Howdy wizards,
Some personal news: I've started doing part-time work with NewsCatcher — who also happen to be today's sponsor. I've been a fan and user of their web search API (CatchAll) for a few months, so it was an easy yes. I made a little demo of their product showing how to use it for monitoring M&A and fundraising in any market. It’s just one use case, but a good way to see it in action.
Now, here's what's brewing in AI.
How fast do companies ship an AI use case?
Of the 2,000+ case studies of AI implementations in Context Windows, 135 currently have data on how long deployment took.
The average company takes 6 weeks from project start to when the AI solution is actually in production.
Here’s how long some popular use cases take to deploy on average:
A few non-obvious things to note:
AI implementations bought off-the-shelf from a SaaS ship almost 50% faster than building custom ones.
The most effective + fast AI projects are when companies use turnkey platforms that have one well-scoped use case. When there’s a solid SaaS available, procurement beats engineering.
Scope matters: the same use case can vary 100x in deployment speed. E.g. a multi-regional rollout to dozens of offices takes longer than something intended for a single team.
There might be lots of benefits to partnering with a system integrator (Accenture, Deloitte, McKinsey, IBM, etc) to build the AI solution — but speed doesn’t seem to be one of them. Companies who had a partner build the implementation took as long as those who built in-house.
Document Drafting is an outlier in terms of deployment time, mostly because it’s a very popular use case in regulated industries. Some examples: Acentra Health (Medicare appeal letters, 6 months), UCB Pharma (clinical research support, 6 months), Siemens (industrial field-service reporting, 6 months).
—
I use CatchAll to find new AI implementations from companies.
With their Company Watchlist feature, I can continually monitor the web for a specific list of companies, and when something is published about them deploying AI, I get structured data back. The use cases of this feature are many.
Check it out:
IN PARTNERSHIP WITH NEWSCATCHER
If you're monitoring a portfolio, a supplier network, or any other curated list of companies, CatchAll's Company Watchlist is built for you.
Upload your companies once. Then attach the list to any search job: product recalls, data breaches, or anything else you can describe. CatchAll figures out what events to look for and returns every match it can find. What you get back is a structured dataset of events scoped to your list, with details like what happened, when it was reported, citations, and a relevance score per match.
Useful for M&A surveillance, supply chain risk, portfolio monitoring, counterparty tracking, or anything else worth watching across 2B+ web pages.
❦
Are GenAI implementations increasingly making a difference on the bottom line?
Looking at case studies across the last 8 quarters, there’s a slow rise in implementations that report specific EBIT/ROI, revenue gains or cost savings.
The trend is currently quite weak. There’s an increase but it’s modest.
And it’s not super clear why there’s an increase — is it because companies are getting better results from AI, or is it because they’re getting better at measuring it?
In any case, knowing that close to 20% of companies are now seeing real P&L movement from their GenAI projects is still useful, and shows it’s definitely not all smoke & mirrors.
❦
📮 Have a question about AI implementations you’d like me to answer with Context Windows data? I’m currently taking suggestions! Just hit reply.
Also — if you’re getting steady value from this newsletter, your team probably would too. Appreciate all of you who share it with friends & colleagues 🤝
You are a delight.
Dario
What's your verdict on today's email?
Know which AI use cases are paying off — with Context Windows Pro
Most companies pick AI use cases by brainstorming internally. 90% of those initiatives fail.
I’ve created Context Windows just so you can pick the winners.
🟦 Subscribe to Pro to find high-performing use cases by industry, from 2,000+ companies. Visit contextwindows.ai or book a demo with me.
Disclosure: To cover the cost of my email software and the time I spend writing this newsletter, I sometimes work with sponsors and may earn a commission if you buy something through a link in here. If you choose to click, subscribe, or buy through any of them, THANK YOU – it will make it possible for me to continue to do this.