6
Know Your Product and Your Audience

There is much about Web3 and related marketing that is novel and revolutionary. “Know your product” and “know your audience” are not: these are practically the first two commandments of marketing. As with most other commandments, however, the hard part of a timeless rule is actually applying it in a particular context. This chapter examines what it means to know—which is to say, to learn about—your product and audience in the evolving context of Web3.

Know Your Product

Sometimes we as marketers have the luxury of choosing a product to market. The most correlated explanatory factor for our team's success bringing Ethereum to market is the genius of Ethereum. We simply couldn't have achieved comparable results if Ethereum were a less mind‐blowing technology. No marketing strategy can replace having an excellent product. However, even the best products can die in darkness if no one pays attention to marketing them. Especially early in a project's life cycle, someone needs to stimulate demand and connect it with supply, otherwise the flywheel of adoption never starts spinning.

If a product is excellent—defined as having a target audience that would use it if they knew about it—marketing hastens the process of creating demand and connecting it with supply. It gives the flywheel a good push or a series of them. But marketing in Web3 is far from a silver bullet. With a low‐quality product—one that users wouldn't want even if aware of it—there's no marketing strategy that can make the project succeed in the long run. The product is like a bone, the solid core of a project. But a body that's all bone and no muscle just lies on the floor. All muscle and no bone doesn't work either. To walk around the block, the body needs muscles to attach to bones and apply leverage on them. A project's success comes out of the tension between the solid bones of a high‐quality product and the marketing muscle that activates it into motion.

At ConsenSys, our Web3 marketing team was deployed to market any project the company built; at Serotonin, we are extremely selective about the marketing projects we take on. Our goal at Serotonin is to help grow the Web3 ecosystem by driving return on marketing investment for our clients. If we believe a Web3 project lacks a solid product vision, we know that no matter how effectively we deploy our capabilities—our marketing muscle—ultimately the client won't succeed or drive Web3 growth, even if we're able to generate temporary buzz. Thanks to tireless efforts to maintain and grow our reputation as not only the first, but the best Web3 marketing team, we have a robust pipeline of potential clients that enables us to be selective. Unlike traditional agencies, we're not guns for hire, and we won't work with projects we don't believe in. We have the luxury of choosing our partners carefully, and we prioritize them based on what we evaluate as their chance of success based on their product and team. A project with an excellent product is primed for marketing to catalyze step‐function growth, and we want to deploy our efforts where they will lead to the most value creation. All Web3 marketers would be well advised to choose the products carefully that they market. With a high‐quality product, they are more likely to succeed and establish a positive reputation.

At Serotonin we learned these lessons the hard way. Early on, we signed with a Web3 utility company whose product was almost ready to launch. This was a Web3 version of a popular type of product that exists in Web2. Users of this type of product in Web2 overwhelmingly prefer paying by subscription, but the payment model for this Web3 product was pay‐as‐you‐go. We conducted market research on our client's behalf that suggested Web3 users, like Web2 users, would prefer to pay by subscription to avoid the friction of reupping their accounts every time they ran dry. But the company's product team dug in their heels; they had already built most of the product and preferred to launch it without making major changes. Introducing a subscription would have required overhauling much of the product. It was one thing to want to launch quickly, but even months later, when the product was live but failing to gain traction, the product team continued to dig in their heels despite user feedback about the friction of pay‐as‐you‐go. They contended it was the job of the marketing team to go find users who wanted pay‐as‐you‐go, as opposed to the product team's job to iterate on the product based on early user feedback. Over time, we would see this pattern over and over and flag it as a formula for failure.

The marketing and product teams both need to be responsive and flexible. Marketing teams need to seek out users for the product that exists today. They also must gather data from users about their preferences and feed it back to the product team. Then the product team must iterate based on that feedback.1 Without a proper feedback loop between product and marketing, otherwise competent product and marketing teams are doomed. We still constantly ask ourselves why products succeed and fail, and we identify and test new patterns.

At ConsenSys, I learned that to market Web3, I needed to deeply understand it. That's why the first part of this book on Web3 marketing is dedicated to presenting the key components of Web3. Anyone marketing a Web3 project should make sure to understand the product. Most marketers aren't computer scientists or developers and can't read the actual code of a Web3 project. This shouldn't stop them from trying to understand how it works as deeply as possible, asking founders or engineers to explain in layperson's terms as many times as it takes. Marketers working on a particularly technical product should consider taking an online coding class like CodeAcademy. I did this during my early days at ConsenSys, not because I wanted to build code, but to understand the vocabulary my colleagues were using and translate it to others without software development skills. The worst thing a marketer can do is pretend to understand the technology behind a product; instead, especially at the beginning, they should find a patient colleague and, in the style of our most naturally efficient learners—namely, toddlers—obsessively ask why or what does that mean until they clarify the fundamental underlying concepts. Here are some good questions marketers can ask of Web3 projects to become familiar with the technology:

  • What type of project is it? (Web3 utility, infrastructure, DeFi protocol, NFT, DAO, etc.)
  • What is the core value proposition to users?
  • How does its technology architecture support this value proposition?
  • How does its technology compare to competitors solving the same problem?
  • What, if any, UX challenges is the project trying to solve for?
  • What rewards are built into the technology for incentivizing behavior?
  • Which Layer‐1 blockchain does it use? (Bitcoin, Ethereum, Avalanche, Solana, etc.)
  • Does it use any kind of Layer‐2 scalability solution? (Polygon, StarkNet, Optimism, etc.)
  • How does it address decentralization? Is there a decentralization road map?
  • How does it address scalability? What kind of user volume could it accommodate?
  • How does it address security? What are the vulnerabilities? Has it been audited?

Though we as marketers need to understand everything possible about the product, this isn't necessarily true of the end users, unless those end users are developers. For B2D (business‐to‐developer) products, developers usually need to understand exactly how a product works, usually through tutorials, thorough documentation on open source code, and inspecting the code base themselves on a platform such as GitHub. For B2B (business‐to‐business) and B2C (business‐to‐consumer) products, the end user often doesn't need to understand how a product works; they simply need to know what the product does. These are not the same thing. Few people who use the internet understand how TCP/IP works, but that doesn't stop them from checking their email. If being able to re‐create a light bulb were a prerequisite to flipping on a light switch, most of us would be sitting in the dark. Knowing how a technology works is essential for marketers, but not necessarily users. It's our job as marketers to be a digestive system for information: to break down chunky technical concepts, extract a clear value proposition, and convey the proposition to its target audience in a way that is accessible and compelling. Sometimes, depending on the audience, that means avoiding blockchain language entirely, even if a product uses the technology—for example, replacing a term such as NFT with digital collectible. Just because we care about understanding the technology doesn't mean our audience will.

Know Your Audience

Understanding the product is only one‐half of the equation to market effectively. The key to success at marketing, and many other parts of life, is to know one's audience. This is hardly unique to Web3. Dale Carnegie's How to Win Friends and Influence People was published in 1936. A lecturer who studied the fields of personality and salesmanship, Carnegie was the first writer in the canon of modern business literature to identify a principle that now seems obvious: to be convincing to another person, one must first put oneself in their shoes. Imagining what it's like to be another person enables us to understand their desires and motivations, and then frame our argument or sales pitch in terms that matter to them. When we try to convince another person, the key isn't what we want; it's the other person and what they want.

A large proportion of the business and marketing books I've read are derivatives of Carnegie's classic, so when a new marketer joins Serotonin, I recommend they go straight to the source and read the original. After nearly a decade of Web3 marketing, we find its essential lesson couldn't be more relevant. As marketers, if we approach our task only from the perspective of the project, we might be able to distill its value proposition into a clear message, but that message is unlikely to be compelling if it's not shaped to fit the target audience receiving it. A good analogy comes from chemistry: an atom must have the right configuration of electrons in order to bond with another atom. In marketing we are trying to forge a bond between a product and an audience. The correct way to do this is study the target audience's “electron structure,” custom configure a message with just the right number of electrons, and deliver it over the right channel. Then the message and the audience connect, catalyzing a measurable action, such as downloading a product.

When I first joined ConsenSys, some of my colleagues were worried about introducing marketing to the Ethereum ecosystem. They were familiar with Facebook advertising campaigns, and other intrusive types of marketing that users wish to avoid. Understandably, they didn't want to pollute the thoughtful and academic spirit of Ethereum with unwelcome messaging. But convincing developers to experiment with Ethereum through a Facebook ad campaign would have been an extremely poor marketing practice.

On our Web3 marketing team, we started by learning about our target audience, and studied everything we could about the computer scientists, developers, and entrepreneurs being attracted to early Ethereum. We noticed Ethereum meetup groups for enthusiasts were forming in cities around the world. Their members were college‐age computer science students, recent graduates working at startups, and hackers working on their own projects from basements or garages. Some were formally trained, and others had taught themselves mathematics, coding, or economics. Some lived in the Web2 technology capital, the San Francisco Bay Area. Others lived in finance hubs such as New York, London, Dubai, and Singapore. Even more fascinating, meetups were springing up in Budapest, Bucharest, São Paulo, Istanbul, Johannesburg, Lagos, and Jakarta.

Our community marketing team reached out to existing meetup groups and organized a network of meetup group leaders to share best practices. We kept an eye out for Ethereum communities cropping up in new cities, and we launched a program to support new meetup group founders. Through small sponsorships, we ensured meetup groups always had the funds to provide beer and pizza. We circulated decks and templates with standard language about Ethereum and consistent visual design that conferred on nascent groups a sense of legitimacy. As new projects were built on Ethereum, we arranged for their creators to go on IRL roadshows to present their ideas and get feedback from the meetup groups. Group leaders could count on regular, high‐quality content, and project teams had the chance to get exposure to early product adopters.

Ethereum meetups grew like weeds. By summer 2017, after the EEA launch and the associated price action and publicity for Ethereum, it was easy in San Francisco, London, Tokyo, Shanghai, or New York to attend an Ethereum meetup every night of the week. For its most devoted members, mostly young computer scientists and developers, meetups took on a semi‐religious air. There were jokes and memes about “the church of Vitalik.” But jokes aside, Ethereum meetups indeed provided to many the benefits they no longer found in traditional religion: a meeting place for people with shared values to work together toward a common purpose. We had learned who our audience was (computer scientists/engineers), the right channels to reach them (meetup groups), and the right message, which was offering them tools and resources to build their own communities, rather than imposing our idea of what they should look like. The culture of IRL meetups and hackathons as fertile ground for innovation persists to this day.

The lesson of learning from one's audience applies to Web3 today just as it did in 2017. At Serotonin, when we think about growing the community, we start by studying the actions its members are already organically taking and seeing how we can support them. Let's take PROOF Collective's Moonbirds NFT series as an example. When a group for women Moonbirds holders called Ladybirds sprung up, the PROOF team helped its leaders arrange their own online and offline events and supported growing their Discord server by amplifying Ladybirds on social media and letting other holders know Ladybirds existed. Another NFT project, Crypto Coven, noticed that its users were tweeting side‐by‐side pictures of their IRL selves next to their Crypto Coven NFT witches with the captions “Web2 me” and “Web3 me.” Many of the witch NFT holders looked like their NFTs in real life. Others, hilariously, couldn't look more different. The project's founders and team started tweeting their “Web2 me” and “Web3 me.” Crypto Coven server members on Discord encouraged each other to share. The meme went viral on Crypto Twitter, strengthening the Crypto Coven community by giving its members their own inside joke and success story. Crypto Coven's founders, similar to the PROOF Collective team and the ConsenSys marketing team, started by observing their audience. Each succeeded in growing communities by learning about the patterns and preferences of their nascent communities and strengthening them, rather than imposing top‐down strategies. Web3 marketers don't always assume they're the ones with the best ideas. Rather, they turn to their communities as their source of inspiration and data, doubling down on what already exists.

Communities in Web3 form from both intrinsic and extrinsic rewards. Projects shouldn't hesitate to experiment with both, if they first establish their unique target audience is open to both. Curious scientists and engineers are often attracted to Web3 projects that advance Web3 technology. They may be curious to learn how the project works, meet like‐minded people, and exchange ideas. At the same time, they might be motivated by a financial incentive to start experimenting with a given dapp or participate in a bug bounty.

Bug bounty programs are excellent tools for projects to incentivize community members to expose vulnerabilities in their code. Hackers collect a financial reward for discovering a flaw in a project's code and revealing it to the team. Fixing a flaw before the code is live spares projects from painful exploits later. Web3 dapps are considered antifragile, which means they become stronger the more they are attacked. Bitcoin and Ethereum are widely considered secure because so many people have tried to hack them for so many years with such a large potential for reward that if it were possible, it probably would've already been done. Newer projects still need to battle‐test their code with bug bounties as well as formal audits before many users will trust them with value.

Community members need not be developers or engineers in order to contribute to a Web3 project's success. Meme contests, for example, are an excellent tool for communities to incentivize content creation. Thousands of fans are just as (if not more) likely than a small, centralized marketing department to surface what is most funny or compelling about a project or team. Then, even better and cheaper than A/B testing, other community members can vote on the best content. The project, its team, and community are then armed to share the winning memes about the project on platforms such as Crypto Twitter, driving project discovery. In Web3, marketers need not do all the work themselves; instead, they should focus on creating the conditions, often through rewards, for others with the right skills and incentives to work alongside them to achieve their goals.

As communities grow, though, there is a limit to what can be accomplished through extrinsic rewards. The Ethereum Foundation still offers financial rewards in the form of grants for projects building essential tooling on Ethereum that helps grow the ecosystem but doesn't necessarily have an economic model that would be rewarding for the creators. It does not, however, have a blanket policy of paying dapp developers to build on Ethereum as opposed to another Layer‐1 blockchain. That is because Ethereum is already the Layer‐1 blockchain of choice for the vast majority of Web3 developers, and also because this simply isn't the practice of the Ethereum Foundation, which figures that it can attract developers by building the best technology. Other Layer‐1 as well as Layer‐2 projects and dapps pay almost any developer to build on their platforms to attract more use cases and activity. This works up to a point. If builders only use a certain platform because they're paid to, as opposed to because the platform serves their use case the best, then they stop using it the moment they stop being paid. Extrinsic rewards for developers at the beginning of a new platform's life cycle can be an effective zero‐to‐one mechanism to drive early use, but if they don't stick with the platform when the cash dries up, it won't drive long‐term return on investment.

One of the best ways to evaluate Web3 platforms is to look at what developers use without being paid to do so, or which have tokenomic models that offer sustainable rewards over time. Eventually, if projects spend more to acquire a user than the value the user adds to the platform, they will run out of fuel. Marketers should take care that extrinsic rewards are actually working to grow their projects without simply creating the expectation for more rewards. They also shouldn't dismiss the power of intrinsic rewards. The early Ethereum meetups didn't work because they offered people beer and pizza. What they really offered was community and identity. Ethereum was a novel technology surrounded by wildly intelligent people. When computer scientists and developers encountered it, they wanted to adopt it as part of their identities. They took up the mantle as local Ethereum organizers and spent their nights and weekends planning meetups. Motivated by fascination with the technology, and willing to integrate it into their identity, their intrinsic motivation often yielded economic rewards. Those with prominent roles organizing Ethereum meetups built their networks and became top candidates for roles at Web3 startups or attractive Web3 founders to venture investors. Marketers should understand how their audience responds to different incentive structures, both intrinsic and extrinsic, over the short and long term.

Knowing one's audience in Web3 marketing doesn't only mean becoming familiar with the existing Web3 community. Consider a Web2.5 transformation project at a traditional or Web2 company, or a Web3‐native project targeting users and industries that aren't yet in Web3—both of these projects' target audiences may be completely unfamiliar with Web3.

Take the example of Audius, a Web3 music streaming platform that Serotonin helped to introduce their token in fall 2020. Most Audius listeners didn't know about Web3, and Audius didn't want to introduce crypto jargon into their messaging. Their users were there for the music, and they wanted to hear about artists and tracks, not NFTs and tokenomics. But Audius also wanted to launch a token by retroactive distribution to reward past users, align them with the team, and build a community going forward. We needed a jargon‐light way to distribute crypto tokens to music fans; our job was to message what they did rather than how they worked on a channel where the message could connect with its audience, while offering a friction‐minimizing UX for users to claim their tokens.

So we decided to host a concert. It was the depths of the pandemic, and everyone was stuck indoors. Music fans missing their favorite artists were tuning in to livestreams on platforms such as Twitch. Audius secured electronic music producer Deadmau5 to stream a show on the front page of Twitch. Once a user signed into the livestream, they could watch Deadmau5 spin. On the screen was a banner with simple text message instructions for how to claim their AUDIO—that is, their Audius tokens. The show was a massive hit, and the retroactive distribution of AUDIO earned some artists who had streamed to fans on Audius more value in a single transaction than they made all year in gloomy 2020.

Neither the artists nor music fans were served technical content about blockchains. If they wanted to learn more about the token, they could search or click from the main website to a separate, designated page. This clear audience segmentation with a specific user flow for the crypto‐curious enabled the users who wanted token information to access it. Most were satisfied with the what of Web3 without the how—that is, with claiming their AUDIO tokens and continuing to use the product. Non‐Web3 audiences don't care that a product uses blockchain technology. The most important message is the value proposition of the product from their perspective, in this case, getting inexpensive and direct access to their favorite music that they can't find elsewhere. For an energy‐focused project, that message might be a cheaper way to buy energy. For a gaming project, it might be access to a favorite game. Most people don't care about the technology underlying the software they use; it's a marketer's job to decide when this is true in Web3.

In our Web3 transformation practice, we even have examples of projects where we removed Web3 language entirely. Serotonin's NFT e‐commerce software spinout, Mojito, did a project with the NBA's Milwaukee Bucks in which fans had the opportunity to claim an NFT by retroactive distribution based on their past engagement with the Bucks, such as attending games. We called the NFTs digital memberships and left out the word NFTs. We and the Bucks agreed we didn't want to make fans feel like we were trying to sell them something; instead, we were merely trying to reward and align them into a community. The digital memberships also really were digital memberships. Owning one enabled the holder to access an online community, perks such as discounts on game tickets, and direct ways to engage with their favorite players. Similarly, Reddit launched a collection of NFTs and abstracted the blockchain language entirely, simply calling the series digital collectibles.

As NFTs become more ubiquitous, eventually there may be no need to use the word NFTs, and we may see more descriptions like digital goods or digital memberships. A store in the early days of the web would be called a web store; today it's simply called a store. As Web3 and the metaverse become the waters in which we swim, we will likely stop calling them by these names, and just call them the internet (or reality).

At this point in the growth of Web3, it's likely that any new Web3 project attracts a mix of Web3 native users and Web2 users. The Web3 natives will come with self‐sovereign wallets such as MetaMask and protest if a platform forces them to use a centralized, custodial wallet. They will expect to engage directly with the team behind the project and to earn rewards for being early adopters. However, Web2 users entering Web3 for the first time won't have wallets. They will fall out of the marketing funnel if they aren't motivated enough to convert through each step of a complicated UX. Used to being treated as consumers, they expect an arm's‐length relationship with the project team and may hesitate to join a noisy communication channel such as Discord or Telegram, which are the staples of Web3 communities.

The Web3 natives can be exuberant bordering on spammy. Web2 converts can be intimidated and skeptical. Armed with the knowledge that both Web2 and Web3 users will show up at their project's doorstep, Web3 marketers can put themselves in the other person's shoes and imagine the mentality and expectations coming in from Web2 or Web3. This helps us design messaging and onboarding that suits both of their needs, and accommodates those coming from Web2 along a journey of learning about Web3.

Mojito offers an excellent example. In building NFT e‐commerce infrastructure that enables vendors such as Sotheby's to sell NFTs on their own, owned websites (instead of on third‐party sites such as OpenSea or Nifty Gateway), Mojito recognized the need to accommodate both Web2 and Web3 users. It pioneered a progressively decentralizing wallet that's easy to set up and takes credit card payments. Although the wallet starts out centralized and custodial as a “multi‐sig” (a multi‐signature wallet with more than one private key), the user can choose to boot Mojito out of the multi‐sig and make the wallet self‐sovereign whenever they want. As a Mojito user learns more about Web3, how they use the platform can evolve with them. Native Web3 users can also come with their own MetaMask or other Web3 wallet. Software, UX, and marketing innovations are all necessary to meet users where they are and facilitate their Web3 journeys.

One pattern that has recurred over my time at ConsenSys and Serotonin is that token buyers buy tokens, and NFT buyers buy NFTs. There are obviously exceptions to this rule. Dapper's Top Shot NFT project with the NBA was remarkable in that a significant number of its buyers were NBA fans, as opposed to members of the Web3 community.2 Other NFT drops by prominent brands have attracted disproportionate numbers of fans of those brands over Web3 natives. Currently, however, this is the exception rather than the rule. Marketers working with brands with existing followings should learn about those audiences, especially whether some segments of them are already onboarded into Web3. Existing audiences that are already Web3 native are most likely to convert to buying a Web3 product. Highly engaged fans of the brand who are not Web3 native can be converted to buying a Web3 product, but unless they are highly motivated, there is usually a higher bar to converting them, including an easy UX for Web2 audiences and clear explanatory content. Projects with and without Web3 natives in their audiences should assume by default that their buyers are most likely to be Web3 natives, and they should take the time to learn about the Web3 community's preferences in order to tailor products and marketing to them.

Which is to say that, today, most people buying tokens and NFTs do so at least in part for their speculative future value, and are savvy about the factors that make a token likely to increase in price. Those factors include team experience, tokenomic design, community size, rate of growth, previous price history, and project road map. An experienced team with an excellent track record is a formula for attracting investment. Tokenomics should demonstrate how demand is stimulated and the mechanism by which an increase in demand increases price. Metrics matter, like Twitter following and Discord community size relative to level of engagement; low engagement can reveal that followers are bots or community members have been purchased. An ambitious yet credible road map that includes catalysts for step‐function growth, such as launching a dapp, metaverse land drop, or fungible token, is also a confidence builder. Projects trying to sell tokens or NFTs to communities that haven't traditionally bought them should be extra careful to verify that these communities would actually buy tokens.

When it comes to moving audiences and understanding how they connect with Web3 projects, we can learn from failures as well as successes. Civil, a Web3 journalism startup that was part of the ConsenSys portfolio, innovated a new business model for incentivizing local journalists. It planned to sell tokens to news readers that, among other benefits, enabled them access to the content generated by its newsrooms. With my history at HuffPost and Slant, I was fascinated by this project and sat on its board. As a result, I had a front‐row seat to the failure of its token launch. Despite attracting more top‐of‐funnel token buyers than any other project I'd seen at the time, the UX of buying the actual Civil token involved too many steps for non‐Web3 users, and they fell out of the funnel (the Web3 marketing funnel is the subject of Part 3 of this book). Web3 marketers shouldn't count on an audience that's never exhibited a behavior in the past, such as buying a token, suddenly taking that action on launch day. Wherever an audience is being asked to do something new, marketers should test and optimize the funnel to make sure it's actually converting. What happens on launch day should be pretested and never a mystery.

Conversely, I've also witnessed happy stories of projects that assumed they'd attract mostly Web2 users and ended up overrun with Web3 natives. At Serotonin, we worked on the Web3 transformation that earned Sotheby's—founded in 1744, one of the world's oldest and best‐known auction houses—inclusion on the Time 100 Most Influential Companies of 2022.3 Sotheby's saw the promise of NFTs early on and decided to make Web3 a core part of their business. Powered by Mojito and with support from Serotonin marketing, Sotheby's launched its own NFT e‐commerce site branded as Sotheby's Metaverse. They were cautiously optimistic about converting their existing clients, traditional art buyers, to buying NFTs on Sotheby's Metaverse. But surprisingly, the opposite happened: Web3 buyers flooded into Sotheby's Metaverse, many of them engaging with Sotheby's for the first time. Web3 natives would soon start bidding on traditional art items at non‐NFT Sotheby's auctions, such as when ConstitutionDAO bid on a copy of the US Constitution. Very quickly, Sotheby's decided to start accepting auction bids in ether, and even built a replica of its New York auction house in the Web3‐enabled metaverse world Decentraland, where virtual attendees anywhere the world can watch its live auctions and bid. With our support, Sotheby's benefited immensely from learning about its Web3 audience.

The best way to learn about your target audience is to observe them and ask questions. Here are some key questions to establish who your audience is and how to appeal to them:

  • Who has the problem that your product solves?
  • Do they already know they have the problem?
  • How are they solving the problem today?
  • What are the flaws or shortcomings of the current solution?
  • What value proposition of your product resonates with them?
  • On what platforms is this audience currently active?
  • Is there a particular type of content or signal they pay attention to?
  • What other products do they like, and what do these have in common?
  • What other communities or groups is this audience part of?
  • Are they allergic to any particular channel or messaging style?
  • How do they respond to intrinsic and extrinsic incentives?
  • In what ways are they able and willing to help with a project?
  • What kinds of activities is our community already engaging in?
  • What do members want from this community? From this product?
  • Why do they leave, and why do they come back for more?

The most powerful marketing strategies in Web3 come from deeply understanding one's product and audience. This is a necessary step toward figuring out how and where to connect the two, with the right messaging on the right channels. But first, projects need to figure out who will do this work, and that means building their marketing team.

Notes

  1. 1. This virtuous cycle is the classic method for product development from Eric Ries's canonical classic, The Lean Startup (New York: Crown Business, 2011).
  2. 2. Elizabeth Lopatto, “NBA Top Shot seemed like a slam dunk—so why are some collectors crying foul?,” The Verge, June 7, 2022, https://www.theverge.com/23153620/nba-top-shot-nft-bored-ape-yacht-club.
  3. 3. Andrew R. Chow, “Time100 Most Influential Companies of 2022: Sotheby's,” Time, March 30, 2022, https://time.com/collection/time100-companies-2022/6159438/sothebys/.
..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.
Reset