Jul 3, 2026 · 49 min · 10 segments
This discussion includes, How to differentiate productive vs. unproductive spend and track real-time savings when cloud provider billing data inherently lags by 24–48 hours.Centralized vs…
yeah as they said right this is your particularly about how to track and review uh recommendations and realize that as a savings right so this is how uh our i will uh start with my pain points so we have almost like uh when we take multi-cloud azure and aws we have 20 types almost like 20 types of recommendations okay okay so assume that you have a 16 business functions for one type of recommendations i should be sharing 16 excel sheets to send 16 actual seats like you know this is a regular scenario before we had the tool and company recommendations you know like per quarter like to you know like we share the recommendations per
so you say like 10 recommendations you're selling and 10 to 13 okay like 10 recommendations and 10 different sheets into that in business function, it will become 100 plus Excel
You open every day to see like whether it's approved or not and whether can it take to implement it or not.
Okay, like that's a huge task that cannot go in every quarter, like every quarter like 100 sheets to open and track the approvals.
So when I say the recommendations is in the tool, like maybe from the tool or it may be from the
We collect everything and sharing and tracking become, you know, like impossible.
so this is the kind of a scenario we were not efficient and also not humanly possible to do that okay so that was the scenario but yes i uh now uh we when we look at the uh tool in the market okay some rooms are coming up with okay so uh filter that by the uh you know like uh by the recommendation type and also filter by the different view for the different application teams by tag so this is all coming up
so this is all coming up but when we get that we miss some other the prime module the chargeback forecasting that modules are missing so okay and when we want to also like the practically difficulties we cannot have the multiple tools also together so so uh we have to find out with some intermediate approach that you know the cost module okay like the cost module is fine the cost transparency module is fine but recommendations and implementation tracking yes definitely it provides a good recommendations by the tool but how do we present it like you know the tool is missing how do we present and to the application team in a crisp manner and how do we track it that was missing in the tool so that is what we targeted and we created our internal tool so that that tool like as i said before it was like uh but 10 recommendation is 130 sheets so now it's a one sheet we created one sheet that we created all the uh 13 recommendations uh will be in the just a small a simple excel sheet but it's a very effective and it's all run by the you know like the visual basic code and the macro and uh we have uh the technical architecture is it's an actual tool the reason for keeping the excel sheet is it's it's no no need to give an introduction to the excel sheet to all the business features so we want to maintain that that's the reason the second reason is we don't want to introduce another uh like you know another server or url already have a tool okay we don't want to introduce another you are creating confusing this one so we want to keep a familiar ui but all the powerful thing we connected the rds like in the backend and rds in the backend and also we accessed that rds through the api in the excel sheet through the stood over like useful basic and macro.
So the step one there would be, we will be collecting all the recommendations, okay, from different sources.
example, like we can database, we can shut down RDS, we can shut down like EBS snapshots cleanup, okay, and volume log level optimization, volume upgrade, okay and uh database right sizing and everything together in a summary okay we load it into the uh rds and when the when the user that's a backend we clean up and give it crisp data in the backend okay so crisp data so the data model that we created is will will be accepted uh can be used for gcp as well as aws and uh also for the azure so we know that for example it's a volume cleanup so all three clouds volume cleanup will be in a single place so from the user perspective so now it's a technically we have loaded all the data in a crisp manner and all the recommendations when the in the from the user perspective that means the application team who wants to optimize this opens that sheet
yeah as they said right this is your particularly about how to track and review uh recommendations and realize that as a savings right so this is how uh our i will uh start with my pain points so we have almost like uh when we take multi-cloud azure and aws we have 20 types almost like 20 types of recommendations okay okay so assume that you have a 16 business functions for one type of recommendations i should be sharing 16 excel sheets to send 16 actual seats like you know this is a regular scenario before we had the tool and company recommendations you know like per quarter like to you know like we share the recommendations per
so you say like 10 recommendations you're selling and 10 to 13 okay like 10 recommendations and 10 different sheets into that in business function, it will become 100 plus Excel
You open every day to see like whether it's approved or not and whether can it take to implement it or not.
Okay, like that's a huge task that cannot go in every quarter, like every quarter like 100 sheets to open and track the approvals.
So when I say the recommendations is in the tool, like maybe from the tool or it may be from the
We collect everything and sharing and tracking become, you know, like impossible.
so this is the kind of a scenario we were not efficient and also not humanly possible to do that okay so that was the scenario but yes i uh now uh we when we look at the uh tool in the market okay some rooms are coming up with okay so uh filter that by the uh you know like uh by the recommendation type and also filter by the different view for the different application teams by tag so this is all coming up
so this is all coming up but when we get that we miss some other the prime module the chargeback forecasting that modules are missing so okay and when we want to also like the practically difficulties we cannot have the multiple tools also together so so uh we have to find out with some intermediate approach that you know the cost module okay like the cost module is fine the cost transparency module is fine but recommendations and implementation tracking yes definitely it provides a good recommendations by the tool but how do we present it like you know the tool is missing how do we present and to the application team in a crisp manner and how do we track it that was missing in the tool so that is what we targeted and we created our internal tool so that that tool like as i said before it was like uh but 10 recommendation is 130 sheets so now it's a one sheet we created one sheet that we created all the uh 13 recommendations uh will be in the just a small a simple excel sheet but it's a very effective and it's all run by the you know like the visual basic code and the macro and uh we have uh the technical architecture is it's an actual tool the reason for keeping the excel sheet is it's it's no no need to give an introduction to the excel sheet to all the business features so we want to maintain that that's the reason the second reason is we don't want to introduce another uh like you know another server or url already have a tool okay we don't want to introduce another you are creating confusing this one so we want to keep a familiar ui but all the powerful thing we connected the rds like in the backend and rds in the backend and also we accessed that rds through the api in the excel sheet through the stood over like useful basic and macro.
So the step one there would be, we will be collecting all the recommendations, okay, from different sources.
example, like we can database, we can shut down RDS, we can shut down like EBS snapshots cleanup, okay, and volume log level optimization, volume upgrade, okay and uh database right sizing and everything together in a summary okay we load it into the uh rds and when the when the user that's a backend we clean up and give it crisp data in the backend okay so crisp data so the data model that we created is will will be accepted uh can be used for gcp as well as aws and uh also for the azure so we know that for example it's a volume cleanup so all three clouds volume cleanup will be in a single place so from the user perspective so now it's a technically we have loaded all the data in a crisp manner and all the recommendations when the in the from the user perspective that means the application team who wants to optimize this opens that sheet
The rest of this transcript — segmented and speaker-labeled, so you land on the exact moment something was said
Search every transcript — by keyword, by phrase, or by meaning, across every show Radar indexes
Trends — what is surging across podcasts, measured against its own baseline
Alerts — when a name you follow appears in a newly indexed episode
No account is needed to search Radar.
