Exclusive Webinar

Decoding Marketing Impact: Navigating Data Challenges in 360-Degree Analytics

Are you ready to unlock the full potential of your marketing efforts? Join us for a transformative webinar where we'll explore the art and science of 360-degree analytics. Learn how to navigate complex data landscapes, identify critical insights, and make data-driven decisions that drive business growth.

With Arpit & Rahul Sachdeva, experienced data analytics experts, as your guides, you’ll gain practical knowledge and actionable strategies to optimize your marketing campaigns. Discover how to measure what matters, improve customer experiences, and achieve sustainable business success.

By tuning into this episode, you can expect to come away with an understanding of:
  • The 360-Degree View of Customer Journeys
  • Overcoming Data Silos and Integration Challenges
  • Measuring Marketing Effectiveness Across Channels
  • Leveraging Advanced Analytics for Deeper Insights
  • Implementing Best Practices for 360-Degree Analytics

Featured speakers

Links & Resources

Transcription:

0:11:
Hello and welcome to another episode of the Revenue focus marketer where we discuss anything and everything related to
marketing as well as data.

0:21:
in today’s episode, my goal is really to help you guys uncover the hidden gems of 360 degree analytics that is literally a treasure full of insights.

0:33:
And we have two really important guests with us today, Rahul and Arpit that are gonna be our trusty guides that are gonna seriously guide us through this experience of learning more about this topic.

0:46:
They both have a lot of experience in this field as well as curiosity that has helped them, you know, become experts in
this matter.

0:54:
So welcome Rahul and Arpe.

0:58:
Thanks.

0:58:
So she thanks.

1:00:
She is good to be here.

1:02:
Awesome.

1:02:
So I’ll just give a brief description about both of them.

1:06:
Sir.

1:06:
Rahul is a seasoned professional with over 14 years of experience in business intelligence, data analytics and data engineering.

1:15:
He specializes in marketing sales and fintech.

1:18:
Besides his garden, he’s passionate about creating end to end data engineering and analytic solutions.

1:25:
Raul also collaborates very closely with cross functional teams and business stakeholders to implement self service analytics and architectures.

1:35:
So I guess you can all tell by now.

1:38:
He has had a lot of experience in this field and to join him, we have our usual in house expert that is Arpit with over 16 years of experience across data analytics, marketing and digital transformation.

1:52:
Ar pt’s approach to marketing is grounded in his belief that data should be used to make informed decision making with a keen focus on data quality and also A I use cases for marketing.

2:04:
He also works with cross functional teams that includes the marketing product as well as analytics team and works heavily on improving customer experiences.

2:14:
So we’ll just dive straight into really learning about today’s topic.

2:20:
We keep saying 360 degree analytics.

2:22:
But for a lot of like, you know, our users out there that are listening in Rahul, maybe you can help us start off.

2:30:
What really, what does it really mean by 360 degree analytics and what does it really cover?

2:37:
Yeah, absolutely.

2:38:
So if you think about so it’s, it’s kind of a panorama button on your mobile phone camera.

2:47:
you know, which you know, reads through and sees through every image that you point to and then kind of stitches it back to create a, you know, a 360 degree view.

3:00:
and it at every point or every image becomes, you know, a piece of a puzzle, you know, which comes together and create a cohesive narrative So it, it’s, yeah, it’s like a storytelling piece where you, you know, we together different pieces of threads of information and, you know, it, it creates a, a beautiful story how it, it, it, it flows from a point to another point and it tells something, it gives some information which is, you know, insightful and interesting to the audience.

3:33:
So, I mean, it has various flavors.

3:37:
you know, I, I, I don’t think we have reached 360 maybe, I mean, we call it 360 but it can be maybe 300 degrees or maybe 320 degrees.

3:47:
because there is a lot when it comes to you know, understanding the audience, understanding the landscape gathering every data touch point that is there outside.

3:59:
you know, we, we do try but again, I mean, it’s, it’s so vast and the technology landscape is so expanding that it can be overwhelming.

4:11:
So, yeah, it 360 degree when I say it’s, it’s start with in understanding your audience where the data is coming from, where the touch points are understanding, you know, how your customer interact with your product.

4:26:
And then, you know, working back to gather all that data in a cohesive form that, you know, you can collect and you know, you can report on, you can analyze later on wonderful and arpit maybe you can help us understand like you know, how does this understanding of the audience or the touch points really help with the marketing aspect?

4:47:
Like how do we really use you know, 360 degree analytics in measuring marketing effect effectiveness?

4:57:
Yeah, that’s a pretty relevant question.

5:00:
Glad that you asked.

5:01:
So I believe like 360 degree view of you talk about is essentially the customer the entire customer data platform that we’re talking here.

5:11:
And for me, marketing effectiveness and tracking along those lines is just one of the use cases that this three review will enable and within the marketing effectiveness.

5:22:
There are three things that generally we would like to sketch.

5:27:
One is who’s the user from where he came from like the channel and what did he do?

5:35:
So what’s the user action and your ability to stitch these three together can really solve the puzzle of marketing
effectiveness.

5:44:
And then there this might sound easy but it’s not.

5:50:
So who is the user essentially itself is a it’s a platform generally under identity resolution category.

5:59:
So you can imagine like the buyer’s journey has become very complex, right?

6:05:
And the data sets around those users and customers and prospects, they are siloed across various platform and identity can be your email address, it can be your anonymous ID, it could be your lead ID, it could be some customer ID, it could be some of your internal way to track users, right?

6:25:
And each of the systems will have their own way to identify a user.

6:30:
So you need a proper process and system and things in place that can help you identify the user in the first place.

6:41:
Now, the second bit is which channels, which channels drove that user to your website or wherever you’re running a
promotion could be some offline campaign as well.

6:55:
Now again, the data is siloed, right?

6:57:
In order to achieve that 360 degree, you need to consolidate all the channel level data.

7:03:
It can be a app platform in form of Google, Facebook, linkedin.

7:08:
It could it could be some of the events that you are attending like trade shows conferences where you would actually inviting the same user.

7:18:
And then finally, the third category which I mentioned is what happened and what was the action that user performed and probably that’s most critical one as a data set because that’s what you’re trying to understand what actions user perform.

7:35:
In terms of micro conversions and macro conversions.

7:39:
You got to have a way to really understand the entire funnel.

7:43:
So you need all the leading indicators as well as lagging indicators right?

7:48:
From like a form fill that can happen on your website to the booking that happened in your C C R M and everything in between.

7:58:
So long story short, back to basics, you need all those three things.

8:05:
Who’s the user, which channel he came from and what actually he performed and you need like a 360 degree view that
really enables you to stitch these three things together.

8:18:
Yeah.

8:19:
I think you’re right like it sounded very easy when you started off like we just need three things and we’re like, you know, resolved.

8:26:
It’s good to go, we can really gain more insights.

8:29:
But the more I understand about this, I realize it’s a much more challenging process.

8:35:
And I’m sure like, you know, there must be some challenges that people face while figuring out their 360 marketing
analytics.

8:43:
So are there any particular that sort of I know you touched upon silo data in different, across different platforms and channels?

8:51:
Are there others that you see as challenges?

8:57:
Yeah.

8:57:
So like like going further into all the channels, for example, let’s say we pick S C O as one of the channels, you
probably have a, if you’re probably investing on search, you will have four or five people and some agency who’s
supporting the whole S E O thing, right?

9:14:
No problem is Google certainly is the main search engine where you get 80% of traction.

9:23:
And Google stopped sharing data on keywords which is like a backbone for any any of the sort strategy.

9:30:
Now, the data exist in their own platform called Google search console.

9:35:
But it is again, pretty much high load in the sense that you hardly have any inability to connect it with the rest of the data sets.

9:45:
So the challenge is to really have some connection between the keywords that are driving traffic to your website and whether it all it is making any impact on the desired user actions, the third bit, whether it’s a harmful, whether it’s a lead, whether it’s a pipeline, so you will have to create some custom situation around that, that can allow you to stretch that journey.

10:17:
Now, other channel would be your social media channels.

10:22:
Now, social media channels come up with their own ways to report.

10:27:
And there are certain metrics that are very difficult to a particular channel and across your paid social strategies as well as organic social, you will have to have a clear way to segment data.

10:42:
And also you need some kind of data harmonization because it might happen that certain metrics are actually more
relevant for paid social versus organic social.

10:53:
So you need to have a well thought process by you are sort of working on your data or your schema the way you’ve
designed in order to get like a clear view, whether your social media strategy is working or not.

11:07:
So likewise, each channel we have to go through.

11:11:
And if I pick like offline as one of the channels, especially from week to week perspective, marketers spend a lot of time, energy and money in attending various events like let’s say you’re attending Adobe Summit or you’re attending a hubspot conference, right?

11:31:
It’s a huge investment.

11:32:
How can you connect the online and offline marketing or, and then really create a view that can tell if the combined back of digital and offline campaigns are actually leading to close one, right?

11:48:
So those are the things that we have to take care of when it comes to the trace it to review and the use case that we are solving for which is marketing effectiveness, tracking and all.

12:01:
And, you know, just to add to that, I think many of these you know, channels are now digital.

12:09:
you know, there, there was a saying from a marketer 100 years ago.

12:14:
you know, we, we, we know that half of my marketing tactics are not working, but we don’t know which half and that, that you know, that thing is still true even in today’s age where we have, you know, almost everything digitized, we have, you know, we have connected digital and physical worlds.

12:34:
and still, you know, those siloed data and disparate systems, you know, make us feel helpless sometimes.

12:43:
So, I, I think, I think it’s pretty relevant in today’s age.

12:47:
You know, how do we actually go ahead and connect those missing pieces?

12:54:
especially when you talk about you know, challenges with C R M data or, you know, marketing automation data that, you know, is another whole in your game altogether.

13:05:
And you know, it, it, it starts with the, I would say, you know, it starts with the lead.

13:14:
And, you know, goes up to close one but you know, the, the, the integration with other systems like you know, social media or it, it could be paid media or it could be trade shows, events.

13:29:
And how are we ingesting that data using technology?

13:32:
So you, there could be challenges like you know, they could be missing pieces of data or they could be inconsistent data.

13:41:
That is a big problem today.

13:44:
You know, people tend to enter you know, all sort of data in the systems but you know, each has their own version of how they feel like should be you know, they should be entering the data.

13:58:
For example, you know, country names or state names, they could be all different when you, somebody else could be using I S O format or somebody else you know, would be using full names of the state.

14:14:
So this is just an example but I mean, there could be any number of you know, version how somebody can write a single piece of information.

14:24:
So, yeah, I mean, it, it, it, it could start with as simple as missing data, incorrect data inconsistent or integration missing.

14:33:
And you know, there are some tactics tactics to handle those kind of situations as well.

14:40:
So for example, you know, o obviously out of the box integrations are one, for example, marketer and sales force
obviously do have a great sync.

14:50:
But what about other platforms which do not have?

14:54:
So, you know, you kind of have to build out you know, custom integration using the API I hope they have API but if not, I mean, it, it, it might lead to another you know, puzzle which is either you export the data or you do you know R P A process where, you know, you know, you AAA robot is basically opening up the application, downloading the data, sending it to your the other application to sort of join that.

15:31:
Only if you find a missing, you know, only if you find this integration point for example, a lead or a contact or
opportunity or account.

15:42:
And this, this elevates up to a point where, you know, if you’re, let’s say running A B M campus, so account can
consist of, you know, multiple contacts.

15:55:
you know, sometimes there are even leads with the same account.

15:59:
And how do you really map the touch points that are coming in on, let’s say on a social media A B M campaign back to your C R M or marketing automation system.

16:11:
So, you need to carefully assess you know, which accounts you’re mapping to which or are we getting any known contacts?

16:22:
are we getting any known leads?

16:24:
Well, if not, you know, we might have to create new ones otherwise, you know, there should be an integration in place which maps the existing touch points to the you know, the existing leads or contacts.

16:41:
And this could happen you know, across I would say whether in the system or outside the immune system.

16:51:
I mean, you can do this kind of analysis outside using some of the E T L tools that are available which could, you
know, join the missing pieces.

17:01:
It, it may not essentially put that back in to the system, but it’ll really help to stitch the data across.

17:10:
put it in a kind of a warehouse, a data warehouse where, you know, we can report on it and you can see the missing
pieces I just want to add like you, you brought in A B M specifically.

17:22:
So one challenge with sales force specifically is that you’ve got leads and you’ve got contacts and essentially it
represent the user only, right?

17:33:
And lead, generally speaking, becomes your dumping ground for any incoming a record of the user.

17:40:
And generally speaking, there’s a lot of duplicates as well and then sales force does not offer a direct relationship between your lead data and your account data.

17:52:
And that’s where the problem arises because marketing team is making a lot of efforts to drive leads.

18:00:
And just because there’s a process gap in terms of how the whole sales process works.

18:07:
And how you convert lead to a contact and opportunity.

18:11:
There, there are a lot of times that there are often leads and the sales team really don’t care about following the process.

18:21:
So coming back to the whole thing, like who’s the user, which channel it brought and what user action was performed, that information exists in the orphan lead object like this campaign that lead was engaged with.

18:39:
Now, since the lead is not connected with the account directly, you’ve lost that information, there’s a massive data linkage.

18:47:
So again, you need some process in place.

18:49:
that allows you to lead to account matching, especially if you’re using sales force that is not coming out of the box.

18:57:
So that, that that thing we have to keep in mind.

19:01:
And another topic around challenges I would say is the whole the browser restriction thing that is coming up.

19:11:
which means that obviously third party party cookies are no more there.

19:19:
And then like at least the latest update from Chrome, it’s it’s going down and the other browser restrictions like
Mozilla, for example.

19:31:
So if you come to a website and there’s an anonymous user, so your tools like Google Analytics or Adobe Analytics or Amplitude, you name it.

19:44:
Obviously, they start with anonymous id for that user.

19:48:
But if the user come back after, let’s say seven days, it’s a net new user because the browsers delete those bookies.

19:57:
So again, you’ve missed that identity resolution thing that I was talking initially.

20:03:
So what can we do?

20:05:
There could be some workarounds, there could be methods like finger printing that still give you some level of identity resolution.

20:14:
But you need to have a clear view how you managing the whole user data.

20:23:
OK?

20:23:
To further add on you know what you talked about missing piece of information and missing in data leakage.

20:31:
to some extent, I think some of these platforms also enable or, you know, also our culprit to that of, for example, sales force it, you know, might not store all the activities for in the entire period.

20:49:
It may store only till 90 days of activity and you know, beyond that, it will just flourish that data.

20:57:
So, and you know, same with moto as well.

21:00:
There are some limits to how much activity data you can retain.

21:04:
So all that user behavior that is gone.

21:07:
And that is crucial.

21:10:
you know, to understand your audience understand the touch points.

21:14:
If your sale cycle are longer, maybe, you know, more than a year or so, that piece of information is crucial.

21:22:
And you know, these are like all online channels.

21:28:
If we talk about A T L marketing, which is more like engaging the mass audience using TV or print ads, the challenges are even bigger.

21:39:
You know, you might have to, I mean, not every person who sees your ad, you know, is gonna call you or is gonna contact you with a touch point, but they might have some impression on the mind.

21:53:
It’s impossible to you know, gather all that data.

21:57:
But yeah, it’s, it’s pa it’s p you know, it’s part of the puzzle in the sense.

22:02:
At least it, it, you know, it, it, it grows up to a level where, you know, they might wanna go onto the website and you know, fill out a form and then it becomes, you know, a, a known piece that you can fit in the puzzle.

22:21:
So, yeah, it, it, it starts off you know, online that is a little bit easier.

22:29:
you know, goes to more rough edges where, you know, we are your audiences, mass audience using A T L marketing.

22:40:
So, yeah, it’s, it’s, it’s tough, definitely.

22:43:
And, you know, all these, you know, rules and regulations, these are really, although, I mean, it’s i i it’s for the good obviously, but it’s definitely brings up another challenge every, you know, year or so that you have to redesign and rethink your strategies, you know, and rethink your how you’re sinking the Red Cross.

23:09:
Definitely, that’s really you touched upon some great points there.

23:14:
I think now our audience is able to really understand like, you know, the type of challenges that come across.

23:21:
But for people that are still in the process of implementing and do see the benefits.

23:27:
What would you guys suggest is like, you know, best practices that can maybe like if not, like, you know, remove all the challenges at least help them like, you know, miss out from the initial challenges that they might face, maybe Rahul, you can start us off.

23:47:
OK.

23:49:
Yeah, I can start a Ra in China.

23:51:
So I think even Raul, you started off with saying that there’s no perfect 360 degree, right?

23:59:
And the whole idea of 360 degree itself is broken in the sense that we’ll have to start somewhere, right?

24:07:
And the whole idea of crawl walk run can be like visualized when it comes to the overall data maturity.

24:15:
So maybe we start with, let’s say I’m not even saying 3, 300 degree.

24:19:
I’m saying maybe 180 degree or even 90 degree.

24:22:
And just to put the this in perspective, it could be, can be stretch, let’s say your C R of data, your behavior data, and maybe your channel data, those three data sources so that at least you can connect online and offline together.

24:39:
And then you also know where you’re spending and how much you’re spending.

24:43:
That could be a great start versus thinking that I have 50 different data sources.

24:47:
Let’s create a big data platform and we run a twelvemonth project.

24:52:
And then in that 12 months, you’re not delivering a single insight, you’re not delivering anything that is actionable.

24:58:
So it’s always good and that’s how generally we work with our clients is let’s have something in place and then let’s we, we have an objective of a North Star to have T 60 review.

25:11:
But what would it look like in the first three months and what insights you will get and what are the use cases we are trying to solve?

25:20:
So once you’re good with the understanding of the use case you’re solving for and then you get an alignment with, with the stakeholders, your client, you know, like in order to solve those use cases, where does that data exist, what are my priority data sets that I should ingest and that sort of model it and make it more actionable and try some insights out.

25:42:
That would be a good starting point.

25:45:
The other view is the understanding of where you are in your overall data maturity.

25:53:
See even within the framework of marketing effectiveness, there could be letters of maturity, right?

26:02:
maybe you have certain clients that are not even doing platform specific conversion tracking, that is the starting
point of any kind of marketing effectiveness.

26:12:
And that’s like a base.

26:13:
at least you’re telling these signals to your to the Google ad words and other platforms that these are the conversion action happening from these campaigns.

26:24:
And so that the machine learning algorithm can kick in and that can optimize the campaign.

26:29:
So that’s the base level.

26:31:
If once you cross that, then you do cross platform attribution, right?

26:36:
So you start with a tool, let’s say like a Gao where you have a last last di click attribution and then you at least have now one view that covers all the channels maybe still you’re missing cost data for certain channels because g airport does not integrate everything in one place.

26:55:
But you know, at least the initial conversion action, the leading indicators of form and leads.

27:01:
But as you mature more, you sort of build a centralized marketing database and create your own marketing data model that not only bring, brings in all the channel cost data together, but also bring in the different levels of user actions, call it leading and lagging the data in terms of how many forms we got how many leads were there are these marketing qualified leads?

27:30:
Are these sales qualified leads?

27:32:
Are we able to create an opportunity?

27:34:
What kind of pipeline we built?

27:37:
And does that pipeline actually converted?

27:40:
And each of on those steps, what has been the fallback?

27:44:
And how much time we are actually investing in each of these stages?

27:52:
How can we actually accelerate the whole process?

27:55:
Where are the friction points?

27:57:
So those kind of analysis can be easily done?

28:00:
Now again, like one thing we touched upon was browser restrictions, right?

28:07:
No, the whole use case of marketing effectiveness and just hopping on the same thing.

28:14:
But you have on one side, multitouch attribution models of the cross channel attribution that fully relies on user
level data.

28:26:
Like if you don’t have ability to stitch user channels and the target metrics or target action, you can’t do multitouch attribution.

28:36:
And there is always something that you won’t be able to practice as long as you’re aware and using multitouch
attribution and gain some insights.

28:46:
And now you want to take it to the next level, there are ways to do it.

28:50:
And the other one of the ways that is getting popular is medium modeling, although it’s it’s old pretty old way of
measuring the marketing effectiveness, but it is getting more traction.

29:03:
Now, given even your online digital tracking has some limitations.

29:09:
So instead of connecting all those three things together in case we are missing the user level data and we are not able to stitch it.

29:18:
Can we do some kind of regression and see if we are spending this much across four or five channels and you’ll be
getting this much of sales, what is the contribution of those channels?

29:29:
And that’s what media mix modeling will solve for.

29:32:
And that’s how again, depending upon your maturity, you can really go to that next level.

29:40:
Now, beyond that, I think one thing that is very important is the culture.

29:47:
So one of the things that we have seen is that you’ve invested a lot on all this data, you’ve got great dashboards, you did multitouch attribution, you also did media X modeling.

29:58:
Now your data analyst or your head of analytics is saying that go and cut down on this channel and invest more on this channel because it’s not performing.

30:10:
But is the marketing team listening to that recommendation?

30:15:
Are they taking actions?

30:17:
And are they actually challenging the model itself?

30:22:
And that is very common?

30:24:
Like you will challenge if it’s not really telling you something that you wanna listen, the data will tell its own
fact, right?

30:34:
And then it might be good, it might be bad, it might be questionable to your end job, right?

30:39:
So what we need here is a culture and that culture will come from the top.

30:45:
So while we are taking on big data projects of 360 degree customer view.

30:53:
We need the blessing from either CEO or somebody at sea level that this is company variety and everyone has to follow and it, it, it is not that at the end of the day, you’re not taking actions or you’re just taking it as one of the reports.

31:12:
Yeah, absolutely.

31:13:
And you know, the such kind of data first, organization culture is is the need of the R.

31:20:
I’ll touch on the other aspect you know, which is data engineering.

31:26:
And you know, once you understand where your audience is and you know, we have kind of integrated all those channels through some or the other way if whether API S or doing, you know, R P A S or, or, you know, some other kind of integration, it is essential to you know, transform the data and you know, do a data validity checks.

31:52:
whether the data is that is coming in is you know, is, is reliable complete.

31:59:
you know, and basically, if it’s missing anything with respect to you know, the, you know, the platform itself is
collecting data which is you know, let’s say inconsistent, it will be garbage in garbage.

32:15:
So you have to you know, obviously implement on the platform first the data layer and you know, then do the data
integration, automate the processes so that, you know, everything that you do or everything that you collect, you know, comes at.

32:37:
you know, it’s, it’s, it’s in a usable format.

32:41:
And it’s you, you should be employing an detection, you should be employing checking for outliers, simple checks like null check or empty fields.

32:52:
You know, those should be good enough.

32:54:
But while you’re you know, ingesting the data and, you know, let’s say putting into a data lake or data warehouse,
there could be you know, there are many coming to technology landscape, they could, there are many tools and technology available to do that, you know, widely used open sources.

33:13:
Python.

33:13:
you know, there are many implementations like Apache, NIFI or Kafka or there are, there are a bunch of other
proprietary tools as well.

33:23:
You know, you can the there are for simpler tools like T prep.

33:28:
If you’re just pulling in from a couple of sources, you can pull in and do some quality checks and then you know,
transform the data, put it into the reporting.

33:38:
You can, you know, use two psychotics as well which have, you know, on the E T L side as well as advanced analytics.

33:46:
you can use nine which is again open source talent.

33:52:
I mean, both have open source and community and paid versions.

33:56:
So, and after we have collected the data, you have checked for quality.

34:01:
you put it into data lake or data warehouse.

34:05:
you know, it, it depends.

34:07:
So data lake is mostly where, you know, you put your raw data.

34:10:
So as soon as you pull in the data from the system, you put it in a place where you can store it.

34:17:
And further you can do E T L operations on top of that to convert it in the format where you can actually use it,
create insights out of it and you know, build reports and dashboards out of it.

34:31:
So that is the process which is kind of a middle layer.

34:34:
And it, it, you know, deals with you know how you integrate your data across the touch points, how you make it
redundant, how you make normalization to make sure that your data is reportable.

34:54:
And those are the things, you know, you’ll need to take care when you use that data for, for the analysis.

35:01:
It could be de descriptive analysis that could be predictive.

35:04:
you know, you might be creating the variables out of the data for predictive modeling.

35:09:
So, and for that, you’ll have to have that all the data in place in a single draw where you know it, it touches all the touch points under the same identity.

35:22:
you know, like you’re talking about having the same fingerprint.

35:27:
So, yeah, I mean, that’s, that’s another aspect that I can think of.

35:32:
It’s really important after you have, you know, identified everything and you know, started starting to go to the next step and that’s really helpful to know.

35:45:
I also had one other question around this.

35:49:
So, like, you know, I think in today’s world A I is really moving along a lot of things within your guys’s field.

35:56:
Are you seeing any impact that A I is making?

35:59:
Like, I know there are so many things to take care of.

36:02:
So are there different tools that people can now utilize or do you think A I hasn’t re tapped this area yet?

36:13:
Yeah, I can maybe start.

36:14:
So I would say within our view of the whole marketing analytics and understanding of user behavior, what customers are doing and doing any kind of root cause analysis, like one thing which is very common is that like, let’s say your S C O traffic has come down a certain channel is not performing.

36:42:
So one question that really comes in why like we always want like why it happened?

36:47:
So generally speaking, it takes like weeks to understand like where all is the data you run a bunch of SQL queries, right?

36:56:
And then and let’s come back with some of their analysis and recommendations and whatnot.

37:03:
So I think with with A I and especially generative A I coming into picture what is really happening is that it is
getting more democratized.

37:13:
Now, a person who does not know SQL can still talk to the entire marketing database.

37:21:
And there are ways to do that in the, in the sense that just like a simple chat interaction you do with CHA GP T, you simply ask that we’re seeing traffic going down by 30%.

37:35:
What are the contributing factors?

37:37:
And then the chatbot will come back with some of its analysis based on whatever data has access to and really give some overall analysis.

37:49:
And what you really saw there is that maybe it’s not 100% perfect, but it is there 60 70% or even 80% in some
scenarios.

37:59:
So you get a good head start and on top of it, you can really build your your point of view.

38:06:
So I think that’s one area where you’re seeing a lot of operational efficiency.

38:11:
And it’s really helpful in making the whole 360 degree view of that data available to even business users so that they can interact directly.

38:24:
And they’re not at the mercy of the technical team who has their own bandwidth challenges, their own turnaround time.

38:33:
The other areas which I think, which are closer to marketing is I would say your ability to grade the incoming leads especially from B to B marketing perspective.

38:46:
The traditional way of doing is more like you define the rules and those rules can be biased that is driven by your, your own gut feeling.

38:57:
But can we make it more data driven?

39:00:
Can we go back and see what all leads converted in the last two years and understand through some machine learning
algorithms that what were the contributing factors?

39:14:
What were the attributes that are leading to it to it?

39:17:
And maybe we put that into two buckets.

39:19:
One is all the demographic factors like what are the industries?

39:25:
What titles?

39:27:
What is the revenue range?

39:29:
What is the company type?

39:31:
What level of like titles within like the person that we’re targeting?

39:37:
And they could be another bucket of factors which are more activity driven, which means how many times he the person came to the website, what all content he consumed the content was at?

39:53:
really which level it was?

39:55:
Is it top of the funnel?

39:57:
Is it middle of the funnel or at the bottom of the funnel?

40:01:
All these signals will really tell you whether the lead is good or bad and you get more accurate way to grade leads.

40:10:
And what it really solved is again, some operational efficiency because the sales team time is pretty expensive.

40:17:
They always have a bandwidth challenge.

40:19:
So can we give them leads that have likelihood to convert on like in range of, let’s say 70 to 80% versus less than 5%.

40:30:
So those, I think those are a couple of use cases which I believe are closer on the data science and the I side of the house and, and just to build on top of that, even, , you know, dashboards which are meant for executive are finding a great use case here.

40:51:
You know, I think the traditional approach has been to create some sort of some dashboards, you know, which and a, a data warehouse which, you know, help them to do self service analytics as well.

41:05:
I think is A I J A I is gonna change that.

41:09:
And in a beautiful way where, you know, they, they won’t have to actually deep dive in the potion of data.

41:20:
They just need to, you know, ask their questions in English like language and you know, they would have their answers.

41:28:
So I think it’s, I mean, it’s getting there definitely many companies are using it and you know, you know, we are
working with a few clients that you know, have started using, we are building that for them.

41:43:
And I, I think it’s, it’s, it’s going to be a game changer in terms of how easy it is to extract those insights.

41:53:
when you write off the you know, that hay stack of data.

42:01:
No, I think that was really wonderful and really helpful for us to really understand like, you know, the impact that AI and specifically Jenny I can also make on really improving operational efficiency.

42:14:
And like, I personally know a lot of people within sales that, that are really like, you know, you as you mentioned short on the bandwidth aspect.

42:24:
So just being able to like, you know, procure, do the work for them and just give them leads that we have more
likability of converting would really impact the whole organization in a positive manner.

42:38:
We’re at the point in our show where I think we’ve gained a lot of insights today.

42:44:
So I just wanna thank you both for taking the time to do this.

42:48:
I think just to wrap up, I also wanna say that we went through a lot of different things today.

42:55:
We’ve dealt into the intricate world of marketing effectiveness and data integration from like, you know, panoramic button on a camera.

43:04:
Starting off.

43:05:
We also kind of mentioned how like, you know, we want it to be at a 360 degree level, but we’re not necessarily there yet.

43:14:
And we even discussed like different challenges that are sort of prohibiting us from reaching that 360 degree level.

43:24:
But regardless of the challenges, we were also able to discover certain solutions that are gonna help us.

43:36:
I also think from like, you know, Rahul emphasized on the need for a top down leadership and robust data engineering process while shedding light on the role of A I in democratizing data accessibility and enhancing operation efficiency.

43:55:
So these were a lot of different use cases that we were able to really understand.

44:01:
I think when we talk about data, a bigger chunk that always comes out is the whole garbage in garbage out process.

44:11:
And here like, you know, the importance of really getting your foundation right, makes a huge difference.

44:17:
I I know while working with the teams within the product is with the growth, we often talk about how this is the first step that we, you know, have to like data governance and cleaning that data is such an important aspect and getting the foundation wrong will not give you the insights that would be actionable or you could really rely upon, right?

44:40:
So I think in today’s world is getting that aspect right?

44:45:
The fact that data is really out there, but making sure it’s reliable data, it’s something that, you know, even
marketers and like c level executives can really trust to make that decision upon is really important.

44:59:
So I think for us to just be able to understand this whole complex landscape today was really effective.

45:09:
And my biggest takeaway is also like things to keep in mind like such as like the stage of the enterprise and like
their data maturity that plays such a huge role in the steps that they should take in the practices that they should follow.

45:26:
So I, I definitely can say here confidently that I learned a lot from today’s session.

45:33:
Any last, you know, pieces of advice or takeaways from your guys’ end before we wrap up, I think like for any, any
larger projects around 360 degree view of customer, The end milestone is the organization ability to how about the culture that will basically determine if we are making any business impact or business value with all those efforts like the foundation we understand, like all those 45 pillars that you have to have like a good data ingestion integration process in place.

46:14:
You should have a mechanism to control the data quality, completeness accuracy, you should have mechanism to model the data the way it can answer the questions that business users have and then your ability to do the proper visualization as well as activate certain maybe A I and machine learning use cases.

46:36:
All that is great.

46:38:
But at the end of the day, are we taking actions?

46:43:
Are we using the data the way it should be or it is just one of the reports that comes to executives email box every week, every month and it is just like a dead tree, right?

46:59:
So I think there’s some some again, thinking required, some focus is required and some direction is required right from the top.

47:10:
Somebody who’s sitting at the sea level, he should be the one who sort of has to evangelize across the organization.

47:19:
Otherwise, everything is a false promise.

47:21:
So I don’t want to end up in a negative note, but it’s critical to understand that why we are doing this and are is the entire organization mature enough to consume this data and take actions and sort of rinse and repeat the whole process.

47:42:
Definitely, I think a lot of us like, you know, in today’s world, focus more on the technology and procuring the
technology and getting the steps, right?

47:52:
But the human aspect, regardless of how much A I I think succeeds and processes things further, it’s still important for leadership and like, you know, the people that you employ to really grasp this concept in order for it to move on.

48:07:
Otherwise it’s just another email, like you said, it’s another slide deck, another report just sitting in there without really like any value coming out of it.

48:17:
So thank you so much Arpit any last words from you, Rahul.

48:22:
Oh I think that was pretty much bang on.

48:26:
Perfect.

48:27:
So again, thank you so much to both of you and also to the users that tuned in to, you know, listen to this episode.

48:35:
I hope this has been one episode that they really gained from and I hope to see you guys again and participate in more discussions with you both.

48:44:
So, thank you so much for today.

48:46:
Thanks.

48:47:
So you’ll be here.

webinar

Additional Resources

Unleash the Power of AI with DiGGrowth. Explore More!