Pax Gray (00:53)
All right, Gallant, thank you so much for being on the show today. We're excited to have you and diving in. ⁓ Gallant, you have said that most people ⁓ feel like they're doing ⁓ data-driven marketing, but they're not actually doing it, or at least not driving all the way that they could.

Where do you feel like marketers are falling short? Where do you see in your consulting organizations that aren't like, where are they falling short of doing true data-driven marketing?

Gallant Chen (01:23)
I think one of the challenges for marketing as a function is that oftentimes it's under-resourced from a data reporting analytics perspective. I work with a lot of companies where, in some cases, you might have a team of 20, 30, 50 plus marketers. And in some cases, they have one analytics person supporting the entire organization and function.

And I think about how important data and reporting and analytics is for the function of marketing, especially a lot of the digital demand gen performance stuff. And I just think that that is not enough in most cases. And so I think like part of the challenge is for marketing teams to build the case and

get the right resourcing. Sometimes that is because those types of resources are centrally located in the organization. There might be a centralized analytics or data engineering ⁓ function, and then those teams are often not as familiar with marketing data and reporting, ⁓ in part because a lot of those teams are primarily usually built

to support a lot of the finance or product or engineering side of the business. so marketing is often a secondary priority for them. And so for the marketing teams to justify the right resourcing can be quite challenging.

Pax Gray (03:08)
If you were to wave a magic marketing wand over a company to create the ideal setup from a data and marketing ⁓ data, data field marketing perspective, what would that setup look like? What roles and functions do you think organizations should build within or dedicate to a marketing team?

Gallant Chen (03:36)
It's a good question. There's a lot of pieces to it, and I think maybe that's part of the problem, which is ⁓ there is just making sure that you have the data that you need and then collecting that data and then actually being able to put it in the right places where it's actionable and usable. ⁓ In some cases, that means building, reporting, dashboards. ⁓

different tools that can support the different functions of marketing. ⁓ So there's some of that, which is just what is the data that is actually needed? ⁓ And from my perspective, to really understand marketing, you want to understand the end-to-end, right? Which is like, we do a bunch of things from marketing perspective. We invest in a bunch of channels. We run a bunch of events. We have a bunch of campaigns.

and we spend a bunch of money that drives, in some cases, traffic that gets us leads from events. And then what does that lead to downstream for the business in terms of, on the types of business, whether that's like if you're a more traditional B2B business, that's leads and MQLs and SQLs and opportunities and pipeline and ultimately close one. And being able to, first of all, collect all that data and then understand

what are sort of all the touch points and the lifecycle interactions that those customers had before or prospects had before becoming customers. Just gathering all that data, connecting it and being able to report on it is quite a significant undertaking, right? Because first off, you need to like architect the systems. You need to understand what you need to collect and then you actually need to build the systems to collect it. And in order to do that,

Pax Gray (05:29)
Hmm.

Gallant Chen (05:32)
Oftentimes, you're working with existing software tools, right? So if you think about your typical B2B marketing tech stack, you probably have some sort of front-end analytics solution that you're using. Maybe that's Google Analytics, maybe that's an Adobe Analytics. You probably have some sort of like...

JavaScript tracking to understand the sources that people are coming in from in terms of, you know, are they coming in direct via organic search, paid search or other referral channels. And then you're tracking things like form submissions and leads. ⁓

and then trying to understand like, okay, where did those come from? And then following those down the funnel ⁓ in a sort of like a marketing automation platform or in a CRM, like a Salesforce, and understanding which of those convert from leads to actually marketing qualified leads that ultimately lead to opportunities and pipeline. And so there's a lot of data pieces that need to be connected, and then you need to build sort of the reporting layer on top of that.

So it requires having the right systems, gathering the right data. And then in order to do a lot of these things, you often need data engineering resources to be able to connect the different data sources or perhaps pull from the data sources and push it into a data warehouse. So I think one of the things that I think is

Pax Gray (07:01)
Mm-hmm.

Gallant Chen (07:05)
ideal, because you asked sort of like, you know, if I could wave a magic wand and say, these are the things that you want, is I typically recommend that clients have a marketing specific data warehouse. And I would say a lot of companies

typically have a data warehouse of some kind, but it's usually a product data warehouse or a finance data warehouse. And they don't always have a marketing data warehouse. There may be some marketing data that is collected in that data warehouse, but it's not specific for marketing. And so I feel pretty strongly that if possible, marketers should have their own data warehouse that captures all of the marketing data, puts it into ⁓ one place where

the marketing team can report on that. And so in order to do that, you need the data engineering resources to do that. Those resources are not cheap. ⁓ And so for marketing to justify that is quite challenging. And then that's just getting the data into the systems and then being able to have it accessible. On top of that, you need data analytics, data science resources to actually figure out what that data means.

Pax Gray (08:03)
Mm-hmm.

Gallant Chen (08:22)
how to present it in a reportable fashion that supports the needs of the ⁓ functions of marketing, and then in some cases to actually understand and interpret it and make sure that the data is actually actionable ⁓ and that in some cases that it's statistically significant, all these sorts of things that again are like relatively expensive resources, right? Like most marketing teams, depending on the size,

Pax Gray (08:47)
Yeah.

Gallant Chen (08:51)
probably don't have a full-time dedicated data science resource. They might be able to borrow part of a data scientist or data science team's resources, but in most cases, they don't have it. But the reality is, if you think about some of the things that they need to be able to do, it's often quite important as a skill set that is therefore lacking in the teams.

Pax Gray (09:14)
Yeah. So I'd imagine part of the barrier to ⁓ getting this level of data within the marketing function is incentives for individuals. And I think it could potentially maybe even boil down to that. ⁓ The CEO or C-suite has one set of motivators that are perhaps different than motivators at the director level within marketing or even the tactical execution level.

And, ⁓ I wonder if perhaps the incentives don't align or when we try to get budget for this data, we're not accurately addressing those incentives at the higher levels. So I guess if you could lay out, you know, I wouldn't expect this to be a complete list, but what are some of the benefits? Like what's the gold at the end of that rainbow or that struggle, I guess, actually of building out that, that data function, like what will marketers

be able to do that they can't do now? And what will that mean for organizations in a dollars and cents? ⁓

Gallant Chen (10:25)
I think to the extent that you have the right data and reporting and analytics to support marketing, then what that will enable is for marketing to understand, you know, to a fairly, you know, in some cases in a fairly detailed level.

what is essentially working and not working. Now, there's obviously some challenges in terms of understanding. Even if I collect all of this data and I have data science resources to say that these things are correlated, correlation is not always causation, in which case sometimes you have to do things like incrementality tests to differentiate.

But I think if you can have all of the data, then it should allow you to understand, you know.

essentially like where you should focus your marketing resources going forward and better sort of allocate the limited resources that you have from a budgeting perspective to go after the areas that are the highest return in terms of a marketing investment. And what that allows you to do is get more for your dollars going forward, ⁓ right, in terms of like optimizing existing spend, right? So to the extent that you have the ability to understand, you know, that these specific

channels are performing better than these other channels and therefore I should shift my dollars, ⁓ that can help you become more effective as a marketing organization. And so there's the ability to optimize the existing spend. That's sort of like ⁓ one major benefit. And then the second major benefit, which in some cases is a little more challenging, is to help figure out where there are opportunities to increase spend and growth.

Right. I think like ultimately. ⁓

Pax Gray (12:15)
Yeah.

Gallant Chen (12:17)
you know, a lot of the sort of like C level folks and especially certainly the CEO and the folks on the sales side are going to be very focused on how can we grow the business? How can you deliver more pipeline? How can we actually make more money as a business? And, you know, from that perspective, it's always to some degree challenging to understand, okay, which of the areas that I'm investing in already have more headroom to grow?

⁓ And to the extent you already know what's working well and you have the data to understand where there is headroom, then that allows you to have the ability to hopefully justify future investments. And so I'll give you a very specific example. So to the extent that...

Pax Gray (13:03)
minute.

Gallant Chen (13:08)
I'm investing in non-brand paid search as a channel, right? And it performs well from a return on ad spend perspective, right? I'm meeting my, whatever my efficiency goals are in terms of a profitable channel, right? Meaning I'm able to acquire leads at a reasonable rate. It hits my target cost per MQL or my target cost per opportunity, or my pipeline return on ad spend is above one, right? Obviously.

That may or may not be profitable depending on how well you convert your pipeline. But if you can look at it and say non-brand paid search is profitable, this is how profitable it is, and then be able to understand, also have the data because I have all of this in my data warehouse and I've looked at it.

Pax Gray (13:41)
Right.

Gallant Chen (13:58)
that can help me understand I actually have a lot of headroom in paid search. And I can say that because I can look at impression share, I can look at what my competitors are spending, I can look at how much I've spent in the past and understand that if I invest these incremental dollars that I can expect this incremental return, right? Meaning like if I invest another X dollars in paid search, I can deliver this out from an output perspective.

in terms of this many more leads, this many more MQLs, this much more pipeline, this much more closed one. And again, that that is profitable. And then what that allows you to do is for marketing to not only better optimize, which was obviously like the point one, but to actually find the opportunities to grow. So I think those are kind of like the two key areas that really having data.

can help you really truly understand, right? As opposed to, we're doing a bunch of stuff we don't know, you know, which is one of the challenges for marketing, right? Which is in most cases, the marketing teams are doing lots of things. And so it can be challenging to figure out of all the things that we're doing, what is it that's actually driving the business to the extent that marketing is having an impact?

Pax Gray (14:58)
Yeah.

Yeah, I've seen, ⁓ marketing teams get caught up when it comes to data. ⁓ you know, data is inherently. Rear facing it is what has happened. And I think, ⁓ kind of a novice move in the marketing world is to place so much emphasis on what has happened and not enough emphasis on. So what will happen? Right. And so I think.

Gallant Chen (15:31)
Yeah. Yeah.

Pax Gray (15:48)
getting the investment for the data. If, if we're making a stronger case for this is going to inform the future rather than this helps me better understand the past. That's just not a compelling standpoint. And I think sometimes we rely a little bit too much on that. ⁓ and particularly around AI, think, ⁓ when it comes to AI, I'm convinced that the, those with more and better data are going to be the ones who.

will catapult forward. And I'm curious if you have any stories or use cases around companies or brands that have taken some really cool data and used it in AI or have you seen anything like that yet to further add another like button or badge to the case of like, you got to invest in gathering more data.

Gallant Chen (16:41)
Good question. don't know if, let me think about that.

I think one area that, and arguably this is an area that marketers, a lot of marketers have used AI ⁓ for many years now, which is ⁓ effectively smart bidding. ⁓

from all the different ad platforms. So if you think about it today, even like your Facebook and meta ads, as well as Google ads, a lot of the way that they are pushing advertisers is to give them the data to optimize and to drive results for you. And as part of that, they are asking you to give up control, which I think for some marketers can be quite...

disconcerting, right? Which is like, hey, it doesn't matter what keywords you bid on. Don't worry about the match types. Don't worry about what the ad copy is. We'll do it for you. ⁓ And I think as marketers who've had those controls and feel like we know what we're doing in terms of choosing the right keywords that make sense for our business or what match types make the most sense or what ad creative is going to resonate most with our customers.

Pax Gray (17:46)
Thanks

Gallant Chen (18:04)
that that's kind of, it feels strange, right? ⁓ Because essentially we are giving up control. ⁓ But what I will say is that I have found that not always, but in many cases, when you give the platforms the right data set to optimize on, they will drive good performance for you, right? And so...

Pax Gray (18:09)
Hit it.

Gallant Chen (18:30)
One of the challenges is making sure that you give them the right signal so that they will actually optimize and get you the best results. And what I have found is that when you do that, they do quickly figure out what works and what doesn't work.

As an example, one of my clients was very focused from a smart bidding perspective previously on essentially just leads. And so they were driving leads. And so the goal in bidding was optimized to drive the most leads for ⁓ their business. And what they found is that ⁓ Google could drive leads at a low cost per lead, but that the lead quality was quite poor.

Pax Gray (18:58)
Thank

Gallant Chen (19:16)
And that's because ultimately, Google is trying to drive as many leads as possible at the lowest cost for you. ⁓ And because of that, they're going to be the lowest leads that are going to exist. Whereas if you can optimize to a more downstream metric, for example, opportunities or even SQLs, then you're giving Google much better signal. And so they can figure out pretty quickly, these leads.

Pax Gray (19:31)
Mm-hmm.

Gallant Chen (19:44)
don't convert to SQL, and those were driven by these keywords and these match types, and I'm gonna turn that off, right? And I'm gonna focus on the ones that actually are actually driving your SQLs or your opportunities or whatever it is that you are now optimizing on. If you give them good signal, then they actually will find you the best stuff. ⁓ And I honestly do believe that in most of those cases, having that type of data.

is actually driving better results for marketers.

Pax Gray (20:16)
Great use

case. That's a perfect example. ⁓ and again, it merits investment then where you can connect the dots between, listen, if I can get better data and cleaner data, then I can feed it into this. I'm to get better output. ⁓ and that's a language that C-suite I think understands.

Gallant Chen (20:30)
Yeah. Yeah.

And I think it's one of these challenges, is sometimes it is difficult for marketers to justify and build the case that, I need to actually push this downstream conversion data from my platforms, right? From Salesforce or from my data warehouse. And I need to push it back into the ad platforms.

Because that requires essentially like an engineering project, right? To connect to the API, figure out what is the data source, how am going to push that data on a regular basis to make sure that Google or Meta has the right data. And in a lot of cases, the marketing teams don't have the right resources to actually do this, right? Because it does require some dev development skills and some data engineering work.

And if they don't have those skill sets, then they need to go figure out what skill sets in the company do have those resources that then can support these efforts. And then part of the challenge is while those resources may have the technical understanding to figure out how to connect to an API, they're not familiar with a Google Ads API platform or a Meta Ads API platform. So they have to learn it too, right? And so...

It's actually a pretty significant lift to actually put in place the infrastructure to do offline conversions or CAPI. And therefore, oftentimes doesn't happen. So I've worked with companies where I'm like, wow, you're spending millions per month, ⁓ and yet you don't have good data that's being pushed back into the ad platforms. You're just relying on the downstream.

Pax Gray (22:09)
Yeah.

Yeah.

so these, ⁓ you know, lot of these organizations, despite this are still going to have a tough time getting the investment they need for a complete build out the way it should be done. So oftentimes as marketers, we have to deal with, okay, so what do do now? If I can't get that investment, what's the next best thing? So what would you say to an organization, you know, mid level, we've got a team of five, seven people doesn't size doesn't really matter, but if I am struggling to get that investment.

but I still feel like I need to get this data. I need to use this correctly. What would be maybe like a good, better, best versions of, I guess not best, but short of best, what would be some things that I could do with my org that I don't necessarily need as much investment for?

Gallant Chen (24:07)
Sure, ⁓ this is pretty difficult, but remember when we first talked earlier in the conversation, we talked about being able to collect all of the data, ideally being able to put it into a data warehouse and then being able to report on that. That's sort of like the ideal state, but that's sort of, in a lot of cases, very challenging to justify and...

quite difficult from a resourcing perspective. So I am certainly very understanding that not all companies ⁓ have the ability to actually make those types of investments. And a lot of my work is primarily with B2B software companies. So this may be more specific to those kinds of companies and sort of their tech stacks and kind of like what they have in place. But I kind of think about...

three sort of levels of what you can do, ⁓ right? And there's different options for how can you kind of get, maybe not a complete picture that you would get with everything in the data warehouse, but ⁓ a somewhat complete picture, right? And so, one option is to try to look at your existing tech stack and figure out, is there one system that I can use to be sort of my system of record or source of truth?

and put most of the data into that. And so if you think about a B2B type environment, you often have a marketing automation platform and a CRM. In a lot of cases, that CRM is Salesforce. And I have a lot of my clients where what they will do is they will put all of the attribution data, marketing data, as much as possible into Salesforce. They'll append that information at the lead level, make sure it gets past the opportunities. ⁓

and ultimately to closed one. And so they're able to use Salesforce as sort of a system of record and then be able to look at marketing performance within the CRM. And that's not perfect, but it can do a pretty good job of capturing most of the stuff that you want, right? The other sort of option that I've seen is for people to essentially push the data to, you know,

spreadsheets or Google Sheets or some sort of intermediate platform. And so an example of this would be taking some of your ad platform data from Google Ads and pushing that to a Google Sheet and then pulling data from maybe it's HubSpot or Salesforce and pushing that into a Google Sheet or an Excel sheet and then pivoting them together so that you can understand and connect the two data sources together. And so...

Sometimes you can build automations to make this a little bit easier. You can use different connector tools like a funnel or supermetrics to make this a little bit easier as well. But then in those scenarios, you're able to at least get the reporting that you need into one place, which is a spreadsheet, ⁓ which still has limitations, but it can be good enough for a lot of different businesses. And then obviously,

The other way that you can do it, which is not sort of the full data warehouse, is you can push the data, which we just talked about, into the ad platforms themselves. So assuming that most of your marketing investments that you're focused on are within paid media platforms, right? To the extent like, you know, and this is true for a lot of businesses where all your money is being spent on Google or all your money is being spent on Meta, then what you can do is essentially push your downstream data.

into those platforms and then you can use those platforms to measure success because you'll have all of their, you know, spend data, all of the impressions, the clicks, and then you'll be able to see your downstream conversions to leads and opportunities and to pipeline and all those kind of things. So then you can have that as a system of record that helps you understand performance. So those are probably like the three options that I think are easier to accomplish than doing a full

⁓ data warehouse of all your marketing data.

Pax Gray (28:21)
Yeah, I love that. And I think that's super actionable. ⁓ and I doable, especially if you have got like a more, more simple, marketing plan and tactics. ⁓

Gallant Chen (28:32)
Yeah, I'm very

cognizant that, you know, it's sort of like what is practical. What is practical given the environment and the resourcing and the teams and the different resources. so, you know, while I am a really a big, am always in favor of a ⁓ dedicated marketing data warehouse. I'm very cognizant that that is not always possible. And so that doesn't mean that just because you can't do that, that you can't have.

Pax Gray (28:37)
Yeah.

Yeah.

Gallant Chen (29:01)
a better sense for what your actual data and performance is.

Pax Gray (29:06)
Yeah. The, in the, especially in the B2B SaaS world, in my experience, the, the, the first wrench that I see that gets thrown into a lot of those more ideal data plans is trade shows. The trade shows are always like a big wrench because it's like, Hey, listen, I, I can just go in and talk to customers and they're, I'm going to get this many demos or this many leads. And, but then, you know, then the battle's like, well,

They only came up to the booth because they saw the ads that we were hitting them with beforehand or the emails or yeah, but we're retargeting them and they actually converted because of that. You know, it's just like, gets infinitely complex by adding this one variable of in-person communication. And that, brings me to my next question, is I years ago, we used to have, uh, um, structured debates within 97 floor. We would just have two sides and debate some topic and

One topic I'll never forget. We debated was, should you do, should marketing be data focused, data-based or should it be gut based? You know, should you just go off my intuition? I just, you know, I don't have data back to set, but I just know they're going to love this. I know this is going to work and there's, you know, there's arguably some merit to both sides and some mix truly. ⁓ but I have seen organizations tie themselves in knots, especially when it comes to attribution models.

trying to get the perfect attribution model and it ends up being so expensive, both from a resource and a time perspective, when they would have been much better off with just saying like, I'm going to accept some degree of blindness to what's happening and I'm in favor of action. ⁓ So I guess what would you say to that? What's your perspective on that? You are so in the data with all these different organizations. I guess what would you say to somebody who says, listen, just don't worry about it, just get a

Get, 15 % visibility. That's good enough. We're just going to charge forward. We'll make mistakes, but it's through the charging forward that we'll actually learn what works and what doesn't.

Gallant Chen (31:11)
⁓ that's a good question. think...

Pax Gray (31:15)
You know it's good when he sighs.

Gallant Chen (31:21)
I will say like it is what you described I see fairly often, which is, you know.

marketing, maybe the broader company even decides that they need a new attribution system to properly understand, evaluate, measure performance so they can do the things that we talked about earlier, which is like optimize existing spend and figure out where to invest going forward to grow. And the problem is that those types of projects tend to be, you know,

across the entire marketing org, often even outside of the marketing org, you have lots of cooks. Everybody has a different perspective. ⁓ Everybody wants a certain way approach. ⁓ And so you have this sort of like, you know...

by committee decision-making process that essentially creates, ⁓ in a lot of cases, an attribution model that is essentially good for nobody, right? Because it's trying to accommodate too many different stakeholders. ⁓ And then it just becomes very complex. then everyone's a little unhappy. Some people are very unhappy. Some people are a little unhappy. Everybody's unhappy. Nobody's happy, right?

because they think, well, this doesn't properly give credit to my channel that I'm investing in and here's why, right? And really this other model would be better. And so that's quite challenging, especially when you have a process where it's done by committee. ⁓ I think if you are going to do that process, you really need a strong sort of like...

leader or group within the company that's going to take a look and understand all the different stakeholders and ideally decide what is best for the broadest broader company based on where the business is, right? ⁓ Because, you know, depending on your business, the type of model and how sophisticated it needs to be can be pretty different, right? So like, if I'm a business that is all digital, I don't do any of the stuff that you just talked about. I'm not doing any offline stuff. I'm not doing any events. ⁓

then I could probably get away with a fairly simplistic model, right? ⁓ Especially if there's not multiple touch points, right? Where it's like someone's, know, seeing many ads or coming to the website many times before converting to a lead or to an opportunity. Whereas if I have a business that I have a multi-channel, I have multi-channel efforts where I've got a large event or trade show effort, I'm doing a bunch of, you know, ⁓

content efforts, have other things that I'm doing from a customer evangelism perspective, maybe I have partners involved as well, maybe I have a business where it's a very long sales cycle, right? Then the type of attribution model that you're going to need is much more complex than for other businesses. ⁓ And so I think the challenge is figuring out for your business,

What is the type of model that is necessary to properly support what you need? And I do think, you know, I'm sympathetic to the fact that most companies ideally like want like one model to rule them all, right? Where it's like people can say, hey, this is the shared model that most everybody uses. But I also think that for a lot of teams, there should be the ability to see different models

in a different way for a specific purpose, right? So ⁓ what's an example, right? If I am running, let's say you create a model that is very first touch introducer focused because you're very focused on top of funnel driving new customers. But if I'm a, know, email lifecycle marketer or I'm a LinkedIn ⁓ marketer who is running campaigns on, you know,

existing prospects that we're trying to close, then that model can't measure what I'm doing, right? Because it's mostly focused on top of funnel, introducers before lead create. When these people are already leads, they're already in my system. I am essentially, you know, doing retargeting of those existing users. And so that needs to be measured in a different way. So even if you build that one model that is tilted towards a specific goal,

Pax Gray (35:39)
Right.

Gallant Chen (36:02)
you should have the ability for different versions of the attribution to be able to be applied for different purposes within the business. Assuming you have that kind of level of sophistication where you have some marketers who are focused on top of funnel versus some marketers that are focused on maybe pipeline acceleration, right? Or in some cases, expansion business, As opposed to new business. And as you get...

to be a larger organization, a lot of those things are true, which is why I think the sort of like build a single model for the entire company is just like, it's too difficult and therefore you end up in a world where nobody's happy with what they have.

Pax Gray (36:43)
Yeah.

Yeah. I think that's interesting. ⁓ and a really great way for these more complex scores to tackle it is like build models to help people do their jobs better. And I think, ⁓ where it can tend to get, go off the rails is when the models are tied to compensation, cause investment inherently then means compensation. And so

Gallant Chen (36:46)
All right.

Pax Gray (37:12)
It's one thing to say the, you know, the search channel is getting credit for this conversion. And it's another thing to say the paid media specialist is getting credit for that conversion, you know? And I think, that idea of like using multiple models, none of them is kind of the BLNL, but they all help everyone to do their jobs more intelligently.

Seems like that that range true is like a really great goal for these organizations, ⁓ as they're building out their, their data. so I'd love to, to, and let's say, if I'm leading a team and we're in what I would call like a nation nascent stage of being like having this, like, I'm not even close to the data warehouse. I'm, trying to get there. And, ⁓ let's say maybe I have a team that doesn't,

quite yet have the skills needed to take us there. Barring the investment, what would you say to a marketing leader, have your team start learning X or have them start diving into this in order to get your team moving in the right direction? What would be that next step for them?

Gallant Chen (38:29)
I'm gonna spin it a little bit differently, right? Which is one of the things that can happen within an organization is like, in a lot of cases, let's take like your typical paid media campaign managers, like maybe someone who's running, know, paid social ads, someone who's running programmatic, someone who's running paid search, right?

Pax Gray (38:31)
Hey, yeah, that's great.

Gallant Chen (38:52)
they're responsible primarily for managing those channels, right? Managing the budget, working with those specific platforms or those specific vendors and driving the business. In a lot of cases, they are not able to effectively do their jobs because they don't have the right data, right? Or they don't have the right resourcing, right? And what

I find is the challenge is like they're typically working in that realm. And that's obviously where they have all their direct control, ⁓ where they don't have the control as much as what reporting I have, right?

whether or not I can set up like, you know, going back to our example before, which is like, if I'm in, you know, doing meta ads and I want to set up, you know, conversions API, I don't know how to set up conversions API, right? But if I look at it and I go, actually, if I can't set up a conversion API, I won't be able to be successful as a marketer in meta.

Right. And I, you know, that's not always true, but in some cases it is. And when those cases happen, you know, I do find that a lot of times those marketers, they're well intentioned, right? They will go and try to figure out how to get the resources to do it. But in a lot of cases, they run into a roadblock. ⁓ this person doesn't want to do it. They are not resourced. They're waiting to hire someone. So maybe they'll be able to do it in six months or another quarter, or we'll put you on the list, but you know,

Pax Gray (39:57)
the

Yeah.

Gallant Chen (40:27)
It doesn't ever happen, and they check another quarter. And eventually, a lot of them give up. ⁓ And I kind of look at it, and I go, gosh, you can't give up. And what they do is they just default back to, well, I'll just optimize with what I have. I'll just use whatever reporting I have. It's good enough, but it's not perfect. ⁓

Pax Gray (40:32)
Hmm.

Gallant Chen (40:52)
And I just look at that and I say, no, like your success is tied to this. That's way more important, even if it's outside of your wheelhouse and not something that is technically your area of responsibility. It kind of is because you're responsible for that channel. And if that channel cannot be successful because you don't have the right reporting, then you need to be spending the time to figure out how to solve for that, as opposed to like figuring out, ⁓ maybe I'm going to like...

change my bids, refresh this creative, do all the things that you're normally going to do because those things are actually more important. And so to your point, if that means like, I need to go learn SQL ⁓ because I need to go figure out how to like pull the data myself out of the data warehouse. Or if that means like, I need to go research tools that will allow me to connect the data, right? So like going back to that conversion API example, like instead of having, you know, internal dev team,

Pax Gray (41:21)
Yeah.

Gallant Chen (41:49)
you know, write to the meta API, I can go work with a platform and go use a reverse ETL platform. So I can go research what reverse ETL platforms can actually support conversion API via Facebook, figure out what tool could actually do what I need to do, and then go try to build the case ⁓ that that tool should be used in this scenario. Right. And I recognize like, again, like that's not easy. You're probably going to run into some roadblocks. People are probably going to say no.

But if a success for you is tied to some of these analytics or data or reporting issues, to me, those things are far more important. To some degree, and I know this is impractical, you should basically say, I don't think as a company it continues to make sense for me to spend on meta ads until I get this done. ⁓ And I think that's...

Pax Gray (42:40)
Yeah.

Gallant Chen (42:45)
that approach is what's ultimately necessary, but oftentimes it's just challenging within the context of these organizations. Because in most cases they tried, right? ⁓ And it's just they met resistance and then they gave up and then they don't try again, ⁓ which I think is unfortunate because ultimately it means that they are essentially hampered ⁓ and handicapped in terms of what they're able to do.

Pax Gray (43:13)
I love that. That is such a great takeaway and, ⁓ building that resilience to continue to push either figuring out how to build it yourself or continuing to like, to push, to get investment, even if you've been told, no. ⁓ I love that. I mean, that's a great skill set.

Gallant Chen (43:27)
Yeah. Yeah. Like the ideal would be that

they have a data analytics partner or maybe a marketing technology partner who would actually go and drive that for them. Right? That's the ideal and really what should be the case. But in so many of these companies, that just is not true. And so, you know, I feel like as marketers, they should try to own that more. Right? You know, it's obviously not always fair for me as an outsider to say that, but...

I do think that if I look at it, I'm just so data focused, which is like, if you don't have the right data and reporting to be effective in your job, you got to fix that. You have to figure out a way. And maybe it's not the perfect solution, like some of the data warehouse solutions that we talked about. Maybe it's some of the put it all in Salesforce or put it into a Google Sheet, ⁓ but figure out something that gets you to actually be able to understand performance properly.

Pax Gray (44:27)
man, great, great takeaway to end on. I love that. This has been a really great discussion. think it's, if you're listening, like think about where your org is at from a data perspective, ⁓ and then work toward the ultimate solution of like, let's get it, get it into a data warehouse, barring that, get it into, ⁓ you know, your central repository, like Salesforce or spreadsheet. And then I think even looking at the platforms as

kind of a pseudo data warehouse. If, if your marketing plan is more simple, think is a really great, second tier solution, barring, you know, barring something more complex. ⁓ and then yeah, empowering your teams.

Gallant Chen (45:06)
Yeah, if you listen to what

the platforms want to do, right? Like if you heard some of the quotes that Mark Zuckerberg has said, it's like his world is his perspective is like, hey, you just tell us what your goals are ⁓ and give us some money and we'll figure it all out for you. And the reality is they're investing a ton in AI to make sure that they can do that effectively for you as a business better than you can. Right? I know that that is a little bit scary for marketers to say like,

Pax Gray (45:10)
Yeah.

Gallant Chen (45:35)
I'm not gonna choose my creative, I'm not gonna choose the landing pages, I'm not gonna choose what kind of campaign type. But in a lot of cases, if you give them the right data, they actually will be better, right? Because they're gonna use machine learning, they're gonna use AI, they're gonna test, they're gonna iterate, they're gonna be able to do the things that we as humans, unfortunately, ⁓ cannot do as effectively.

Pax Gray (45:46)
Yeah.

Yeah, I mean, they've made some significant investments. And if you're giving them the correct data, you get to benefit from those investments without making any of them yourself in the advancement of AI. And so why not take advantage of that when it's there? I love that. ⁓ Gallant, thank you so much for joining us today. What would be the best? You do tons of consulting a lot in the ⁓ B2B tech world. ⁓ What's the best way for people to connect with you?

Gallant Chen (46:12)
Yeah, definitely true.

you are. ⁓

I'm on LinkedIn, so if folks are interested, I work with a lot of B2B SaaS companies. I've worked with folks like DocuSign, Shopify, Intercom. And if you are a B2B SaaS company that is looking to improve your marketing efforts from a data-driven perspective and use those insights to help optimize and drive growth for your business, please do reach out on LinkedIn.

Pax Gray (46:53)
Okay, great. Thank you so much for being on the show today, Gallant. It's been a pleasure.

Gallant Chen (46:56)
Great.

Thanks Paxton. Have a good one.

Pax Gray (46:59)
You too.