SPEAKERS

Meghan Hodge, Christian Klepp, Aniket Deshpande

Christian Klepp  00:00

Here’s a controversial statement: Your return on ad spend or ROAS (Return on Ad Spend) is lying to you, and nobody in your team can fix it. This statement highlights the reality and challenges many marketers face in an ecosystem that’s generally broken. They must contend with and address the truth gap, the talent gap, and the usability gap. Today, we’re talking to two experts who will tell us how marketers can successfully address this challenge through strategy, technology, and data? Welcome to the Marketing Demo Lab series, and I’m your co-host Christian Klepp. Today, I’m joined by Aniket Deshpande from Azalea and Meghan Hodge from Supermetrics, who will talk to us about why ROAS can be misleading, marketing data, and how to resolve the visibility challenge. Let’s go. Okay, and away we go. Welcome to another episode of Marketing Demo Lab. I’m your co-host Christian Klepp, and I am joined today by Aniket from Azalea and Meghan from Supermetrics. Welcome to you both.

Aniket Deshpande  00:59

Thank you.

Meghan Hodge  01:00

Thank you, Christian.

Christian Klepp  01:01

Fantastic. So, like I said, we’ll just dive straight in. I’m really looking forward to this conversation because, wow, we’re going to start with something slightly controversial, but I’m sure you have an answer. When I first spoke to you both, you mentioned that your return on ad spend or ROAS is lying to you. Please explain what you meant by that, and what exactly did you think is misleading? Is it the metric itself, the data going into it, or how teams interpret it?

Aniket Deshpande  01:31

So the answer is as controversial as the question is. So basically, the data isn’t lying to you. Data never lies to you. It’s the interpretation of the data that is lying to you. That is the crux of where Azalea comes in. The ROAS that you get today is from the platforms that you’re sending your ads to, and obviously they would they buy for the same sale, so the platform ROAS that you get from these platforms is built on pixels. is built built on different strategies from where they do a lot of guesswork in what constitutes a sale for a particular ad. They do do it out of their own self-interest, and you have no way to find out which channel actually led to the sale. What? So the ROAS that you see is is going to be 10x from what you actually see in your bank account. So, have you actually lost money on the on the campaign, or actually have you made money on the campaign? That is the gap that we see as as when we say that your ROAS is lying to you.

Christian Klepp  02:53

Okay. Okay. And fantastic. And Meghan, do you have anything to add to that?

Meghan Hodge  02:57

I would just share that you know different industries, organizations, lines of business will measure it in different ways, and there might be other components that are being need to be considered in terms of like a retail organization might want to you know make sure that they include return percentage or maybe gross margin, so there might be other elements that executives want to bake into a real return on ad spend versus you know just looking at okay here’s my spend and here’s my profit. So there might be other factors that might be considered in measurement as well.

Christian Klepp  03:41

Right, right. You might have brought that up earlier, Aniket, but why do you think platform-reported ROAS doesn’t tell the whole story to marketers, especially when the marketing data-and it sounds like it’s fragmented across different channels?

Aniket Deshpande  03:53

Correct. So you you actually hit the nail right on the head. We are seeing more and more channel touches that a prospective customer is looking at. Your ad will be published on maybe 9, 10, 15 channels, right? How do you actually attribute the ad or the sale to a particular ad, right? And it’s not going to stop here. You’re going to see more and more channels getting fragmented, and what will happen is these channels will start taking their own slice of the of the revenue, but and they will take full credit for the sale. So if two channels or three channels are taking credit for the sale, which channel do you actually spend money on, or how do you decide which channel is working and which channel is not working? So, the honest truth that you get is from revenue numbers, and that is what Azalea is trying to do.

Christian Klepp  05:00

I see. I see. Okay. So, what evidence would you say should a marketer look for before concluding that their ROAS is incomplete or misleading?

Aniket Deshpande  05:09

So, it’s a very simple test, Christian, and it is something that each marketer should do. So, can you map each dollar spend on the marketing side with the sales number that you’re getting from your POS (Point of-sale) or or any of your sales sources, can you match dollar to dollar? If you can’t, then you have a problem. If you can’t, you don’t know which channel is working, what channel led to the sales. You you have no clue about how to how your campaigns are working, and what that when that happens, you might be spending more time and more money on a particular campaign that would have worked anyways. For example, retargeting campaigns we see that all the time, right? Retargeting campaigns sometimes just work. You don’t have to do any campaigns; they just work. So, are you spending the money on the retargeting campaigns because the ROAS from the re-targeting campaigns reported by the platforms is showing you a really gorgeous number, or do you do your own analysis and try to find out what is actually leading to the sale? So that is where I think B2B (Business-to-Business) marketers or even B2C  marketers. I know for B2C marketers, it becomes a very big task to actually match the sales. But for B2B marketers, it is actually a very easy task to map your dollar spent on marketing with your sales report. It

Christian Klepp  06:35

sounds almost like you’re trying to navigate your your way in the darkness without a light, some is what you’re describing, and I know both of you have a solution to this, but let’s keep the audience in suspense a little while longer, right? So the big reveal will come at the end, right? But I’m gonna segue or move us on to the next topic of B2B marketing data and the visibility challenge. So, and to both of you, based on your experience, what besides what you’ve already mentioned, what difficulties do you think marketers face when data is challenging to access or siloed?

Meghan Hodge  07:11

Sure. Yep. So, just in speaking to marketers daily, having siloed data really just really inhibits their their insights that they can gather because they don’t have a complete picture of what’s coming in from individual platforms, or maybe they have to go into each individual platform and download a CSV (Comma-Separated Values) and then copy and paste. Really opening up to a lot of error-prone reporting, not being able to have speed to insights, and not being able to stitch everything together to get a true understanding of what’s happening with their spend and what’s really driving campaign success and results.

Christian Klepp  08:00

Right, right, exactly, and and and it’s it it it also delays the analysis, doesn’t it? Right, like if you have to access data that’s siloed, and it it it just prolongs the period, right, of like you know being able to aggregate the data and so forth.

Aniket Deshpande  08:13

Correct. And to add to that complexity, imagine you have to. This is related to the previous question, right? How do you match dollar to dollar? Now you can’t do that on a daily basis manually. You get the marketing data, you find out what your marketing spend is, marketing campaigns are, and then you get your sales data from somewhere. You try to match it together. It is an impossible task, and that is where Azalea comes in, powered by Supermetrics. If we didn’t have Supermetrics, it would be an impossible task to do what we are trying to do. So, absolutely, superimposing your marketing data with your sales data is where you actually get the power of understanding where your ROAS is.

Christian Klepp  09:01

Exactly. This this might sound like a little bit like a leading question, but I’m going to ask it anyway. Why is it important to have reliable current data for understanding a campaign’s performance?

Aniket Deshpande  09:14

The answer is as simple as the question. I mean, if you’re if you’re working on past data, stale data. You are not going to make the right decisions, and you are going to work on fiction rather than what the reality is.

Christian Klepp  09:30

Exactly. So I I suppose that that’s where we go into the realm of the importance of data hygiene, right?

Aniket Deshpande  09:36

Correct.

Christian Klepp  09:37

To make sure that you have that right that clean data in order to like conduct the proper analysis,

Aniket Deshpande  09:42

yes, and and to add to it, you’re making decisions about budgets based on these metrics, right? Now, if you’re getting wrong data or you’re getting incorrect data, you’re going to set incorrect budgets. Either for the campaigns or for your next quarter, so making sure that your data is right and the data is current is very important.

Meghan Hodge  10:08

And even if you think about like in the retail sector, you know we’ve got peak season coming up with the holidays, and I know many retailers and CPG (Consumer Packaged Goods) brands that set up war rooms around Black Friday and Cyber Monday and getting hourly refreshes, and that makes all the difference in whether or not you you actually hit forecast or go above forecast in being able to measure in almost real time campaign performance.

Christian Klepp  10:41

Absolutely. Would you say that bringing more data together creates a more complicated report without, you know, improving decision making? Like, what do you think makes a difference there?

Aniket Deshpande  10:54

So, more data is always good, right? It’s how you reconcile the data and how you present the data that is what is important. So yes, bringing more data is a complicated task, but in in in today’s world, volume isn’t the point problem or it isn’t the actual point. You you have to be able to derive value from that data, and so bringing more data in is complex, but if you contextualize the data correctly, it becomes really easy to deal with that data.

Meghan Hodge  11:31

Yes, and just having more data points and being able to surface those data points in context with the original baseline data just opens up more opportunity in terms of you know other audiences, other types of ads that are performing better, and why they’re performing better, and then being able to make informed decisions to drive revenue.

Christian Klepp  11:55

Got it. Got it. At the time of this recording, it’s October 2026, and I I’m I’m gonna say sorry in advance. I can’t help myself, but you know we can’t avoid talking about AI (Artificial Intelligence). What role do you think AI plays in what you’re talking about now? About like you know about analyzing data and capturing data, collecting data, perhaps even to a degree cleaning the data. Like what what role does artificial intelligence have in this in this highly complex ecosystem.

Meghan Hodge  12:25

I think that that story is being written daily, and I think it’s a bit of a moving target. But from being on the front lines and seeing how super metrics evolution around being an agentic first product and platform. It’s been a pretty incredible ride to see pointing APIs that have hundreds of metrics and dimensions into an LLM (Large Language Models), and then being able to build out reports instantaneously. So the speed to insights is is incredible, especially with the APIs (application programming interface) that Supermetrics has. The depth of our connectors are much more granular than some out there. So, you know, you can imagine really the sky’s the limit in terms of the number of different reports that you can surface. You know, typically most marketers start with a ROAS or a cost per acquisition report, and it kind of stops there because they just don’t have the resources and the teams that can maintain these reports. And now, you know, really, it’s it’s only going to be really around the the type of what your imagination could be really-that’s going to inform the kinds of reports. So, anything from anomaly reports to you know looking at social engagement and creating baseline reports for that.

Christian Klepp  13:56

No, I think you were onto something there, and I’m not going to ask both of you to prophesy because I know that that’s not within your area of expertise. But where do you think that this is going to go with AI and data? How like like looking looking 10 years into the future?

Aniket Deshpande  14:16

10 years is a big time frame that you’re talking about with AI coming in, I don’t think anybody can can look at more than a year in the future, because what we are seeing right now is that, and specifically on the marketing front, right, you will you will be able to synthesize a lot of data, past data, how your campaigns have performed, and to be able to synthesize that data and get insights from that data. Earlier, maybe a year ago or two years ago, it it would have been a difficult task. And to be frank with you, marketing data is the. Is the single most data that has like millions and millions of rows of data that need to be synthesized. I don’t. I have not seen that kind of volume in any other industry, and the the the columns of data and the rows of data that come in from marketing channels is is humongous, and AI is going to play a very big role in helping synthesize that data.

Meghan Hodge  15:26

I would just add, in terms of like, you know, because if this story is being written daily, yes, I would say, you know, my conversations with clients are going to change, you know, and they already have changed dramatically in the last six months or so, but I certainly see marketing teams, you know, building out autonomous agents to help with their daily functions and to help with their ad spends and continue to, you know, be able to build out models and and you know being able to respond instantaneously to some of, you know, the responses to ads and the performance of ads, etc. But I do think that there also will be the need for human element as well in managing everything because there just always is. And 100% I think that will be that will be something that would be will be valuable in the future as well.

Christian Klepp  16:25

100% We we we can’t let the AI just be left to its own devices. That would put us in a very precarious position. I think. Right. Okay. So we’re moving on to the next topic, and if you can, please cite an example of a client you’ve worked with, where you’ve used the approach you’ve been discussing in this conversation. So I’m going to break it down like this, and I’m happy to repeat. So first of all, what challenge or question did they have that they couldn’t answer? What did they believe about performance, and what changed when they had a fuller view? And if possible, share outcomes.

Aniket Deshpande  17:00

Sure. So I’ll go first. So obviously Azalea is incomplete without Supermetrics. So whenever I talk about case studies or whenever I talk about anything, you have to understand that Supermetrics is the base on which we have built Azalea. They had a question about which are your profitable channels, and not just depending on the ROAS that these channels are reporting, but which channels have the highest return on ad spend. And they couldn’t answer this question because each of their channels were reporting high ROAS, and they were not seeing the actual mapping of marketing dollars with sales on each and every campaign that they were running, so this was something that we couldn’t answer. What they found out, and this was their belief, right? So, for example, one of the campaigns that they were running continuously was the retargeting campaign for one of their SKUs (Stock Keeping Unit). Now, those ROASs were like really awesome, and when we have when they ran these campaigns through Azalea, we have this campaign compare feature where you can do your own A/B testing on your campaigns, right? So they found out that these retargeting campaigns were actually useless because without those campaigns, people were still signing up and buying those SKUs without having to spend money on those those campaigns. So what we actually did was we actually saved around 30% of their budgets for retargeting. And what what then happened was those 30% were diverted. That 30% of the cost was diverted to other campaigns, which were either performing poorly or they were they were not retargeting campaigns, and they were trying to target net new customers. So that is something that the customer was really happy when they when they saw this outcome on our platform.

Christian Klepp  18:58

Fantastic, fantastic, and I was did it did it take them a lot of convincing in in terms of like you know using the approach that that you’re using or or that they feel like perhaps that that the pain point was so amplified that they were willing to go down a different route and excuse me for oversimplifying it but you know what I mean

Aniket Deshpande  19:20

no yeah right Christian and it’s it’s a very good question. Actually, this is a very new concept that we are introducing to the market, right? That you you have to depend on your sales numbers for for finding out which channels work, and this is a completely new topic for even marketers because marketers were only dependent on that one ROAS figure that you get from all these paid channels, and that idea is changing. So it took a little bit of convincing, but when we showed the customer the kind of reports that they are going to get, the kind of insights that they are going to get from each campaign, it was after. It was an easy sell because then it it is like a switch that you that switch on switches on in your brain that it says because marketers have been doing this for multiple years and they can’t see anything beyond that. So when you actually convince them that you know this is the right way to do it, that is when after that it was like smooth sailing for

Christian Klepp  20:20

Yeah, that’s interesting, and and and perhaps that’s a great segue to the next question. Like, what would you say to somebody that’s a potential client that looks at what Azalea and Supermetrics is doing, and and they see that and say, like, you know what, that’s not applicable to us because that’s not the way that’s not the way our systems work. Like, what what would you say to those people? Like the doubters, what would you? How would you put you? How would you put their doubts to rest?

Aniket Deshpande  20:46

To be frank with you, I would say that maybe Azalea does not cater to every marketing team. For example, if you’re if you know your channels are not fragmented, you’re only working on two or three channels, and it is easy for you to map your sales numbers to your marketing numbers. You don’t need azalea. To be very frank with you, it’s when these channels fragment. If you’re if you’re campaigning on multiple channels, if you’re multiple sources of sales, POS systems, your e-commerce websites can be multiple sources, right? Now, to make sense of all that data, you need something like Azalea. So, I would say that you know, in in one sentence, Azalea cannot may not be the right answer for everybody, but it it would be the best answer for people who want to map their marketing spend with their sales,

Meghan Hodge  21:44

and I would add that Supermetrics or a tool like ours is needed for any marketing department that’s spending, you know, over $30,000 a month in ad spend. We are watching about 20% of the global ad spend being piped through Supermetrics, and without the visibility into multiple APIs and having a way to aggregate that data and then report on that data, you’re really looking at individual marketing team members going into each individual API and dealing with the idiosyncrasies of them, and it’s just very labor-intensive. So, on average, we see a three to 15% and sometimes larger increase in ROAS as clients adopt our product, and then we typically see a reduction in the hours spent, anywhere from 10 to 40% depending on the size, obviously of the client.

Christian Klepp  22:44

Both of you brought up a good point, and I want to build on that a little bit, right? And I think this is also important for the benefit of the audience. What type of company, from your experience, would benefit from using this Azalea and Supermetrics platform. Like, let’s talk about size, and I think Megan, you also talked about budget, right?

Meghan Hodge  23:08

Yeah, I would say any client that obviously has high ad spend, they have multiple distribution points. They might have, they might have. They might be a manufacturer that is also a retailer, so more of a complex go-to-market where you know they’re really answering to the business as a whole, and they want to pull in multiple different. They want to report on just something larger than just you know marketing performance, they want to see what products that generate the greatest gross margin. You know are how they’re performing in marketing campaigns, and then pushing out you know additional dollars to those campaigns. You know some more complex use cases that require many more data points that may have more complexity in being able to measure ROAS.

Christian Klepp  24:10

Got it. Got it. All right. And so, just in the interest of time, and I’m going to give both of you 30 seconds to talk about this wonderful collaboration between Azalea and Supermetrics, and what differentiates you from other platforms out there that claim to be doing something similar?

Aniket Deshpande  24:29

So Azalea is incomplete without Supermetrics. To be very frank with you, Supermetrics gets us the data in a very reliable and and rapid method. It’s the and we also integrate the sales channels with with our data warehouse that we create, and that is impossible without Supermetrics. What Azalea does is, on top of that, it creates this layer. Which superimposes your marketing data with sales data, and it is able to show you your real ROAS, right? So there there are pipes which get the data in, and there is a layer on top of it which is the truth layer that we that that we call as the truth layer, and you you can’t separate both of them, anyways.

Christian Klepp  25:22

Fantastic, fantastic, and rumor has it that you’ve got a webinar coming up shortly where you’re going to do a full demo to highlight what Azalea can do. Any chance you can give us a quick like sneak peek?

Aniket Deshpande  25:35

Yes, I I think I can. Let me just share my screen. So imagine you are a marketing agency and you are working with, say, multiple customers. So you have these three customers that you can configure. You can configure more, but for this demo purpose, I’m configuring these three. What you see different from your other marketing platforms is the ROAS that you see is highlighted separately, and it actually is the inter the superimposition that like I talked about with your marketing the sales data. So this ROAS is your real ROAS. So the ad spend is is mapped against the real sales, and you get the ROAS that you are looking for. You can deep dive into each of these each of these agencies that you’re working with or companies that you’re working with, and you can get multiple types of reports. You can understand the the campaigns that are going on, how they are performing, and you can compare your campaigns. So I talked about A/B testing. You can compare these campaigns on various aspects. So as you can see here, this has a problem, right? So the plat platform reported ROAS and the POS reported ROAS is completely different. So you will be able to do those those comparisons. Then there are a few of your comparison screens that that you get, and there are some configuration screens as well. The way we do our configurations for for getting all these metrics in place. So this is like a sneak peek into it. the the The main feature that we are working on is this. So there is an Ask Azalea interface where it is like a chat interface. You can ask any question you want, and it will be contextualized within the data that is present within your data. Basically, your platform reported data, your sales reported data, and you can ask any questions here, and it will give you these answers. So this is just a sneak peek into what Azalea is all about.

Christian Klepp  27:53

Fantastic, fantastic! Thank you for sharing that. And if people are interested in attending this webinar, how can they register?

Aniket Deshpande  27:59

So there is a link that we can share with you,

Christian Klepp  28:02

and we can yeah we can drop in the show notes.

Aniket Deshpande  28:05

Absolutely,

Christian Klepp  28:06

that brings us to the end of this interview. I want to thank you both for your time, and please quickly introduce yourself and let us know how people can get in touch with you.

Aniket Deshpande  28:15

So I’m Aniket. I’m the co-founder of Azalea, and I would want you you can look me up on LinkedIn, and I welcome each and every one of you to attend the webinar, which is happening on the 27th of October, and it’s it has a very fun theme. It has this Halloween theme that I’m sure everybody is going to have a laugh and and have a really good time. So I welcome each and every one of you to that.

Meghan Hodge  28:44

My name is Meghan Hodge, and I’m with Supermetrics. And you can join us at the webinar and bring some candy for a little trunk or treat. And we look forward to meeting with all of you and talking about our collaboration and seeing if we can drive some success for you.

Christian Klepp  29:03

Fantastic, fantastic! And I really hope that the scary aspect of it is limited to the to the background, right? Like the background decoration and what have you. No, and the ROAS. So the ROAS monster. So yes, that’s the ROAS monster. So you guys are going to talk about how to get rid of the raw ass monster. Exactly.

Aniket Deshpande  29:22

Exactly.

Christian Klepp  29:23

Yeah. Make sure the monster doesn’t come back. All right. Well, in that case, Aniket and Meghan, thank you so much for your time. Take care. Stay safe, and talk to you soon.