Volume bot versus paid advertising

Where crypto ads can actually run

The first thing that surprises teams new to this is how small the legitimate inventory is.

Mainstream advertising platforms restrict cryptocurrency promotion heavily, and promoting an individual token is the most restricted category within that. Policies vary by region, change without much notice, and enforcement tends to be account suspension rather than a warning. Building a plan around mainstream reach usually ends with a disabled account and a lost deposit.

What remains is crypto-native inventory. Screener promotion slots. Placements inside wallets and aggregators. Newsletter and community sponsorships. Direct arrangements with sites that cover the space. These audiences are far smaller than a mainstream platform and far better qualified, since everyone seeing them already holds a wallet and knows what a mint address is.

The pricing model is different too. Rather than an auction with transparent clearing prices, most of this inventory is negotiated, which means two buyers can pay very different amounts for the same slot and neither will know.

What you are really paying per visitor

Headline pricing in this market is close to meaningless without one adjustment: what fraction of the audience could plausibly act.

A placement quoting a large number of impressions may reach an audience that is mostly other projects, bots, or people who have never bought a token on Solana. A smaller placement in a focused community may reach a fraction of that number with several times the relevant share. The second is frequently the better purchase at a worse headline rate.

The only diligence that reliably works here is historical rather than promised. Ask what other tokens ran the same placement and look at what happened on-chain around those dates. Trade counts, unique participating addresses and whether anything persisted afterwards are all public. A placement that has never produced measurable activity for anyone is information, and it is information the seller cannot suppress.

Ad traffic is the least forgiving audience you can buy

This is the part that decides whether advertising works at all, and it is structural rather than a matter of creative quality.

Somebody arriving from an advertisement did not seek your token out. They arrived sceptical, they know they were advertised to, and they will spend a few seconds deciding. In those seconds they read four things: the chart shape, how recently anything traded, the liquidity figure, and the holder concentration.

A token that shows a handful of trades in the last hour, a jagged chart and a pool that would punish a real order fails that inspection every time. The advertisement worked perfectly; the page did not. And the specific damage is that impressions are not renewable. Those people have now formed a view, and a second flight next month reaches many of the same eyes with the memory attached.

Compare that with somebody arriving from a screener ranking. They were browsing tokens by activity, which means they self-selected into looking, and the ranking itself is a form of endorsement. That traffic is more forgiving because the visitor arrived already interested rather than interrupted.

Side by side with activity

The last row is the honest limitation on the right-hand column. Activity produces figures on surfaces that people browse; if your token's audience does not browse those surfaces, the figures are produced into an empty room. That is the case where advertising genuinely adds something no amount of activity replaces.

There is a second-order cost to advertising that rarely gets counted and belongs in the comparison. Running ads for a token invites scrutiny of a specific kind: people who see an advertisement for something they have not heard of go looking for reasons to dismiss it, and the first place they look is the on-chain record. Holder concentration, liquidity depth, whether the pool is locked, when the mint was created. All of it is public and all of it takes under a minute. Advertising therefore raises the standard your foundation has to meet, rather than substituting for it, and a project that would have passed unnoticed can fail a check it invited.

When ads make sense

• The foundation is finished. Depth that absorbs a realistic order, working metadata, a holder list that does not end the conversation.
• The page already looks alive. Recent trades, distinct participants, a chart that is not a single spike.
• There is something to announce. Attention with no news attached converts poorly however it arrived.
• The audience is specific. A placement in front of people who plausibly want what you built, rather than the largest number available.
• You can measure it. Meaning you know what your baseline organic activity looks like beforehand.

Miss the first two and the spend is not merely wasted, it is counterproductive, which is the same failure the week-one guide describes happening for free with launch-day attention.

One sequencing note that saves money. If you are going to do both, run the activity first and let it establish for several days before the advertising flight begins, rather than starting them together. Two reasons. The page needs to already look settled when the traffic arrives, not to be visibly changing while people inspect it. And you need a stable baseline to measure the flight against, which is impossible if both variables move at once. Teams that launch everything on the same morning end up unable to say which spend did anything, and usually repeat both next month for the same reason. The boosts comparison makes the same argument for screener promotion, where the timing question is identical.

Measuring either one honestly

The measurement discipline is identical for advertising, influencer placements and screener promotion, and it comes down to one rule: count only what you did not pay for.

Activity you purchased proves nothing, because you purchased it. Impressions you purchased prove nothing either. The number that matters is on-chain participation from addresses with unrelated history during and after the flight, and whether any of it persisted once the spending stopped.

Establish that baseline before you start, because otherwise you cannot tell the difference afterwards. Then read the transfer history rather than a dashboard. It takes a few minutes and it settles most arguments about whether a spend worked, including arguments with the person who sold it to you. The influencer comparison works through the same measurement problem from the other direction, and the conclusion is the same: the interesting number is always the one you did not buy.