Latest / Inside DuckDuckGo / Duck Tales: Marketing at DuckDuckGo — how we practice the privacy we preach (Episode 9)
Transcript
- 0:00SPEAKER_00: Hi,
- 0:00SPEAKER_00: and welcome to DuckTales,
- 0:01SPEAKER_00: where we go behind the scenes at DuckDuckGo and discuss the stories,
- 0:05SPEAKER_00: technology,
- 0:06SPEAKER_00: and people that help build privacy tools for everyone.
- 0:09SPEAKER_00: In each episode,
- 0:10SPEAKER_00: you'll hear from employees about our vision,
- 0:12SPEAKER_00: product updates,
- 0:13SPEAKER_00: engineering approach to AI,
- 0:15SPEAKER_00: or how we operate as a company.
- 0:17Today we're going to chat about how most companies collect a ton of
- 0:20information
- 0:21SPEAKER_00: through their marketing activities and how DuckDuckGo,
- 0:23SPEAKER_00: given our privacy policy of we don't track you,
- 0:26SPEAKER_00: do things like attribution very differently.
- 0:29SPEAKER_00: I'm Cristina.
- 0:30SPEAKER_00: I'm on the marketing team.
- 0:31SPEAKER_00: And today I'll be interviewing Chuck.
- 0:33SPEAKER_00: Chuck, you wear a lot of different hats.
- 0:34SPEAKER_00: Can you introduce yourself and some of what you work on?
- 0:37SPEAKER_01: Sure.
- 0:38SPEAKER_01: I...
- 0:40I'mtechnicallyonthefrontendteamandworkonthefrontendofoursearch
- 0:45SPEAKER_01: products and our subscription products.
- 0:47SPEAKER_01: But I ultimately do whatever I need to do to get the job done,
- 0:52SPEAKER_01: which is kind of our DuckDuckGo ethos.
- 0:54SPEAKER_01: I do some product management, some data science, back in engineering.
- 0:59SPEAKER_01: I work with the marketers.
- 1:01SPEAKER_01: It's fun.
- 1:02SPEAKER_01: And I need hats.
- 1:03SPEAKER_01: So getting to wear lots of them is.
- 1:06SPEAKER_00: Fair enough.
- 1:07SPEAKER_00: Well, thank you.
- 1:08So much like our product philosophy,
- 1:11privacy is core to the ethos of our marketing.
- 1:14SPEAKER_00: There are so many common practices we don't do.
- 1:16SPEAKER_00: Identifying and targeting individual users,
- 1:19SPEAKER_00: retargeting,
- 1:20SPEAKER_00: using behavioral data,
- 1:22SPEAKER_00: using third-party cookies and pixels,
- 1:24SPEAKER_00: the list goes on.
- 1:25And we've also declined working with a lot of vendors because they don't meet
- 1:29our
- 1:29SPEAKER_00: privacy standards.
- 1:30SPEAKER_00: As a consumer, that's something I really appreciate.
- 1:33SPEAKER_00: But frankly, as a marketer, it makes the job very hard.
- 1:38SPEAKER_00: But it's getting a bit easier,
- 1:39SPEAKER_00: thanks to work from people like Chuck,
- 1:41SPEAKER_00: which is why I was so excited to talk to you today.
- 1:43SPEAKER_00: So Chuck,
- 1:44SPEAKER_00: when you first started working with the marketing team,
- 1:47what was your reaction to our limitations and what we were hoping to achieve?
- 1:52SPEAKER_01: Honestly, I was a little shocked.
- 1:54There's a pretty well understood playbook for how marketing in a space like
- 1:58this should look.
- 1:59SPEAKER_01: A playbook of tactics and tools that are well understood.
- 2:02And every company will do it differently and every brand and product will
- 2:05have
- 2:05SPEAKER_01: their own personality.
- 2:06SPEAKER_01: But we pretty much use none of those tools.
- 2:10SPEAKER_00: Yeah.
- 2:10SPEAKER_00: Can you help people understand what the industry norms are for marketing
- 2:14SPEAKER_00: attribution and data and how we do it differently?
- 2:18SPEAKER_01: Yeah,
- 2:18so when you visit your favorite social media site and it's trying to decide
- 2:21howto
- 2:22SPEAKER_01: fill the ad slot in your feed,
- 2:24SPEAKER_01: the ad platform will take what it knows about you as a person,
- 2:26SPEAKER_01: your search history,
- 2:27SPEAKER_01: who you follow,
- 2:28SPEAKER_01: and what it knows about your situation,
- 2:29SPEAKER_01: like where you are and who you're with,
- 2:31SPEAKER_01: and line it up with their ad inventory.
- 2:34They'll do some very complex math to determine the perfect ad to show you that
- 2:37will
- 2:37SPEAKER_01: maximize profits for the platform and the advertiser.
- 2:41SPEAKER_01: So the more better data they have about you,
- 2:43SPEAKER_01: the better they can target the ads and the more money they can make.
- 2:46I know that's something you've talked about with Peter on a previous
- 2:48episode,
- 2:49that the financial incentive for the trackers that are ubiquitous online is
- 2:53data
- 2:53that feeds the machine that helps them make more money off of your ad space.
- 2:56SPEAKER_01: Right.
- 2:57SPEAKER_01: That entire ecosystem just flies in the face of our privacy principles.
- 3:01In fact,
- 3:01some of our apps will block those trackers to keep your browsing private.
- 3:05SPEAKER_01: So when we advertise,
- 3:07werefusetousethosetoolslikeyoujustlistedthatarecommonindigital
- 3:10marketing like retargeting or reporting different types of conversions
- 3:14afterthead
- 3:15SPEAKER_01: click just to protect the privacy of our users.
- 3:19SPEAKER_01: Instead,
- 3:19we'll collect limited data only when there's like a very clear and urgent
- 3:23rationale
- 3:24SPEAKER_01: for it.
- 3:24And when we do,
- 3:25we're transparent about what we collect and how we use it.
- 3:28SPEAKER_01: And we're possibly most important,
- 3:30we're really careful never to let those logs link two different events to the
- 3:33same
- 3:34SPEAKER_01: person.
- 3:35SPEAKER_01: That's really difficult to do.
- 3:38We have a really fantastic privacy engineering team that reviews every
- 3:41projectin
- 3:42their implementation to make sure that the work we're doing is aligning with
- 3:45our
- 3:45SPEAKER_01: principles.
- 3:46We've also gotten really comfortable making decisions with just imperfect
- 3:49or
- 3:50SPEAKER_01: incomplete data,
- 3:51trying to identify the solutions that meet 80% of the business needs with 20%
- 3:55of
- 3:55SPEAKER_01: like the potential input.
- 3:59Yeah,
- 4:00it feels like a lot less than 20% of what's actually available to us.
- 4:05SPEAKER_00: That's probably fair.
- 4:07SPEAKER_00: Yeah.
- 4:07SPEAKER_00: So thank you for unpacking that.
- 4:09SPEAKER_00: That's a helpful foundation.
- 4:10Can you go a bit deeper and talk about what that looks like in practice at
- 4:14DuckDuckGo?
- 4:16SPEAKER_01: Yeah,
- 4:17so we largely don't work with other vendors in the marketing space and rely on
- 4:23the
- 4:23SPEAKER_01: tools we own and build ourselves instead.
- 4:25That makes sure that we aren't incidentally feeding the machine with our own
- 4:29user's
- 4:29SPEAKER_01: data,
- 4:29SPEAKER_01: which is really easy to do if you're not careful.
- 4:33SPEAKER_01: We have a couple of like tools in our toolbox too.
- 4:36We'll do as much summarization and analysis of data locally before we ever
- 4:40send it back.
- 4:41Soratherthansayingthatauserofourbrowsersearches15timesinadayand
- 4:47SPEAKER_01: sending 15 different events for those searches,
- 4:50we'll send a periodic report that will say they searched 15 times during that
- 4:54day.
- 4:55SPEAKER_01: We'll reduce the precision of those signals even further.
- 4:58SPEAKER_01: So instead of saying that that person made 15 searches,
- 5:01SPEAKER_01: we'll say they're a medium volume search user.
- 5:04SPEAKER_01: And then when we do our analysis on an ad campaign,
- 5:07we'lllookatthesummariesofthedataratherthantherawdatasothatwe're
- 5:10SPEAKER_01: looking across our users rather than the individual humans.
- 5:14SPEAKER_01: And if it comes down to it,
- 5:15SPEAKER_01: we are willing to redact data that might be too identifying for a person,
- 5:18SPEAKER_01: whether it might contain PII or
- 5:21SPEAKER_01: If it looks too unique and may be able to be traceable back to a person,
- 5:24we'd rather delete it and not use it than jeopardize that person's privacy.
- 5:32SPEAKER_00: Well, thank you for sharing how our ethos really comes to life there.
- 5:36And I'd love for you to touch on one of your claims to fame at DuckDuckGo,
- 5:41SPEAKER_00: which is creating a better,
- 5:43SPEAKER_00: more privacy-respecting system that we call Origin.
- 5:46SPEAKER_00: Can you talk about how you got the idea and how you brought it to life?
- 5:50SPEAKER_01: Yeah,
- 5:50so we were struggling to run small scale campaigns that test new ad platforms
- 5:55or
- 5:55SPEAKER_01: creatives.
- 5:56SPEAKER_01: With the tools that we have,
- 5:57the only way that we could do that without jeopardizing user privacy is to run
- 6:01big,
- 6:02SPEAKER_01: broad,
- 6:02SPEAKER_01: expensive,
- 6:03SPEAKER_01: scaled campaigns.
- 6:04SPEAKER_01: But we're a small company.
- 6:06SPEAKER_01: We want to move nimbly,
- 6:08and that made it really difficult for us to quickly validate our direction
- 6:11andmake
- 6:11sure that we were dedicating our resources in the right time or in the right
- 6:14place.
- 6:16SPEAKER_01: So I spent some time with our marketing leaders,
- 6:17SPEAKER_01: including you,
- 6:18SPEAKER_01: Christina,
- 6:19SPEAKER_01: trying to understand the norms and the challenges they were facing,
- 6:23SPEAKER_01: the tools that weren't in their toolbox.
- 6:25SPEAKER_01: And I brought that to the privacy team.
- 6:27SPEAKER_01: We worked backwards,
- 6:29starting with user privacy as a first principle to the business goals and
- 6:33landedon
- 6:33SPEAKER_01: a solution that kind of looks like this.
- 6:37You see an ad and you click on it for DuckDuckGo,
- 6:39and you install our app from it.
- 6:41SPEAKER_01: When that app first runs,
- 6:43we will send one signal that says that you installed the app from that ad in
- 6:46that
- 6:46SPEAKER_01: location.
- 6:48SPEAKER_01: And then once a day,
- 6:48we'll build a summary of those signals that give us pretty coarse insights
- 6:52that
- 6:53SPEAKER_01: say,
- 6:53SPEAKER_01: you know,
- 6:53we had 10 users install our app from that ad on that ad platform on that day.
- 6:59Then we'll line that data up with other information that the ad platform
- 7:02gives us,
- 7:02like how many impressions there were of the ad and how many times it was
- 7:05clicked
- 7:06SPEAKER_01: and how much that cost us.
- 7:07And that'll give us some high level insights we can use to start making
- 7:10decisions,
- 7:10SPEAKER_01: like how much it costs us to per install from that ad.
- 7:15SPEAKER_01: There's nothing groundbreaking here technologically.
- 7:18SPEAKER_01: It's actually intentionally very simple.
- 7:20And that helps us maintain the privacy properties because we have a high
- 7:23elevation
- 7:23SPEAKER_01: view of everything that's happening.
- 7:26We never share data outside of DuckDuckGo,
- 7:28so we aren't feeding that machine.
- 7:30SPEAKER_01: There are never person-level insights.
- 7:31SPEAKER_01: We're looking at broad signals across our audiences.
- 7:35SPEAKER_01: There's no risk of PII,
- 7:36SPEAKER_01: and we're only collecting the data that we need to make those decisions,
- 7:39SPEAKER_01: nothing more.
- 7:40But it still lets our marketing team make informed decisions while working
- 7:43quickly
- 7:44SPEAKER_01: and doing their job well.
- 7:47SPEAKER_00: Well, thank you.
- 7:49More importantly,
- 7:49thank you for the months and months of work you did on that.
- 7:53SPEAKER_00: You know,
- 7:53SPEAKER_00: you say it's nothing revolutionary,
- 7:55SPEAKER_00: but actually I think it's a pretty novel approach,
- 7:57SPEAKER_00: right?
- 7:57SPEAKER_00: Like we don't know of any other companies using technology like this.
- 8:01SPEAKER_00: Typically, they use the entire suite of tools available to them.
- 8:05SPEAKER_00: But,
- 8:06SPEAKER_00: you know,
- 8:06hopefully one day it won't feel like such a novel approach and this will
- 8:09become
- 8:10SPEAKER_00: more of the industry standard.
- 8:11SPEAKER_00: At least, you know, my naive perspective can hope for that.
- 8:14SPEAKER_01: I hope so.
- 8:15SPEAKER_01: We'll see if capitalism agrees with us.
- 8:18So any parting thoughts you'd like to leave on the future of privacy
- 8:22respecting marketing?
- 8:25One of the things that I really love about and appreciate about DuckDuckGo is
- 8:28the
- 8:28SPEAKER_01: example that we set for other companies.
- 8:31SPEAKER_01: On the search engine side,
- 8:32SPEAKER_01: could we collect data at massive scale and hyper-target ads to our users?
- 8:37SPEAKER_01: Absolutely, but we don't need to.
- 8:39And we love being an example of a sustainable business that respects user
- 8:42privacy
- 8:43SPEAKER_01: in their searches.
- 8:44SPEAKER_01: And I like to overlay that to our marketing efforts, too.
- 8:47Would we benefit by using really invasive tracking like the industry
- 8:50standards?
- 8:51SPEAKER_01: Yeah, probably, but we don't need to.
- 8:53We're a good example of how you don't need to participate in that
- 8:56data-intensive
- 8:57SPEAKER_01: ecosystem to market your business,
- 8:59SPEAKER_01: and I'm really proud to work on it.
- 9:01SPEAKER_00: Well, thank you, Chuck.
- 9:02SPEAKER_00: I feel the same way.
- 9:04SPEAKER_00: And for those of you listening,
- 9:05if you or someone you know considers themselves to be a privacy-centric
- 9:09marketer,
- 9:10SPEAKER_00: including ad networks and measurement partners,
- 9:12SPEAKER_00: would love to chat,
- 9:13SPEAKER_00: reach out to me on LinkedIn with what you're doing differently.
- 9:17SPEAKER_00: Chuck, it was really great chatting with you today.
- 9:19SPEAKER_01: Always, Chuck.
- 9:19SPEAKER_01: Great to chat with you, too.
- 9:21SPEAKER_00: Thanks to everyone who took the time to listen.
- 9:23We have so many more episodes planned on a wide variety of topics,
- 9:27so stay tuned.
- 9:28SPEAKER_00: Bye.