10 Gemini Productivity Hacks That Make ChatGPT Obsolete - YouTube

https://www.youtube.com/watch?v=gMchbSDsCwU

Transcript:
(00:00) Chat GPT is really good at sounding like your best friend. I got so much
more done thanks to Gemini because it's built into the apps that I'm using every
single day. And there's plenty of new AI features that nobody has been paying
attention to that are freaking amazing. And in this video, I'm going to show you
exact workflows and exact prompts that you can copy to get more done thanks to AI
everyday.
(00:21) If you don't know me, my name is Gail Breton and I'm one of the co-founders
of Authority Hacker and System Stack and we help companies learn and leverage AI
and automation to get more done every day. But for now, let's get started. Now,
we're starting in Google Meet because I know a lot of you guys are already using AI
note-taking tools that allow you to essentially just get the cliff notes of
whatever happened in your calls.
(00:42) But a lot of people are paying for it. They're using tools like Auto.AI.
And this costs 17 bucks per month if you pay monthly or 30 if you're in a team.
It's literally like half the price of Google Workspace. And Google Workspace Gemini
Noteaker that you can find on the top right of your screen here.
(00:58) and you just click on the start taking note button is actually really good.
I'm going to show you an example in a second and I'll make the argument it's more
useful than these tools like auto.ai because you'll be able to use the context of
your call including everything you've said the entire transcript for AI uses later.
I had this real call with a real potential client the other day and Gemini just
took all the notes for me and if I open it you can see I have a summary.
(01:21) I have all the detail of what was discussed and then the next steps
including a full transcript on the second tab. But the point is nobody really reads
these things. However, inside Gmail or any other Google app, I can use the context
of the transcript of the call and the notes taken by Gemini to compose emails to
documents etc.
(01:38) So a real example would be that Mark wasn't on this call with Zafir is
here. So I could just compose an email say that I want to write to Mark open the
Gemini chat on the right. Select my email replies chatbot that I've already
configured to write like me and I'll show you how to do that in a minute. And then
I can just paste my prompt here, which is very short and easy.
(01:58) Draft a short and sweet email to Mark explaining what happened and what's
next for this client. And then I'm going to tag the actual meeting transcript. So
I'm going to say at Zafiris. I'm going to press enter and enter again. And Gemini
is going to take the entire context of the call and write me a short email saying,
"Hey Mark, quick update.
(02:17) Zafiris, he was looking into his scraping architecture information, but I
pushed for API access blah blah blah blah blah blah." And I can just click on
insert. And then I can just add my subject line and send it to Mark. And just like
that, I can keep anyone I want updated based on my meeting notes without having to
write any complex prompt because Gemini has the entire context of everything.
(02:34) Now, when you read your emails, you can also ask Gmail to actually
summarize your emails here by just clicking the button and you'll see they'll be
summarized here. But also, you have the quick reply thing with AI, which is really
handy. So, for example, this person is asking for guest personal site, which I get
about 50 a day. They're really annoying.
(02:50) So I could be like clicking here help me write option that I could click on
and then I'm going to say tell them no and create and Gemini is going to make a
polite answer much more polite than I would have and could insert at send and
that's how AI can help me just essentially get rid of useless emails. Now one more
thing that you can do is using this chat on the right by clicking the little star
on the top you can actually ask questions about your overall inbox and potentially
important things that you have missed. So, for
(03:17) example, I can ask what are emails I should urgently reply to right now.
It's going to scan my inbox, find some emails potentially need to reply to. So, you
can see I got an email from SCM Rush and I got some guest post definitely not very
important, but there is some that could be interesting and I can actually see the
emails here, click on them and actually just go and check them right away.
(03:39) So, that's really handy if your inbox is getting flooded with a lot of
emails. AI can help you find the ones that are actually useful. But it's not just
Gmail that got a big blow up. Actually, Google Sheets is one of those that most
people are ignoring. That's super powerful. So, actually, here is a spreadsheet
that I should probably not share with a bunch of our affiliate commissions over
time and how much money we've made from each affiliate program.
(03:58) I'm not going to scroll through because I don't want to give away all the
data, but it's a bit annoying to go through all that data and analyze it yourself.
So now I can click on the little Gemini star on top and I can ask questions about
my data just in natural language when I can say so for example I can ask what
affiliate programs are declining the fastest right now and what it's going to do is
it's actually going to run some proper analysis and run some formulas and do some
reasoning actually so you get a good model here and it's
(04:24) telling me exactly which affiliate programs are decreasing the most now
which is flying press sharell SCM rush p matters WP rocket which is a lot faster
faster than actually having to go through all of this or even res resorting the
spreadsheet or anything like that. But you can also ask Gemini to visualize the
data for you.
(04:42) So I can ask to make me a graph of revenue per affiliate program of all
time. And you can see that it just made me that graph. And if I wanted, I could
actually just insert it. So I could click here, it would make me a new tab. And you
can see some of these. I'm not going to give you exactly which ones, but you can
see some of them.
(04:59) But I can actually visualize all that data for me if I want to, which is
really, really handy. But wait, there is more. Let's go back to our previous tab.
You can see that we have a category that's WordPress, funnels, etc. And that's
something that we had to do manually because AI didn't exist when we started this
spreadsheet.
(05:12) It's pretty old. But now I can actually use the new AI function to have AI
automatically categorize companies for me so that I can create chunks and see which
categories make me the most money. So let me show you how this works. I can just
press equal and then I can type AI and then I open a parenthesis. Then I write my
prompt in quote.
(05:32) So I'll say categorize this company in one of the following categories only
output category categories to choose from WordPress affiliate funnels SEO MISK and
then I just close my quotes I press comma and then I select the company name. So
I'm going to select the first one here which is A2 hosting and I'm going to close
my parenthesis and now you can see I have this little hover thing.
(05:55) I can click on it and click generate and insert. So you can see it came up
with affiliate which I don't fully agree with but let's assume it did a good job.
The point is then you can take this and just drag it down and then categorize every
single company if you want. So you can see I have this little hover thing and I'm
going to click on refresh and insert and you can see that now it's actually trying
to categorize all these companies.
(06:15) And it's not too bad. It's not perfect. For smaller companies I guess it
might struggle a bit but the point is you can have AI mass process data for free
for you in Google Sheet right now. But wait, there is more. Because this was AI
filling data, but you could have AI write formulas for you so that it calculates
things properly.
(06:32) If you click on the little star here, you can actually let's reset the chat
and you can ask it about a formula that you want to do. So I have this prompt here
that says create a formula that finds the name column A of the highest earning
program column B. And if I press enter, Gemini actually understands in Google
Sheets and is going to write a formula for me.
(06:51) And so you can see a formula I could not write myself, but I can just click
on the insert button, click close, and you can see it came up with the name of the
effect program that supposedly made us the most money. We are rich. So you can see
AI can really do a lot for you inside spreadsheet. But I know a lot of you are
probably using Google Docs more often.
(07:09) So let me show you some really cool tricks that you can do inside Google
Docs with Gemini now. So let's open a new document. And if you click on file and
then new, you will see you have this new help me create. And let's go back to a
real life situation here. Let's imagine I had a call with my client and I need to
prepare a proposal for them.
(07:26) It's pretty annoying to prepare proposals. It can take time. So, we're
going to have AI prepare the proposal for us. And then we're just going to put the
final touches on it. What I'm going to do is I'm going to click on file, then new,
then help me create. And here you see you can essentially prompt AI to build a
document that you want.
(07:41) But where it gets really powerful is that you can reference other documents
to help you create the document that you want. So the prompt says let's write a
proposal for Doug based on and then here I'm going to press at and I'm going to
search for my call with Doug and I'm going to select it and now the AI is going to
have the context of the transcript of the call that I had with Doug talking about
the project that he wants to work with us on.
(08:04) Then I wrote the prompt saying the tone of this the proposal should match
G's tone in the call which is going to be inside the document. We will do the
following. One, research technical implementation to document the process and
estimate the cost and present it to him. Three, build upon his approval. And four,
offer 30-day maintenance in case anything breaks. And a total will be $6,000.
(08:23) Remind him of the value of the process for his business. And the direct CTA
is to respond to this email and we'll draft a scope of work contract. And if you
just click on the create button, AI should take everything that we gave it, which
is the context of the call, the prompt, but also my tone of voice and everything.
(08:39) And you can see in just a few seconds we now have a proposal that is
written for us that takes a lot of the details from the call explaining exactly
what he wants. So he talks about NA10 and HTTP calls etc like stuff that we talked
about on the call and what you can do is just have the finishing touches.
(08:56) So for example, I think this is a little bit long. So we could just
highlight this. I could click on refine and I could just say shorten. And now I can
click on replace. And now you can see the intro is a little bit shorter. You can
see that the AI also followed the prompt explaining exactly what we're going to do
with the exact steps I gave it. It told it reminds him of the value.
(09:15) It's a bit long maybe. So again, I could probably shorten this. Yep,
perfect. Except there's a few extra bullets, but that's fine. It's easy to fix. And
it gave the right price as well. Now, what if I want to add an extra section here?
Do I need to actually write it? No, not really.
(09:30) All I can do is I can click on my Gemini on the right. And again, I could
pick one of the gems that I have already. I'm just going to pick my email replies
one here because I don't have one for proposals, but I would probably make one. And
I say add a section that explains that the payment should be half up front and half
on delivery.
(09:48) Only write that section. But what I can do is I can refer back again to
essentially the transcript of the call. So AI has all the context that it needs. So
I'm going to say at Doug find the call again. And in principle, AI should be
writing a section that I can directly add. So permanent terms 50% up from payment
is required to initiate the project.
(10:08) The remaining 50% will be due upon successful completion. And then what I
can do is I can go wherever I want to go in document. So maybe after investment
here and I press insert and you will see that this section has now been added. And
you will see that now I've created my proposal. Now if I don't like this image on
top, I can also reprompt that.
(10:24) So I can click generate image and maybe I'll say something about newsletter
because that's what he wanted to some help with. Now I pick a style. Let's say
sketch. And now you can see we have a bit of a newsletter graphic for example that
I can put here if I want to position it and that maybe like this. And that's it.
(10:43) I have my beautiful proposal that I can send to my client and it didn't
take me any work really. One more thing that is not necessarily very useful for
proposals in this case but that you might want to know about Google Docs is if you
go back to the initial window of the chat you have this summary of content that
gives you a summary of what happened but you can insert that inside your document.
(11:01) And this block is going to automatically update as the document gets
updated, which can be very handy for very big document. And did you notice one
thing since the beginning of this video? We've been using Gemini, but we haven't
even opened the Gemini app yet. You can do a bunch of stuff with it.
(11:15) And I think now it's time to finally open it and show you some really cool
tricks that you can do with it. Let me actually switch to the Flash model initially
because I want to show you some quick stuff. And I like the Flash model because
it's fast. It's the equivalent of 40 on GPT. So, I know a lot of people are
teaching you how to build fancy AI agents to add calendar invites thanks to N810
and spend days and days doing this.
(11:34) But did you know that if you use Gemini, you can just add calendar and say
add a planning meeting with Mark at 8:00 p.m. And because this is Gemini and is
connected to your Google environment, it's created that meeting and I can just
click on it and edit it in my calendar if I want to. And this works with pretty
much every Google app.
(11:54) If you type at, you will see that you have Google calendar, Google Keep,
Google Task, Gmail, Google Docs, Google Drive, and everything in workspace. And
this is particularly handy for the next trick I'm going to show you. If I open
Gemini 2.5 Pro because I want a bit of a smarter model here, and I add Gmail, I can
ask it to read my recent email replies and prepare a comprehensive prompt to teach
an AI agent to write like me, include examples, and follow prompt engineering best
practices.
(12:19) And if I press enter, because Gemini is connected to Gmail, it can go in my
inbox, it can read my emails. And that prompt is going to be used inside a custom
gem that we can call inside Gmail to write replies for us. So you can see my prompt
is coming back. And you have a decent guide with a bunch of examples. Now you'll be
asking like where do these examples come from? Let's let it finish to reply.
(12:44) But you will see, for example, that the examples are actually links to my
real emails. you see all the sources that it actually used. And so that's a pretty
personal guide here. So what I'm going to do is I'm going to cut this prompt and
we're going to go in the sidebar and we're going to go into explore gems. And here
we're going to create a new gem.
(13:04) And here we're just going to paste that prompt. And we're going to call it
email replier. You don't need to do anything else. You can click save. And now the
gem has been created. Now, you could chat with it from the chatbot, but that's not
very handy. What's very handy is to go back to Gmail, open an email that you have
that you need to reply to.
(13:24) Click on the Gemini icon, go into gems, and find the gem that you just
created. And now I can say reply to this email and say, "I can't come." But the
thing is, this is not just Gemini writing. It's Gemini trained on my style. And you
can see that the reply is like short and sweet, the same way I would write.
(13:42) Thank you so much for the invite spotlight. Unfortunately, I won't be able
to make it this year. Best girl. And then I could just click reply, insert, and
then send. And that's really showing you the power of this ecosystem. Now, I also
really like the Gemini app to like plan content and big projects before I actually
get started.
(14:00) And I'm going to show you how I do that as well. Let's imagine that we're
planning a podcast on cloud code and v coding because I've been spending way too
much time using cloud code right now. It's literally walking on my laptop next to
me. So what we're going to do is we're going to first launch a deep search be like
I want to create a podcast on code find me advanced tips and tricks for it.
(14:25) I'm going to let Gemini go and research a bunch of that stuff and meanwhile
I'm going to show you how to do some advanced brainstorming. Okay, the research
plan came back and I clicked on start research and now Gemini is working. So we can
start a new chat again. And this time I'm going to use the canvas and I'm going to
show you.
(14:40) I'm going to put a bit of a bigger prompt inside Gemini here. And this job
is really to one poke holes into the things that I want to talk about in the
podcast. So I can just like blab and talk. And it's going to organize our chat into
tidy notes on the right that it's going to update as we chat.
(14:56) So I'm going to press enter so you will see exactly how this works. So you
can see it's reasoning. It's a reasoning model. Jamite 2.5 Pro is not super fast,
but it's super smart. And usually the waiting time is worth it. So you can see that
I told it that I want to make that podcast on VIP coding and it's going to poke
back against me and it's telling me what the listener might feel when they actually
see the episode.
(15:14) So they're like, "Oh, you and I know VIP coding, but as a brand new
listener who saw the title, my first thought are, am I smart enough for this
episode? Is this some highle developer talk that's going to go over my head?" And
then it challenges me. It says, "How can we restructure the very beginning of the
episode to lead with a jaw-dropping story?" And within 60 seconds, the listeners
know exactly what's in it for them.
(15:34) And then I would just reply like I would reply to someone I'm talking Mark
for example. So the story is really the fact that I'm not a developer. I don't know
how to code. I'm pretty technical, but I don't know how to code. And I was able to
build an app that I paid $12,000 for in three days. And it's live on the internet
and people can use it and it's working.
(15:55) And really I think I want to highlight the paradigm shift on the fact that
people can now build things on the internet unlike maybe even a year ago because AI
is so good right now. And so like my transcript would go through to Gemini and what
Gemini is going to do is it going to understand what I said update the structure on
the right and then eventually push back on me again and be like okay that works but
maybe this doesn't work.
(16:20) How do we perform a deeper dive into a segment that crushes the self-doubt?
What is the very first most concrete least intimidating step a complete beginner
could take to experience the magic themselves? Well, I think the first step they
could take is actually using the Gemini app and the canvas function. You can
describe an idea that you have for a very simple tool and it will build it for you.
(16:40) And what's really cool is the Gemini canvas app allows you to connect it to
actual AI API. So you can build a small AI app just by describing it. and it
requires not even seeing the code. So I think that's the first step people could
take. And then Gemini again is going to update the canvas on the right with some
elements that are interesting from what I said and that's a really easy way to get
ideas out of you if you're not really good at note takingaking or organizing your
ideas and I think for a lot of you guys before you jump into
(17:10) something that's a really cool way to use the chatbot for that. Now before
we wrap up this video I want to show you one more really cool trick that works
again with Google Workspace. So I'm going to go and use notebook LM this time and
I'm going to be using notebook LM in a way you don't think about it. So I'm going
to create a new notebook and here what I'm going to do is you can bring Google Doc
and remember that Gemini did all these call transcripts for us, right? So I'm going
to go and find a bunch of sales calls transcripts.
(17:40) So I'm going to type Gemini. Hopefully I can find a few. So I think this
one is a sales. Yeah, this is also I think this is also Let's take these four only.
So now I have four sales calls transcript inside my Gemini inside my notebook LM.
And what I can do is I can go in the settings here and I can change how it behaves
and what it tells me basically.
(18:01) So I'm going to click on the custom here and I'm going to paste my prompt
here that says act as an expert sales coach. Analyze this call transcript.
Synthesize the patterns across all calls to find what helps and hurts the close
rate. Give feedback on two list recurring strengths and key improvements. For every
point you make, you must back it up with specific direct quotes from the call.
(18:22) And I'm going to make it longer and such so that it can take its time. And
then I can click save and I can say analyze these calls and tell me how to close
better. And I'm just going to send it. And essentially, it's going to have the
context of all these calls. It's going to act as a sales coach. It should tell me
what I'm doing good and what I'm not doing so well.
(18:44) So you will know if you ever go on a sales call with me what you can expect
and you can see the answer is now coming in recurring strength proactive program
identification demonstrating expertise bl that's real cause as well so it's quite
interesting to read but key improvements key improvements so we can see like
proactive pre-flight research for deeper tailoring so basically research your leads
more okay and I give you some quotes like what I said for example earlier clarity
on pricing frameworks for initial engagement again yeah we're not very clear on
(19:13) stronger differentiation against alternatives. Okay, makes sense. I manage
the scope time of expectation and initial discovery. Okay. And yeah, that's pretty
much it. And what I can do as I get more calls done, I can bring them in and I can
ask what I'm improving or not improving. And I can even generate a podcast about my
sales calls if I want to.
(19:33) But I have a creative way to use Notebook LM to analyze you and to help you
be better. So, I hope you enjoyed this video. This was how I use Google Workspace
to get a bunch of stuff done. There are other tools like on Slides and on Vids as
well, etc., but I don't really use them, so I didn't want to talk about them.
(19:50) But honestly, Google Workspace is a powerhouse now if you actually learn
how to use these tools. And I would argue because it's built into Sheets, into
Docs, into Gmail, etc. It's actually more useful than chat GPT because it prevents
from context switching and having to switch tab and having to switch app which is
really distracting when you're trying to get work done.
(20:10) So, I hope this helped you be more productive. Don't forget to subscribe
and I'll see you in the next