warning
Payload mismatchWarning
14 errors in last 24h
Dependencies
Intercom→Zapier→Slack
HubSpot→Zapier→Sheets
6 downstream services affected
Coverage Gap
31% coverage · 5 endpoints unmonitored
Notion AI
Notion AIclaude-4AgentStable1
82%
22 min ago
HubSpot Workflow
HubSpot WorkflowcustomWorkflowDrifting3
71%
38 min ago
/Type a command or ask a question...|
⌘K
Suggested
ControlPause Salesforce Einstein across all channelsControl

Suspends all active sessions without data loss. Queued interactions route to fallback.

- status: "active" // running on 3 channels
+ status: "paused" // graceful shutdown, queue held
DiagnosticsShow error logs for Zapier Sync — last 24hDiagnostics

Surfaces errors grouped by type with frequency and severity.

- errors: unknown // no visibility
+ errors: 14 (3 critical, 11 warn) // last 24h
AnalyticsCompare Copilot usage this week vs lastAnalytics

Side-by-side delta of tasks completed, time saved, and error rate.

- tasks_completed: 847 // last week
+ tasks_completed: 1,203 // this week (+42%)
OrchestrationRoute Intercom tickets to backup workflowOrchestration

Redirects incoming tickets to backup. Original agent stays online for non-ticket tasks.

- routing: "intercom-fin" // primary
+ routing: "backup-workflow" // failover active
Natural language or slash commands · Changes proposed as diffs, you approve before deploy
How the Index Works
Sara Maldon, Co-Founder of Persist
“2 years leading AI transformation at Make, I was obsessed with how you actually measure the impact of AI on a business — beyond adoption numbers. No methodology existed. So I built one from the ground up, together with practitioners and operators who'd lived the same problem. That framework is what powers the Persist AI Index — now being validated with researchers and academics.”
Sara Maldon·Co-Founder, Persist
01

Connect your platforms

Secure API integrations pull quantitative usage data from every AI tool your teams use. Zero disruption, full visibility.

NotionZapierChatGPTCopilotIntercomSnowflakeZendeskand many more
Syncing 0 data points
02

Survey your teams

Short, structured surveys capture the qualitative side -- how teams actually feel about AI, what barriers they face, and where they see impact.

AI Index Survey
8 of 12
How often do you use AI tools in your daily work?
NeverAlways
Rate the impact of AI on your team's output quality.
No impactTransformative
How well has your organisation prepared you to use AI effectively?
Not at allVery well prepared
What makes it hard to tell whether AI is actually improving your team's work?
Honestly I wouldn't know. Some people use different tools, but nobody really measures if it's increasing productivity or not.
Team response rate
84%
03

Score & benchmark

Quantitative data meets qualitative insight. Your organization is scored across 6 dimensions and benchmarked against industry percentiles.

Security52Adoption80Throughput70Environment35Impact74Quality75
Percentile Benchmarksvs. comparable peers
P25
P50
P75
TOP
Security
52
P44
Adoption
80
P78
Throughput
70
P72
Environment
35
P45
Impact
74
P79
Quality
75
P75
Your org
Comparable avg
Board-ready report generatedPDF + Live Dashboard
04

See your team heatmap

Instantly spot which teams lead and which lag. Color-coded scores across every dimension reveal exactly where to focus your next investment.

TeamAdoptionImpactQualityThroughputSecurityEnvironmentAvg
Engineering
88+19
82+13
79+6
85+17
72+4
64+5
78
Sales
74+5
68-1
71-2
76+8
55-13
48-11
65
Support
62-7
71+2
80+7
64-4
68+0
52-7
66
Marketing
80+11
76+7
65-8
72+4
50-18
45-14
65
Operations
55-14
60-9
72-1
58-10
78+10
70+11
66
Finance
55-14
58-11
68-5
52-16
82+14
75+16
65
Avg69697368685968
Strong above avg
Above avg
Neutral
Below avg
Critical
05

Track the trend — and act on it

See whether AI spend is translating to results, quarter over quarter. Spot what's working, cut what isn't, and know exactly where to invest next.

Quarterly Trend
64+8 ptsAI INDEX
£18.4k+14%MONTHLY SPEND
120+2%HEADCOUNT
+36%QoQTASKS /MO
Q2 '25Q3 '25Q4 '25Q1 '26
AI Index score
AI spend (normalised)
Automated tasks (% growth)
Q2 '25Q3 '25Q4 '25Q1 '26
AI Index40485664+8
Spend /mo£9.2k£12.8k£16.1k£18.4k+14%
Headcount112115118120+2%
Automated Tasks /mo—+51%+46%+36%+36%

Efficiency signal detected

Automated tasks grew +200% over four quarters while spend increased +100%. As a proxy, output is scaling ~2x faster than cost.

Founding Cohort

Founding Cohort Programme

We work alongside your team to deliver a full AI index assessment — and a clear path to act on it.

4 of 10 spots remaining60% filled

What You Get

Answer
  • “Is our AI investment working?”
  • AI Index report scored across 6 dimensions — understand how you compare to your benchmark
  • Team-level heatmap: who leads, who lags, where to focus
Action
  • Prioritised suggestions: where to double down, where to cut
  • Visibility across your entire AI stack in one view
  • Quarterly re-scoring to track progress over time
Access
  • Early access to EU & UK AI Impact Report benchmarks
  • Priority access to Observatory and Console as they ship
  • Dedicated support — we embed with your team from day one

10 design partners. Hands-on. Results-driven.

Book a 20-Minute Discovery Call
FAQ

Questions we get asked

We help companies figure out if their AI investments are actually working. One score, benchmarked against peers, broken down by team.

Yes. We’re working with our founding cohort of companies now.

A scored report your board can read, a heatmap showing which teams lead and which lag, visibility across your entire AI stack, and quarterly re-scoring to track progress. The score combines quantitative platform data with qualitative team surveys.

Read-only access to your AI tools plus a short survey to your teams. We see metadata, never content. Never code. Never anything about individual employees.

First real results in 6 weeks.

Companies spending on AI across multiple teams who want clarity on where to double down and where to cut.

Ten companies getting early access to benchmarking data, plus inclusion in the EU & UK AI Impact Report we’re publishing this year.