There’s a curious thing about DMPs: Publishers generally understand it’s a good idea for them to be aligned with one. A DMP is supposed to help publishers get the most value for their proprietary data, and to pass along the value of its audience extension capabilities to advertisers. But at the same time, publishers generally can’t say much about how DMPs have driven revenue for them. At AdMonsters, we’d gone forth into the publisher community in search of useful DMP case studies. While we certainly heard a few, they were voices in the wilderness. Far more publisher ops said they’d enlisted the services of a DMP, and they were at a loss to figure out what to do next and how to make any money from these things.
That’s not how it was supposed to work, we thought. So we sat down with Rishit Shah, Lotame’s VP of Product and Strategy, who offered to clear up some things for confused publishers. His advice: Publishers need to put in some real work if they want DMPs to improve their revenues. I asked him to help draw out a plan of action for publishers.
BRIAN LaRUE: If a publisher client were to mention to you they didn’t know what to do with their DMP to make more money, what are some beginning suggestions?
RISHIT SHAH: These are some of the first action items we try to accomplish in the first 30 days of the onboarding: What problem are you trying to solve? Some of the smaller publishers have scale issues. You need a certain kind of scale to be able to incentivize an advertiser to buy your audience: “You’re trying to reach people in this age category who are in market for a Toyota. I can deliver you this many uniques on my inventory, but I know these same people also browse across these other inventories that I can help you deliver.” That’s how they can increase their scale using the DMP, by creating audiences combining their first-party data and our third-party data sets and essentially extending their audience.
The other problem, from a strategic and process standpoint: What is your internal process to leverage the DMP? How are your teams structured in order to be able to sell audiences? Who creates those audiences, and how do you communicate across teams to execute on an audience-driven strategy?
The DMP by itself doesn’t become a turnkey solution. It requires thorough thinking—what kind of one-sheeter sales people will use to go out in the market and sell audiences, who creates that list of 20 or 30 audiences they should go to the market with. If those are the same generic audiences that an advertiser can buy pretty much everywhere, then it’s not going to work out for them.
BL: Can you tell me more about what it means for a publisher to play an active role with the tech vendor?
RS: Engagement between the teams is one of the most important driving factors behind a successful DMP strategy. I think it starts with a publisher understanding there needs to be a resource investment on their end. And there are two kinds of resources we see publishers most needing. The first is somebody who is dedicated to finding out what granular data points we are collecting and making some sense out of it, and then creating those 20, 30 niche audiences that could be a value proposition. All the nuts and bolts of where the tags are set up, what data points are coming in, how many uniques—all of that is something the account management team is setting up. The way we have done this is having a weekly call for the first 60 days to go through all the stuff the DMP is capturing on their properties, what tags are being set up and so on.
The second is a training piece. We have a DMP playbook where we start with the basic core DMP functionalities and then eventually go toward higher-use cases around cross-device and campaign optimization and so on.
The other piece is getting the sales people involved. Some of the most successful publishers we have seen across our client base—they are successful because their sales people get it. They speak the language of the audience versus context. Helping them with the reporting tools to create a story around how they should be talking to an advertiser is key, as soon as the audience-building is done.
That active engagement is absolutely necessary. A DMP by itself wouldn’t create revenue for you unless and until you really start utilizing the audiences and spread it out to your teams across the org. And there are so many examples where we are seeing publishers whose advertising revenue increases by 50, 60% the next year, because they are able to bring so much more intelligence to the table for an advertiser.
BL: When last we spoke, you mentioned the importance of publishers getting away from thinking about their revenue in just a contextual way, and more about their role in an audience targeting strategy. Where do you see publishers struggling with that thinking, and what are the benefits of getting past those challenges?
RS: Fundamentally, they need to think like advertisers. If a publisher right now is just focused on delivering x number of impressions, it’s an extremely short-term thought process and an incorrect mindset. Publishers need to start thinking proactively about what KPIs advertisers are trying to get out of this campaign. And this requires some probing, really getting in sync with the decision-maker on the advertiser’s end: Is this a performance-centered campaign or a brand campaign? Secondly, talk to the agencies representing advertisers. Say, “If your KPI is conversions, how about we bring in some more machine learning optimization? Give me some of the core users that you think you want to reach. Let me create a lookalike on my inventory, so that we can reach the best user base, based on people in the past that have purchased your product.” If the advertiser can get 1,000 conversions out of 2 million people instead of 500 conversions out of 10 million people, the advertiser would take the first one!
BL: I want to ask you about the differences in approaches publishers might have toward using a DMP, with the legacy publisher brands that rely heavily on direct sales at one pole, and bootstrapped, 21st-century, purely digital pubs at the other. Seems like it might involve different strategies depending on where you are on the spectrum.
RS: There are different tactics around how to go to market. In my mind, the big guys are those who have at least 50 million uniques a month, who have been around for years and who have decent engagement, a strong content muscle. But they don’t have the digital chops or the social skills, or aren’t reaching the millennials. I think for them, in terms of a DMP, a lot of it is about monetizing one-hit wonders—the people who come to them because they’re high up in Google searches. These people are not necessarily loyal, but they are still a decent chunk of uniques. A lot of the DMP infrastructure they utilize is to monetize their traffic by serving ads based on the third-party data that we bring to the table, because they don’t have enough first-party data on those users.
Now, the smaller guys, who do not have reach, but are pretty tech-savvy—they are extremely active on social, and they are really reaching millennials. I see them utilizing DMPs for increasing scale in a marketplace setting. We have a few clients who have created a co-op in order to increase their reach, because that’s what eventually makes them interesting to advertisers. They can combine their inventory and bring at least 50 million uniques to the table, millennials with really unique interests. And they go to market using a DMP’s capabilities to combine their inventory and create audiences across their properties.