On-chain attribution & decentralized identities pave a new way for marketers to measure the success of their campaigns. Unlike cookie-based data, marketers are able to measure with precision who purchased which products and what path they took to get there.
However, this promised land of measuring marketing activity on-chain will take some time to come to fruition. If we use the web2 growth stack as a proxy, there are three key gaps in the growth stack today that we need to fill: solutions to ad fraud, global adoption of wallet-based web browsing, and web3-native media channels.
Poor match rates between publisher-reported impressions via server side platforms (SSPs) and programmatic-reported impressions via demand side platforms (DSPs). The ISBA documented that only 4% of the ads (61 million / 1.3 billion) were matched from participating advertisers to participating publishers. If we even put 10% of impressions on-chain, we will truly begin to measure marketing metrics without sacrificing user privacy in the process.
Why do we need better match rates between SSPs and DSPs?
Let’s use an example: imagine being told you’re buying 100 bananas when your vendor only gave you 61 bananas. As a buyer, you’d be able to verify whether or not you have the 100 bananas you bought by counting the quantity on delivery.
With the delivery of digital impressions, there is no feedback loop to reduce publisher supply of impressions unless the buy-side forms an alliance to unlock media cost savings by fighting ad fraud with blockchain technology. Matching impressions served by publishers and impressions purchased by DSPs create a tamper-proof recording of marketing performance, which is something that does not exist today in the adtech industry. Current advertising systems lack the ability to audit the campaign data*.* The ad network exclusively determines how much advertisers will be charged, as well as the revenue share that the publishers may get.
The ISBA 2022 study did highlight concrete improvements like
The study noted that these improvements came from better data access, cleaner data that could be matched from end-to-end through the supply chain. Brave’s Themis ad platform is a great example of how smart contracts can be utilized to have both a campaign manager and advertiser agree on ad campaigns, and I’m sure we’ll see more solutions in this category this year.
So what do we need to start leveraging as an ecosystem to drive the industry to this standard?
Publishers that leverage blockchain technology (example: Solana’s ‘proof of transaction history’) will provide an on-chain record of an impression — from the moment the ad was served by the SSP to the sale & attribution of that impression by the DSP. Over time, we would be able to publicly identify websites that have not registered verified human traffic and as an advertising industry, push back on ad fraud and save budget on paid media campaigns.
Why are on-chain impressions are so important? fake news manufacturers typically setup a ton of vanity URLs (eg: BestMomLife.ca), purchase cheap bot traffic from black hat vendors to make the publisher seem like it is generating traffic to the site. Then, list the phony ad site impressions to the DSPs for real-time bidding. There is currently no solution that exists for this.
However, if we pair Solana’s proof of transaction history working with publishers recording on-chain web visitor metrics, we can stitch together the supply of impressions with the demand for impressions. Website visitors read a Mirror.xyz blog posts, an on-chain website visit is recorded as a verified impression, from a verified wallet-based browser, to record the exchange of value between author and reader.
The majority of web2 advertising relies on pixeling a website visitor, app user or customer. While we are increasingly token gating decentralized app experiences, the reality is that the majority of our web browsing still is on web2 browsers like Google Chrome which track and monetize website behavior.
Web3 wallets like Metamask and Phantom have web browsers in their native apps (eg: Metamask’s mobile web browser), but the user experience to get to these browsers are not ideal and usage is likely low.
Because wallet-first browsing isn’t a widespread behavior, web3 marketers have two options. They could leverage the existing ad tech to reach a wider audience with cookie data… or they could ignore it altogether and only by media on-chain today. Which brings me to my next point…
One in three people on Earth use Facebook on a monthly basis. By comparison, only 30 million people use Metamask, the most popular wallet dapp on the Ethereum blockchain. If wallet-based browsing is too difficult to imagine a normie to adopt, then we can think through web3-native media channels and outlets as a way to generate verified human impressions through a token-gated content experience.
Web3 social platforms (eg: Lens.xyz), blogging platforms (eg: Mirror.xyz) and messaging platforms (eg: Notifi, Dialect) all have verified wallets. There is an opportunity here to assemble a data co-operative that shares on-chain user bases as a data set to feed a wholly-community owned Demand Management Platform (DMP). A DMP that has verified wallets browsing the web will be able to match against Advertiser IDs (ADIDs) in the programmatic landscape — kind of like how a custom audience in Meta is able to identify your customers on Facebook and Instagram.
The more web3 users browse the web using Brave, share updates on Lens.xyz, blog on Mirror.xyz and communicate with friends on-chain, the more we will be able to defend against fake impressions from bot traffic that obfuscates the growth marketing funnel.
GDPR, ITP 2.0 and recent regulatory policies around the world are wising up to protect a user’s data privacy . Users are increasingly able to “opt-in” to web tracking, rather than having to “opt-out” of it. However, for these privacy laws to become effective, lawmakers need to ban data co-ops that enable companies to enrich customer profiles using data that does not comply with current data privacy laws.
Now, pair that with self-sovereign digital identities (DIDs). Digital identities are a new way for consumers to protect their data privacy by creating a psudonymous profile. This may appear to be a useless feature of web3, but a psudonymous profile enables a person to browse the web and share different parts of their identity, without having to share their entire identity.
Here are a few opportunities how web3 users could be compensated for sharing their behavioral, psychographic, location and IP address with an advertiser:
Until there is enough content & user adoption for web3 native media channels, brands will continue to rely on web2 social media platforms, content publishers and affiliate networks. This introduces a new risk into the growth stack — ad verification.
In the traditional advertising stack, a brand that is investing heavily into paid marketing will work with a vendor like DoubleVerify to spin up an independent 3rd party ad server. This is important because the publisher, ad network and media agency all have an incentive to inflate campaign performance metrics. The publisher and ad network want to serve as many impressions as possible to maximize advertising revenue. The media agency wants to spend as much of their client’s media spend as possible, since media agencies typically charge a percentage of media spend. A brand that is able to serve ads through it’s own independent server will be able to compare the publisher’s, ad network’s and media agency’s campaign reporting with their own advertising data to audit the report.
Currently, most advertisers are blindly trusting that the other players in the industry are telling the truth — which we all know is not the case. This is why at Myosin, we are excited to build a data co-operative, where brands can safely share campaign data to identify fraudulent publishers, share knowledge on what’s working and drive efficacy for web3 marketing campaigns through a community-owned 3rd party ad server.
Most marketers don’t know that Google Analytics is based on survey data. Google Analytics does not track every single site visit deterministically — it uses a set of sampled data to report web analytics. Google Analytics uses last-touch attribution to record conversions. And yet, Google Analytics is the most popular web analytics tool that digital marketers use today. This might work for e-commerce products with short consideration cycles (eg: a yoga mat you buy on Amazon), but it doesn’t work for most on-chain assets (think cryptocurrencies and NFTs). On-chain assets, by comparision, most likely are longer-consideration products and services, which means that marketers need to use multi-touch attribution models.
Recording on-chain web analytics through a tool like Crux significantly improves the trustlessness of a data source and tamper-proof data cleanliness that a multi-chain attribution model (MMM) can use to calculate attribution. Attribution models are only as good as the quality of data fed to the model, and as an industry, if we share data that feeds multi-touch attribution models, we identify how each channel performs by web3 product category (eg: DeFi, NFT platforms, Layer-1s, DAO tools, Loyalty programs, etc).
Right now, because there is no safe way to share performance data through a community-owned data co-operative in a safe way, each dapp and platform is relearning the same marketing optimizations over and over again rather than sharing knowledge to minimize media waste for the industry.
It’s an exciting time as we build out the growth tools for a new web3 world, and we already have talented builders create solutions like on-chain publishers, affiliate networks, on-chain web analytics, attribution. We need more on-chain data in the top and middle parts of the marketing funnel to truly track marketing performance, eliminate ad fraud, and protect user privacy in a user-first built ad stack.
Of course, the web3 growth stack will have new and novel channels that do not currently exist in the web2 growth stack (eg: quests & bounties). I’ll cover web3 native growth products in a future blog post.