A founder with a fixed marketing budget reaches the same fork every quarter. One path buys another round of creator posts and a new quest campaign, and the follower count jumps. The other path fixes the onboarding flow and publishes something useful, then waits on slower results that are harder to screenshot.
The past year made that choice easier to judge. According to Chainalysis’ 2026 Global Crypto Adoption Index, total crypto market capitalization fell about 50% in the 12 months ending June 30, 2026, a loss of roughly $2.1 trillion. Over the same period, on-chain economic activity slipped only 1.6%, from $9.5 trillion to $9.4 trillion. Prices and attention swung hard. Usage barely moved.
That gap will shape crypto marketing in 2027. a16z crypto’s State of Crypto 2025 report estimated roughly 716 million crypto owners worldwide, yet only 40 million to 70 million active on-chain users. Awareness is part of the problem for most projects. The bigger part is getting people who already know about a product to use it a second and third time.
What follows is INORU’s reading of the past year’s public evidence, drawn from published research, on-chain data, platform policy changes and regulatory updates. Our argument is simple: campaigns tied to a specific product action held up through 2026, and they are the ones worth funding in 2027. Campaigns measured mainly by attention deserve hard scrutiny before next year’s budget is set.
What is working going into 2027
The approaches below have one thing in common. Each ties marketing spend to something a user does after they see the campaign.
Clear positioning before paid promotion
Many projects still buy reach before they can explain themselves in a sentence. When a visitor lands on a token page and cannot tell what problem the product solves, the money spent getting them there is gone.
Positioning work rarely gets attention, but it raises the return on everything that follows. Once a team has named its category and its user, creator briefs get sharper and ads become easier to test. Content gets easier to cite as well.
- Start by writing one sentence on who the product is for and what they can do with it today, then test it on five people outside the team before any campaign goes live.
- Track the share of landing-page visitors who reach a meaningful first action, such as connecting a wallet or completing a test transaction.
- Positioning cannot rescue a product that is not ready. If the core flow breaks, clearer messaging only sends more people into a broken experience.
Communities measured by what members do
For illustration, a Telegram group with 40,000 members can be quieter than a Discord server with 3,000 active contributors. The value of a community sits in what its members do: they answer each other’s questions, test features, report bugs and stick around through a slow market.
The practical step is to pick two contribution behaviors that matter for your product, for example governance votes and testnet feedback, and reward those. Message volume is easy to farm and says little. Weekly active contributors make a better headline number, and the share of month-one members still active in month three tells you whether the community is real.
This takes time and steady moderation. An engaged community rarely produces the overnight growth curve that a member-count campaign can buy.
Creators chosen for audience fit
Creator marketing still works in crypto when the creator’s audience matches the product’s likely user. It fails quietly when the brief says “post about us” and the selection metric is follower count.
The rules around paid attention also shifted this year. On January 15, 2026, X revised its developer API policy to stop allowing apps that reward users for posting. Its head of product blamed those apps for AI-generated spam and reply flooding. Kaito sunset its Yaps leaderboards the same day. Projects that depended on paid posting volume lost a channel overnight, while projects with real creator relationships carried on largely unaffected.
Regulation matters too. In the EU, MiCA’s marketing communication requirements have applied to crypto-asset promotions since December 30, 2024, and the transitional period for unauthorized service providers ended on July 1, 2026. When an issuer pays a creator to promote its token, that post can count as a marketing communication, and the compliance duty sits with the issuer or offeror.
- Before signing, ask for audience geography and past campaign examples, and check how real the engagement looks.
- Give each creator a unique tracking link or referral code tied to wallet-level actions, then compare creators on funded wallets or completed first transactions. Impressions alone tell you very little.
- Niche creators with strong audience fit are harder to find and slower to scale than large accounts, so plan for a longer sourcing process.
Research-led content that AI answers can cite
Search still sends traffic, but it sends it differently. Ahrefs’ follow-up study on AI Overviews compared click-through data from December 2023 and December 2025 across 300,000 keywords. It estimated that an AI Overview on the results page was associated with a 58% lower click-through rate for the top-ranking page.
Content still earns its budget, but its job has changed. Pages that get cited inside AI answers, and brands people search for by name, matter more than generic explainers chasing informational keywords.
Ahrefs’ separate study of 75,000 brands found that YouTube mentions had the strongest correlation, about 0.74, with brand appearances in ChatGPT, Google AI Mode and AI Overviews. The number of pages a site published showed little relationship, at roughly 0.19. Ahrefs cautions that these are correlations and do not prove cause. The pattern still suggests that being discussed widely and credibly matters more than publishing volume.
In practice, that means fewer and deeper pieces built on something original, such as your own protocol data or a documented comparison. Write important passages so each one names its subject and states the finding with the evidence attached. Pair major pieces with video explainers, since YouTube is where the strongest correlation showed up.
Measure branded search growth and how often your brand appears in AI answers to core category questions. Assisted conversions from content belong in the same report. This work is slow, so expect months, and expect no guarantee that any AI system will cite a given page.
Incentives tied to retained behavior
Airdrops and points programs are still useful for distribution. What matters is the behavior they reward.
Delphi Digital’s June 2026 research tracked 3.7 million wallets over five years across six major token distributions: UNI, ARB, JUP, PENGU, VVV and MON. Between 78% and 94% of recipient wallets sold most of their allocation within 90 days. Delphi also found the 90-day sell rate ran 4 to 11 percentage points higher than the 30-day rate, so a team that reports only 30-day retention is overstating how many recipients stay.
Hyperliquid is the usual counterexample, and Delphi’s explanation is telling. It credited the outcome mainly to revenue-funded buybacks that absorbed selling pressure. Product economics carried the result; the distribution design played a smaller part.
- Reward behavior that repeats over several weeks. One-off tasks are easy for farmers to script.
- Use sybil filtering and vesting where they fit the token design.
- Compare 90-day cohort retention of incentivized users against organic users.
- Accept that sybil detection is imperfect, and that stricter eligibility rules can frustrate genuine early users.
What is underperforming
These practices persist for understandable reasons. Most produce numbers that look good in a weekly update.
Follower totals as a headline metric
Teams lean on member counts and follower totals because they are easy to track and easy to show investors. The trouble is that these numbers can be bought or botted, and incentive hunters inflate them further. A rising count can sit alongside a shrinking user base.
Audience totals still give a rough ceiling on reach, and first-time visitors do read them as social proof. Report active contributors and retained wallets next to every audience figure so no vanity number travels alone.
Paid campaigns that stop at the click
Ad platforms report clean numbers quickly, which makes paid campaigns feel accountable. Optimizing for clicks or email sign-ups tends to reward the cheapest traffic, though, and cheap traffic often never funds a wallet. Crypto ad policies also vary by platform and region, which can narrow targeting further.
Paid media still earns its place for retargeting past visitors and for time-sensitive launch announcements. Point paid traffic at one specific product action and judge each campaign on cost per completed on-chain action. Cost per click belongs in the appendix.
Launch-week bursts with no plan for day 31
Many budgets concentrate almost entirely around the token generation event. Delphi’s finding that selling keeps climbing between day 30 and day 90 suggests holder retention is won or lost in the weeks after launch, which is exactly when many teams go quiet.
Launch visibility does help with listings and early liquidity. It needs follow-through: a 90-day post-launch calendar with product updates and usage guides, plus plain reporting on roadmap progress and treasury decisions.
Reporting impressions as if they were attribution
Reach is the easiest number to report, because connecting campaign exposure to wallet behavior is hard. Without that link, nobody can tell which spend produced users and which produced noise.
Impressions do confirm that distribution happened. The upgrade is to tag every campaign link and capture the wallet connection event on your site, then join that with on-chain data through analytics tools such as Dune or Nansen, within the limits of applicable privacy rules. Attribution in crypto will never be complete because users switch devices and wallets. Partial attribution still beats guessing.
What founders should change next
Before 2027 budgets are locked, run the campaign review as a sequence.
- Name the one action that defines a real user. For a DEX it might be a second swap within 30 days. For a wallet, it might be a funded account. Everything else becomes supporting data.
- Audit the last 90 days by channel. List spend and reported results, and where possible count how many users each channel delivered to that action.
- Sort every channel into continue, fix or stop. Keep what produces the action at a cost you can defend. Fix what shows promise but lacks measurement. Drop what still cannot connect to the action after a fair test.
- Repair measurement before adding budget. Tracking links and wallet connection events should be live before the next spending increase.
- Check positioning and product readiness. If conversion stalls at the first product step, the next dollar belongs in onboarding or messaging. More reach will only make the leak bigger.
- Review again at 90 days, using cohort retention as the yardstick. Launch-week peaks flatter almost every campaign.
This sequence avoids universal budget splits on purpose. The right mix depends on project stage. A pre-launch protocol needs positioning and community foundations first, while a live product with users needs retention and attribution work.
The decision that matters most
The evidence from 2026 points to one test for every line in a 2027 crypto marketing budget: can this spend be connected to a user doing something that matters for the product? Attention still has value. This year showed how fast it moves while real usage holds steady.
Teams that apply the test will cut some comfortable metrics and fund some slower work. They will also know which parts of their budget are working.
INORU’s crypto marketing team works across positioning, community management, creator vetting, incentive campaign design, and search and AI visibility. If you want a second opinion on where your current campaign stands, we are happy to talk it through.