Source-backed interpretation.
Editorial interpretation is derived from the original DMAasia/WION video description and transcript. Statements by speakers remain attributed to the source; no unverified current-role or performance claim has been added.
The durable question is how AI and automation should improve relevance without losing sight of data and measurement.
Source signal, edited only for readability.
Three takeaways worth carrying forward.
- 01
Read AI and automation as the primary operating question in this record.
- 02
Test its connection to data and measurement and AI against the speaker’s reasoning in the transcript.
- 03
Keep the 2023 source context visible before carrying the lesson into a current decision.
What the transcript puts on the table.
How do you retain the customers you already know is the number one task that anybody in any business has to answer right now.
The problem with business is that he is getting customer attention.
The moment I say Ashurva data available at 10% discount, 43% open rate.
Original context. Current reading.
Recorded in 2023, this captures the language, platforms and constraints of its time. Read it now by separating those period-specific details from the enduring operating question: how should relevance and AI improve the decision?
Durable tests for the work.
- Start from the customer’s situation, not the channel.
- Use AI to improve a decision or outcome—not to decorate the plan.
What would you keep, change or measure differently if this event intelligence were put to work now?
IP09 · READABLE TRANSCRIPT1,222 words
Do you have a suggestion or a recommendation what should CMOs or agency edge do in the next 30 days to get their act around AI right? The problem with business is that he is getting customer attention. When we talk in the context of AI, we are assuming that all the things that could have been done is already done. Open rate is something like 11%, which is great. But the moment I say Ashurva data available at 10% discount, 43% open rate. I can probably help in doing heavy load work. Part of the messaging is the context and the content that's going out, what's adding value to somebody's life.
So with all this that's happening in the background, do you have a suggestion or a recommendation what should CMOs or agency edge do in the next 30 days to get their act around AI right? About two years ago, someone told me that the next thing after water is voice generated ordering for grocery. So I put all my resources, I cut everything else, put everybody behind it. We put a mic next to our search bar. In the last three years, I've got 10 orders. The issue is that what it is used for, right? You got to see how that works out. To answer your question, what you do for the next 30 days, I think 30 hours is a better thing. The problem with business is that the kind of business we run, which is really grocery online, is getting customer retention.
It's extremely expensive getting a new customer, we all know that. If your retention of the customer month one, we have a retention of something like close to 46%, which may be very high compared to other businesses. Being grocery, you can get that. But if it's furniture, there's no way you can be at 46%. How do you retain the customers you already know is the number one task that anybody in any business has to answer right now. So I think when we talk in the context of AI, we are assuming that all the things that could have been done is already done. Now I have to go to AI because there is nothing left. As a company trying to sell something, you need to arrive at the communication channel that's good for you.
So if you do SMSs, your open rates are some 0.4% single digit. And everybody says SMS is very cheap, it's 10 paisa, but divide by 0.4, it's a lot of money. For everybody who opens, forget about buy. So I think it's very expensive, 15 paisa is too expensive when your open rate is 0.4%. But when you did mail, when we send a mail out which is talking about what we do, which is 99.9% fill rate, great quality fruit and vegetable, the best meat in town, the lowest prices in town. No one wants to hear all that stuff, right? Open rate is something like on mail. But the moment I say Ashrava data available at 10% discount, 43% open rate. So you don't have to go as far as AI. You've got to understand what the consumer wants to see.
We now work with a company which now allows you to put a catalog on the mail and you can actually browse the catalog and put things into your cart without leaving the mailbox. Open rate is 37%. Imagine, the first thing you want to do is in this whole myriad of so much of information, you want somebody to open the damn message, right? Now that means that you've got to apply intelligence about nothing to do with AI. It's only to do with what you understand in your business and what the consumer is buying. To look at those trends of what the consumer wants to see, you make them an offer, they are suckers for it.
Yeah, I can probably help in doing the heavy load work in terms of segmenting the data, what email, what time of the day and studying all those additional things that are there. But at the heart of the messaging is the context and the content that's going out, what's adding value to somebody's life. Yeah, but even before you go to AI, for the last several 10 years, we know we've been tracking the customer's journey about when he's opening, what he's opening, does he open 8 o'clock to 10 o'clock. This is peering into somebody's toilet. But the point is that's what we are. Everyone's looking at a mailbox. And some of these mail companies are not allowing your mail to get into a spam. You put in the spam, it'll bounce back in the inbox.
That kind of stuff is cheating. But that's what we're currently living with and we don't know. I want to leave you with two thoughts and from my own personal experience. I was born in Delhi, though I was from Kerala. Somewhere down when I was a teenager, I was airdropped into Trivandrum. And in Trivandrum, TV had not arrived then. So during that time, my friend and I, we went and saw a movie. And that movie, if I tell you the year, you'll guess my age. But at that risk, I'll tell you. This is way back in 1974. We went and saw a movie. And this movie was about the Cold War and how there was only talk of supercomputer. But it was not called supercomputer. Only Russia and America had these computers. And the idea of a satellite was there.
And the story was about how the supercomputer of Russia and the supercomputer of America got together to screw humanity. Now, I'm not joking. I'm not kidding. If the world is connected far more than it is in 1974 today, there is nothing stopping the intelligence of computers. If it is trained to see that you must negate every politician who will act against the interest of the human being, in whichever way it is designed, there is nothing. All the nukes are controlled by the network. The problem is much more if it were to get out of hand. I'm talking about a movie made in 1974. The other thing I tried, this is my personal experience, is Netflix. Netflix. In Netflix, you can make your profile, right?
You, your father, mother, uncle, aunt, whoever it is. In my home, we don't have any profile. My wife, my two sons, one guy is in Philadelphia, one guy is in Bangalore and me. We use the same profile. And you should see the crappy recommendations that Netflix gives me. Absolutely crappy. Not one. I watch Mexican crime and Delhi crime. I'm watching crime all the time because that's where you learn the most. You're seeing that all the time. My wife is watching they and us and stuff like that. Shit, I can't even spend five minutes on that. My sons are watching F1 and they're watching stuff like that. Somebody is watching baseball. My son in Pittsburgh, so he thinks now he'll become American, so he's watching baseball.
So now, the thing is, Netflix is so confused. They have no clue. Not a single recommendation is what I want to watch.
