Rubi Das

Myself, I have more than 7 years of experience in digital marketing and 6 years of dedicated experience in programmatic advertisement. I started my journey from Google, which was an outsourced project to Accenture. I was working as a subject matter expert for DV 360 ad exchange. At the same time, I was collaborating on other digital solutions, the Google Marketing Solutions. I was partnering with Centimeters 360 and other solutions to troubleshoot campaigns across agencies and brands, and collaborating with Google account managers to help and share resolutions, supporting the creation of new content for help center articles. I also had a team of 5 members, where I shared our learnings on different cases and taught them about the GDPR process. So, I was leading and dealing with a lot of marketing solutions in the Google marketing portfolio. From there, I switched to PubMatic. PubMatic is a SSP platform. In an SSP platform, it ideally brings publishers' inventory onboard, and allows many demand partners to access that inventory. I managed business for the demand side. I was an agency account manager, and I looked after top holding groups and agencies like Mindshare, Publicis, and Havas. I helped them buy efficiently through the PubMatic SSP platform in a programmatic fashion. Then I moved to The Trade Desk. The Trade Desk is a DSP platform, a demand-side platform. I was the first account manager at The Trade Desk in India, and I managed a portfolio of accounts, typically top holding groups. My idea was to pitch and help them with different strategies that might help deliver their best marketing objectives. To meet their marketing goals, I devised strategies across platforms and helped with end-to-end execution, and also allowed them to have a seamless experience across the solution.

So to ensure exact pacing of a programmatic campaign, we need to have a clearer understanding of what pacing is about. Pacing is the way you want to spend your budget in a particular period. So let's say if one wishes to spend a budget equally among the total flight days when even pacing has been set up, then there is ahead pacing and there is ASAP. Now to ensure accurate pacing, we need to understand what the goal of the client's marketing campaign is. The accurate pacing for a campaign budget would be even or ahead. Even the budget will be distributed evenly among the days. In ahead pacing, the idea or algorithm is set to spend more than 100% for a particular day. So if we distribute our total campaign budget for a day and see it daily, it will only ensure to keep the benchmark up to 100% consumption for that day. But then it does a bit more serving or buying for the client for the day so that at the end of the campaign, you don't have a deficit and you still have a smaller portion which you can spend on the last day of the campaign. Hence, these are the ways to accurately do a pacing check. Of course, nevertheless, you have to do a lot of reporting to understand how the pacing has been done and what optimization steps are needed to ensure the pacing is up to the

In terms of a complex programmatic campaign, I would like to highlight one campaign which had multiple platforms where it was required to be served. It was a campaign of both video and display creative assets for a client to G and J. The idea was to do a full funnel programmatic campaign wherein the 1st batch of the 1st phase level of the campaign was to bring brand awareness. And there we had display creatives as well as video creatives. So very effectively, we had to manage that budget. What is the display budget we need to keep and what is the video budget so that my brand awareness is very accurate, and I'm able to get a bigger chunk of audiences which are quite relevant to the brand. And then the next leg of the campaign was creating a user pool or the audience pool and then targeting them. So it was a branding, consideration, and conversion. In consideration, the idea was to retarget those sharp audiences which we collected in our 1st leg of the campaign during awareness and then retarget them with a shorter edit. While the messaging was different in the branding, we wanted to bring a different messaging for the clients who have already been exposed to the branding campaign and bring more likelihood towards the product. So the consideration campaign was set up and then followed by the conversion where we had pixels placed. And there were products which were brought for the brand. And the most complexity was that it was not just for a mobile platform. It was across desktop, in-app environment. It was also on the CTV platform and the OTT's platform. So the whole media mix was multidimensional, and it had a lot of components where we had unified reach to be brought for the client, total number of users to be dedicated, and then bringing a conversion for the campaign.

To maintain effective communication and coordination among the remote ad operations team, it is very important to have regular checks. Communicate and build a strong relationship with those teams, whether they are internal stakeholders or external stakeholders. Have a regular check to understand where the pain points are. And then try to coordinate and help them in solutions that can effectively result in a better outcome. So for us, we had a team based in Singapore, and there was an operations team. To manage that team very effectively, we ensured that we were overcommunicating at times so that there was no miscommunication happening. There were regular connects with the client or internal stakeholders where we were able to share the current progress and where we were heading for the next project. And then, we streamlined it. The idea was to streamline it as much as possible so that there were no loopholes. There were no delays in the execution of the project, and the outcome was the best. A weekly check or regular check would be very essential. Communicate as much as possible and have a very smooth conversation window so that the person can also reach out to you and coordinate the status, and keep the people in the loop so they are also aware of the current status of the project.

So whenever we need to do a test to evaluate a new bidding strategy, we first have to allocate the test budget, then test out the creative. If that is a set of creative - display or video - allocate that, check your markets, what you want to test it up, and set up a bidding strategy which you have been doing for the past campaigns. For example, this test needs to be and also ensure that you have a defined timeline within which you have to produce a test. A test cannot be done in a day or two to conclude a result. A test is something you need to have a defined or substantial time period because the bidding strategy keeps on changing based on historical experiences. So to understand how the bidding strategy in DV360 is working, we need to understand what are the inputs and targetings we have fed in. And over a period, how it is reacting or learning from those inputs which have been fed. So while setting up, ensure that the tags are in line, ensure that the creatives are appropriate. Ensure your budgeting and pacing have been set appropriately. Check on the other targetings which have been set up in terms of geolocation, in terms of any demographic, in terms of any audience segmentation. And then try out that building strategy for over a 7-day period. And then evaluate and test the results.

These kinds of scenarios do come where there might be a conflict between the client's marketing objective and the limitation of the platform's specific capabilities. One classic example is a client who wanted to have a unified reach across multiple platforms, including DV three sixty, and other platforms which I have worked on. The client wanted the user to be mapped across CTV, mobile, desktop, and any other digital devices they were using, and have a unified reach across them. Unfortunately, at that point in time, the mapping was not feasible to make it a unified reach across many platforms like CTV. And then the client also asked for a unified reach when doing work in DV three sixty and YouTube, and integrating all platforms into one panel, which at times becomes very tricky. We need to keep educating the client, and to tackle such situations, the best approach would be to let the client know exactly what the capability is all about. It's just a matter of learning, which the client might not be able to have because they are coming from a different background and are dedicatedly working on a platform. So, they know the nitty-gritty of the platform. In that situation, let the client understand, give them a learning session on how the platform works, show them a walkthrough, and then help them understand what the platform capabilities are and how we can do it. To meet that marketing objective, we can devise a different strategy altogether. That could be another solution that, if this strategy or this approach is a platform limitation, we can try out a different strategy that can also support or complement the marketing goal. But in such a conflict where the client is unable to understand the limitation, we should definitely take some sessions with the client and let them understand what the capabilities say. These are the case studies, how we can run those campaigns, and how we can actually bring the client's objective to a successful outcome. So I believe the education session and connecting with the client is the most effective way, so that you keep the client in the loop and are able to deliver what the platform is dealing with.

In terms of A/B testing, creatives, we were actually doing a display creative check with different content. And we were also doing a check across how display creative works with respect to the video creative, with the same messaging across exchanges. So while targeting different audience segments. So while doing an A/B testing with the creative, within a programmatic ecosystem, we had different sets of creatives in terms of measuring capabilities. But the idea was to see how it works across the exchanges and what's the best efficiency in terms of bringing or using that creative for the campaign. For example, if we have to use pixels placed in the creatives and we have the same pixels for this same set of creatives, but then the messaging was different. Although the brand was the same, the messaging was different, and we had different pixels. In that case, we need to understand how to segment the audiences to see which creative works best. And hence, you can conclude that some of the creatives are able to bring more CTR. Some of the creatives are not having sufficient content that can attract users, and hence, it's not able to deliver better CTRs. Some of the creatives are not quite viewable, not showing the content was not viewable enough for the client where the page is delivered. So these are the parameters where you're doing the A/B testing with creatives, which is performing best, and see how it has worked with different geographies and then conclude that around which are the creatives, what are the attributes that are helping them to deliver the best.

Every time a new inventory partner is onboarded, either on the SSP platform or DSP platform, the effectiveness of a new inventory is evaluated based on what pricing they're able to deliver, what key performance attributes we are able to achieve through those inventory. For example, if there is an inventory where we are able to get great CTRs on display campaigns, that's one way to understand if my inventory is effective as per the advertiser. If I'm able to get the required KPIs fulfilled, if the pricing is good enough, and if my messaging is getting delivered to the right segment of audiences. So these will be the primary KPIs to understand if my new inventory partner is effective. And the last point is very crucial: how much inventory is available. That is very small. Although there are relevant audiences available as per your brand. But if the inventory itself is very low, it might not deliver the purpose. So, having inventory availability in mind, how easy it is to access, in what volume it's available, how the audiences are on the inventory, and what performance I'm able to get while delivering on those inventory. These are the parameters on which one should evaluate the effectiveness of new inventory partners.

In terms of advanced optimization techniques, not just in DV360, but in the Google Marketing Platform, I would say building up a first-party data solution is also one of the key factors that one can use through Google Marketing Solutions. So many people play a very limited understanding of how they can use the pixels and collect data from their first-party audiences. Now since we know that cookies might get duplicated in the future, the idea should be to build a first-party database. And leveraging the Google Marketing Solutions, you can actually place pixels and collect data and understand your users – who are your fine users, who are coming to your page, and are honest to the brand. And then targeting and retargeting them would be one very advanced optimization technique. One can use it. Doing audience segmentation is very important. So not just in the readily available Google audiences, you can also access a lot of third-party audiences which are very reliable at times. And they have a very beautifully defined signals through which they actually collect those audiences. So that could be one way to reach out to a very defined but targeted audience. And other optimization techniques would be diversification of the campaign. When a brand has been delivering on a particular platform through a certain format, the best way to get results is nowadays to diversify – do a mobile and CTV and desktop targeting, which helps you to give a more unified perspective into your reach. So that is one optimization. Check regularly the reports. What are the inventories you are able to deliver? What are the segments you are able to deliver most? What are the markets you are able to do best and do? So that is one optimization one should be able to deploy while working in Google Marketing Solutions. Keyword marketing is also a very great optimization – what are the competitors to which you want to show your ads? What are the keywords where you wish to deliver your ads? What are the sets of audiences? And the final would be doing a site-level targeting. Doing a site-level targeting can always help ensure that these are the sites where a certain set of audiences are coming. And I want to ensure that my messages are reaching that set of audiences. So that would be one. And brand safety, definitely – doing brand safety through multiple platforms is very essential. And for some complex campaigns, I would say that brand safety is a must.