Microsoft AI Study Reveals Productivity Gains
Key Points
- The Microsoft study shows non‑technical workers using Copilot cut email volume by 11% and boost document throughput by roughly 10%, shifting more time into Word, Excel, and PowerPoint.
- Technical roles report less immediate behavior change and instead highlight AI’s potential, with 44% seeing value in automated test generation and 37% in documentation rather than full code‑writing assistance.
- Customer‑success and sales professionals feel more fulfilled, likely because Copilot automates repetitive language tasks that previously dominated their workflows.
- Users report a noticeable drop in mental effort (30/100 vs. 55/100), though the study notes it cannot yet quantify long‑term productivity or wellness impacts and calls for further research.
Full Transcript
# Microsoft AI Study Reveals Productivity Gains **Source:** [https://www.youtube.com/watch?v=T58_gs7mSRE](https://www.youtube.com/watch?v=T58_gs7mSRE) **Duration:** 00:06:30 ## Summary - The Microsoft study shows non‑technical workers using Copilot cut email volume by 11% and boost document throughput by roughly 10%, shifting more time into Word, Excel, and PowerPoint. - Technical roles report less immediate behavior change and instead highlight AI’s potential, with 44% seeing value in automated test generation and 37% in documentation rather than full code‑writing assistance. - Customer‑success and sales professionals feel more fulfilled, likely because Copilot automates repetitive language tasks that previously dominated their workflows. - Users report a noticeable drop in mental effort (30/100 vs. 55/100), though the study notes it cannot yet quantify long‑term productivity or wellness impacts and calls for further research. ## Sections - [00:00:00](https://www.youtube.com/watch?v=T58_gs7mSRE&t=0s) **AI Boosts Knowledge Work, Not Code** - Microsoft’s study shows AI co‑pilot lifts productivity for non‑technical employees while developers see modest gains, mainly in test generation and documentation rather than full code writing. ## Full Transcript
all right real quick Microsoft released
a study on AI in the workplace and we're
going to get into it I want to cover
some of the big stats here across job
families and then I want to get into
what what we should think about it so
number one if you are someone who is in
a non-te role then on average it's it's
influencing a double digit change in
where you spend your time so people who
were using Microsoft co-pilot read 11%
fewer emails on average and they were
handling documents and had document
throughput that was about 10% higher so
they're putting that time back into word
excel and
PowerPoint now if you're in a tech role
it's a little bit different there the
focus is less on the actual observed
change in behavior and it's more about
the potential and that's really
interesting to me because it suggests
that developers actually adopting GitHub
co-pilot are behind the curve relative
to knowledge workers adopting co-pilot
for knowledge tasks and that doesn't
surprise me because technical tasks in
writing code is a more complex nuanced
and precise art form and it's not
something I would expect co-pilot to get
to right away in this case developers
are highlighting potential in a couple
of areas potential mind you 44% are
reporting that generating tests is an
area where AI can help 37% see ai's
potential for documentation what you see
here is that they're not focused on AI
as code writer which has been something
that people have been made a lot of hype
out of but you don't actually see in the
survey results moving on to customer
success which is sort of a very customer
facing role you see that there is a lot
of agreement from surveyed CS
professionals that co-pilot is making
them feel more fulfilled I am
hypothesizing that this is because
co-pilot is picking up some of the
repetitive language tasks that they
would otherwise have to do you see a
similar effect with sales professionals
who also have a lot of repetitive
language tasks where co-pilot or another
AI tool would be really
effective I want to call out that one of
the benefits that we are not seeing come
through here is the general effect on
Wellness longterm of a reported decrease
in cognitive load and frankly that's not
because the study didn't think about
that it's because we just haven't had
enough time the study did call out that
using co-pilot is reported to be less
mentally demanding than just doing it
yourself substantially it was like a
score of 30 out of 100 on mentally
demanding I don't quite know what that
means versus 55 the point is it's less
so I've now gone through sort of a quick
study you can actually go and get this
uh this study and read it yourself I'll
link it underneath the YouTube here so
everyone can go and find it and read it
for themselves the thing that I want to
call out
is the stud study
itself is not something that we can
really use to understand actual
productivity in the workplace and the
study knows that in fact they call out
as one area for further Discovery
further research how AI is affecting
team
productivity because it's actually not
clear when you look at overall team
productivity that we're seeing the gains
that we would see when individuals
report processing more documents and
that kind of makes sense
because if the documents were busy work
to begin with doing more of it doesn't
really add value if the emails were busy
work to begin with doing less of it
doesn't add value the point is that real
team productivity gains have to be
associated with the team being able to
do more meaningful work and in Tech
that's usually finding a way to ship
value to customers that's what really
matters and this study basically calls
out that we don't have any idea yet of
how a is actually helping us ship value
to
customers and that is concerning frankly
and it's not concerning in the sense
that I worry that we won't have an
impact of AI on this because if if AI is
being adopted as widely as it's being
adopted I think it was up to the study
said 78% of respondents had used AI at
least
once we're getting to a point where AI
is being tried in a lot of different
places and
I would expect that eventually we are
going to find applications in use cases
where we see effects at the team level
simply because of the number of people
who are experimenting with this large
general purpose
technology I do not know where that team
impact is going to come from and I think
that there are real question marks
around some of the things we have
traditionally considered work that could
affect and Cloud that team picture so
let me give you an
example if I have traditionally been
responsible as a manager to write a
report every week about what my team is
doing and if the llm can do an
equivalent job on that report in one/
100th of the time yes I have saved
time it is not quite as clear if I have
added value other than saving time
because at the end of the day what is
the opportunity cost of that time what
am I doing as a manager now that I
wasn't doing
before and that gets at the Team
Dynamics piece the question really
becomes at that point is there a
strategic context that an individual AI
user is operating within that allows
them to leverage the time they gain back
to really Drive effective value for the
business I think that is one of the
biggest questions we have in Tech right
now and I don't think anybody has a good
answer to it I I think in some ways AI
is exposing that we do a lot of busy
work in Tech and in other white collar
industries and that we haven't had a
strategic context to operate within that
would allow us to truly leverage more
time in the workplace to generate more
value I'm curious what your take is I
know we're going to be getting more
studies on this I think that Team level
productivity is going to be a really hot
area for AI in
2025 uh let me know what you think in
the comments