Your Platform Is as Strong as Its Identity Layer
Analyzing platforms through a layered cake of identity, social networking, and super apps.
Elon Musk recently turned down his offer to purchase Twitter at a significant premium due to the large number of bots on the website. Twitter does not have a generalized identity verification process for all users. Instead, the platform’s verification resources are used to provision “blue checkmarks” to those with high visibility. Information tweeted from these accounts can be trusted to come from the named source.
At a fundamental level, this choice is a mistake. As long as Twitter does not make the effort to formally verify all of its users, they will struggle to lock in the value of their social network. While this may be a big ask, it is critical to building a trusted platform that can be ultimately monetized. In fact, the crux of any platform is identity.
Subscribe to receive new posts in your email 👇
We can model platforms like a layered cake, similar to the TCP/IP networking protocol. Each successive layer relies on the durability of the previous one to provide increasing sophistication.
Each layer asks a different series of questions. Some examples are listed below.
Identity. Who are you? How can I keep you accountable for your actions? How do I reach you?
Social Proof. Who do you interact with? How do you interact with other members of the network? How open are you to novel interactions?
Personalization. What are you using this platform for? Are your behaviors similar to other users? What are some exciting experiences you could be presented with that are unique to your profile?
Super App-ification. What other services could be useful to you? How can we help you profit from your own accrued value in the network?
Layer 1 is identity.
On financial platforms, identity is explicit. KYC (Know Your Customer) legal regulations require that banks and fintechs can accurately pinpoint their users to prevent money laundering and fraud. KYC includes information such as names, addresses, SSNs, etc. Financial institutions have to make the effort of verifying their customers upfront and periodically afterwards to ensure that any risk of inappropriate behavior is minimized.
Identity on social platforms is clearly different. But note that social identity is equally as valuable as financial identity. A network with verified users creates a feedback loop of trust and engagement, which leads to more value creation for everyone involved.
Verifiable information on social media is often less explicit, and therefore identity is a more challenging task to prove than with standard KYC. For starters, knowing a user’s real name is not as important as you would expect. Consider Discord, Reddit, and Twitter, all of which have sizable anonymous, non-bot account usage. This means internet-native identity features like email, text, and device sensor data, such as geolocation, are imperative.
A great example of a platform that built a strong identity layer which unlocked future revenue streams is WhatsApp.
Looking at WhatsApp today, you could imagine that this was the plan all along: bootstrap a universal identity system and communications network, and use them to create smartphone-based payment rails that skip legacy payment systems and enable more transactions.
From Communications, Commerce, and Money as the High-Order Bit
Poor identity verification can lead to high spam and bot infestations that drive away authentic users. As a rule of thumb, user identification should err on the side of being too strict. Not only does this allow for a more trustful community, but it also empowers the following layers in the stack, the next of which is social proof.
Layer 2 is social proof.
The primary goal of this layer is to provide proof-of-work for spam and inauthentic activity. It bolsters the first layer of identity through data and pattern recognition while also determining a user’s motives for engaging with a platform.
The financial analog here is transaction history. As you go about your day and rack up charges on your card, there are inherent patterns within the data that are implicitly baked into your identity.
Maybe you buy a double-shot cappuccino every day at roughly 10 am, or go to the same poke place next to your home a few days every week. The Spotify Premium membership with a student discount that shows up at the beginning of every month and the fact that you shop at Urban Outfitters tell me a lot about who you are. Without revealing any information about yourself, like your name, phone number, or email, I can already infer several things about you (with probabilistic bounds).
Any purchase that differs from the norm is not necessarily fraud per se (maybe you are traveling), but the likelihood that it is might be a bit higher than usual. Hence, the bank might send you a phone notification to alarm you about your unexpected purchase.
Just like how financial platforms track spending patterns, social platforms track communication patterns. Are you a lurker or actively posting and interacting with others? Do you post publicly or just share content with close friends? Which communities are you part off, and what are the features of these communities? Are they insular or are they open to others? Anomalies in communication patterns should be quickly processed and verified. In the example of bots, weird capitalization or grammar should be investigated immediately.
With the addition of layer 2, platforms can incorporate implicit social patterns and context with a user’s identity to get a better sense of who they are. The patterns users display may not be something they are consciously aware of. Regardless, they are unique and serve a purpose in building a safe, connected platform.
Layer 3 is personalization.
Layers 1 and 2 work together to accurately fingerprint a user. How can that unique sense of identity be leveraged to create a better product? The answer is personalization.
The traditional software approach to personalization has been with ads. While lucrative, ads have to potential to destroy a platform. Take Google Search for example. Google has an amazing identity layer. What does Google not know about me (check this out)? It could have used this data to build more personalized web experiences. Instead, we find ourselves in a world where we have to append " reddit” at the end every search to avoid SEO optimized garbage.
The canonical of personalization example is TikTok. Every TikTok user has the exact same app. At the software level, all of the binary and installation data is the same. However, the TikTok experience varies dramatically from user to user. It is almost as if the app evolves to your identity with only a few initial cues on your preferences, almost to an unsettling degree. As a byproduct, TikTok’s search has also become incredibly personalized; searching on TikTok is the new adding “ reddit” at the tail end of your query.
Imagine if Google Search behaved like this, preemptively finding content, whether it was a website or video, and showing it to you on the Google homepage. It’s interesting that many new companies have taken notice of Google’s poor personalization layer and are looking to change the way we think about search.
Amazon is another interesting example. While their personalization layer involves directly recommending products to users based on prior behavior, users are swamped by poor quality goods and reviews written by bots. Dropshippers with illegible names often sell the exact same AliExpress item with a different brand sticker slapped onto it.
The root of their problems is in their identity layer. As a retailer, they have entirely KYC’d their user base. However, they have failed to build a strong identity layer on the vendor side, leading to a suboptimal customer experience. This deeper issue makes personalization much more difficult.
Financial platforms traditionally haven’t had a layer 3 until recently. Many banks partner with Cardlytics to offer personalized spending experiences. These aren’t ads (deals often have a page of their own), but rather use customer identity and their social footprint to deduce appropriate discounts.
The goal of the personalization layer is not to make money. It is to create fulfilling, recurring, and distinctive experiences for the user. Just as startups seek to find product-market fit, platforms with a strong identity and social proof layer can aim to achieve product-user fit and attempt to monetize the value of their platforms directly through super app-ification.
Layer 4 is super app-ification.
YCombinator’s “make what people want” is a common mantra in Silicon Valley. It’s one thing to build something that people use, but in order to financially succeed, it’s equally important that platforms sell what their users want as well. Platforms with a strong layer 1, 2, and 3 can focus on building premium product lines within their platform that sell directly to their user network. This turns their app into a “super app.”
The term “super app” was coined by Connie Chan of a16z when referring to Chinese apps like WeChat, AliPay, and Meituan.
In China, they also have WeChat and Alipay. In Southeast Asia, there’s Gojek, which is used not only for transportation and food delivery, but also to book massage appointments, find moving services, and more. You can even use it to top up your mobile payment plan. And in the US, Uber is transforming itself from a transportation company into the operating system for our everyday lives. You see, super apps are here to stay, and the big internet giants in the States already get it. This trend is coming from East to West.
Here’s one example of how you can discover your hidden data asset. Spotify, a music streaming service, knows a lot about my musical interests. They know which artists and podcasts I like. What else can this data be used for? Spotify could use that data to help artists and concert promoters figure out which cities they should be touring in. They should use data to inform their strategy. (“Should I spend one more night in New York or book that day at STAPLES Center?”) In fact, Spotify should be able to work with artists and tell them how to personalize their playlist for every single city. These are new revenue streams that Spotify can unlock by discovering their hidden data assets.
The example of Spotify above walks through all of the four layers. Your identity is shaped by where you live and who you are, as displayed in the “Profile” page of Spotify. Some artists are more popular in some areas than others, which is one vector for personalization. Your social proof is the artists and podcasts you listen to and how you interact with them (e.g. do you share songs with your friends through the app?). Spotify uses that data to personalize future recommendations. The leap to super app occurs when Spotify engineers build out a tour management software for the musicians and show hosts on the platform. Arguably, Spotify Car Thing is one avenue for super app-ification that the company is currently looking into.
Instagram is no longer just the photo-sharing app Kevin Systrom dreamed of. It now has aspirations of becoming the digital equivalent of a shopping mall, which directly sells personalized items to users through Instagram Shop. Due to its strong identity verification, WhatsApp can circumvent traditional payment rails, as described in an earlier section above. Twitter users with specially verified accounts can earn money from their followers through the Super Follower program. Snapchat engages users with Snap Minis and social shopping opportunities.
The common thread between all these experiences is that they are transparent, unlike intrusive ads. Users consent to these “meta apps” on the platform and are therefore more willing to spend money.
We can expect this trend to become more and more common in the coming years. Platforms like YouTube are already investigating how they can build paid streaming services through a “channel store”.
Some closing thoughts.
Platforms need to recognize the value of identity and take the responsibility to build trust within their networks. Only then can they unlock the flywheel outlined in this article. A strong sense of trusted identity leads to more engagement. Social platforms can use this engagement to provide uniquely personal experiences for each of their users. This ultimately unlocks more opportunity for growth through super app-ification.
Any writing on identity in 2022 is incomplete without a digression on crypto, but I’ll keep it short as it could easily be its own essay.
Crypto fundamentally lacks an identity layer. There are plenty of apps (or dapps) available to users, making protocols like Ethereum a “super app” by definition. However, there are zero repercussions for poor behavior patterns. Without identity, there is no way to contact others in the crypto world. There’s no analog to a text message or email.
Moreover, it is incredibly challenging to build trust within a purely decentralized community, especially when identities can be splintered across different wallets. Crypto users often use multiple wallet addresses for transactions (and can easily make more as needed), making it impossible to identify individuals with social proof and spending patterns. Approaches to bringing identity on-chain (e.g. through soulbound tokens) exist, but large scale adoption is yet to occur and they do not incentivize the use of a single address across all transactions.
In the case of Elon Musk and Twitter, Ben Thompson of Stratechery argues that Twitter could be used to provide identity as a service, or in other words, serve as a layer 1 for the entire internet.
[If] there is one lesson from mobile, it is just how important notifications are; a secondary consideration is how important identity is. If you can know how to reach someone, and have the means to do so, you are set, whether you be a critical service, an advertiser, or anything in-between. Twitter has the potential to fill that role: the ability to route short messages to a knowable endpoint accessible via a centralized directory has far more utility than political signaling and infighting. And yet, thanks to Twitter’s early decisions and lack of leadership, the latter is all the service is good for; no wonder user growth and financial results have stagnated!
A truly open [Twitter] has the potential to be a new protocol for the Internet — the notifications and identity protocol
At the core of this layered model is the importance of identity. User identity and social proof together form a unique sense of digital identity, which unlocks authenticity and engagement within online spaces. Identity is the electricity that powers platforms.