Three thresholds of perception: why a message fails, and which one your Russian campaign is stuck at

A message can be noticed and still not work. It can land on the first try, or need months of repetition before anything shifts, or be heard dozens of times and move the receiver by nothing at all. These are three different outcomes, and behind them stand three different doors that a signal has to pass through before it has any chance of an effect.

For a company selling into Russia from abroad this matters more than it does at home. The message crosses a language, a platform and a buyer who has no prior context about you. When nothing happens, the usual verdict is “the Russian market did not respond”. That verdict is useless, because it does not say which door stayed shut.

The three thresholds a signal has to clear

Perception filters are the thresholds through which a receiver lets a signal in or screens it out. There are three, and they work in sequence, like checkpoints with different criteria at each one.

The threshold of perception decides what gets noticed at all. The threshold of tolerance decides how much strength and persistence the receiver is willing to bear before rejecting the whole thing. The threshold of sensitivity decides which signal is capable of causing a change, and which one passes through without a trace, even after it has been noticed and endured.

It pays to keep the three apart instead of treating reception as one blunt “did it get through or not”. A message can clear the first threshold, be noticed, and fail at the second right away because it is too pushy or too loud for this receiver. It can clear both and never touch the third: it is noticed, tolerated, filed as a fact, and that is all. From the outside the three failures look identical. The message “did not work”. The causes are different, and so are the cures.

Why each threshold needs a different fix

Not noticed. If the message does not clear the threshold of perception, the problem is strength or contrast. The signal is too weak, too quiet, too similar to the background to be picked out of the noise. The fix is in form, not content: raise the contrast, change the moment of delivery, separate the signal from whatever competes with it. Rewriting the content of a message nobody saw is pointless. Nobody has read it closely enough to judge the content yet.

Noticed and rejected. If the message was seen but fails the threshold of tolerance, the problem is the reverse: too much pressure. Repetitions come too often, the tone is too insistent, the contrast too sharp. The receiver reads it as an attack, and the system defends itself by rejection, sometimes harder than if there had been no signal at all. The fix is to lower the pressure, give space, spread the repetitions out in time. The paradox is that intuition says the opposite. If it did not work the first time, repeat it more insistently. In practice that pushes the message closer to the point of rejection.

Noticed, tolerated, and nothing happened. If the message was seen and endured but fails the threshold of sensitivity, it did not touch anything that matters to this receiver. This is the failure that is hardest to see from outside. Formally everything went right: the receiver nodded, agreed, rejected nothing. And nothing changed, because the signal never reached the level where decisions are made. The fix is neither strength nor frequency. It is precision: find the one sensitive point that can move this particular receiver’s decision rather than register as one more fact.

Where foreign campaigns get stuck in Russia

The thresholds are the same everywhere. What differs by market is where the signal typically stalls, and for foreign companies in Russia the pattern is fairly predictable.

The first threshold is where most translated campaigns quietly die, and it rarely looks like a failure in the account. Yandex Direct reports impressions, Yandex Metrica reports visits, so the signal has technically been delivered. But a search ad translated from an English original keeps the contrast logic of a Google results page. On a Yandex results page the surrounding ads follow local conventions, and the translated one reads as background. The same happens with email: a sequence built for a Western inbox may never be noticed by a buyer whose working conversations run through Telegram and WhatsApp. The message is not rejected. It is simply not where the receiver is looking.

The second threshold is easier to trip in Russia than many foreign teams expect, precisely because the channels are more personal. A nurture cadence that is normal for email gets copied into a Telegram bot or a channel, where every message arrives as a push notification on the buyer’s phone. A chat is a closer space than an inbox. The same frequency that reads as diligence by email reads as intrusion in a messenger. And the exit is one tap, with no “manage your preferences” page to slow it down. The rejection is quick, final, and invisible to the sender, who sees only that the channel went quiet. Retargeting in the Yandex advertising network behaves the same way: the buyer who sees your banner on every site they open does not conclude that you are serious. They conclude that you are following them.

The third threshold is the one translation cannot fix by definition. A value proposition is built around the sensitive point of a specific buyer. At home that point might be heritage, design, or an environmental standard. A Russian buyer of a foreign product in the current environment often has different questions before any of that: how payment will actually go through, who services the product locally, whether the supplier will still be reachable next year. In my experience a message that argues brilliantly to the wrong sensitive point produces exactly the third failure: attentive meetings, polite agreement, no decision. The buyer heard everything and nothing in it touched the place where the decision lives.

How to tell which threshold your message is stuck at

Before changing the wording in the hope that it will finally work, diagnose where it stalls. If it was not noticed, raise the contrast and change the form, and leave the text alone. If it was noticed but, by all signs, pushed away, lower the pressure and frequency instead of arguing harder. If it was noticed and tolerated and nothing changed, look for a more precise formulation rather than a louder one: the single thing that is sensitive for this person.

The data for this diagnosis is usually already in your accounts, as long as you read it as behaviour of a receiver rather than as campaign metrics. Impressions without clicks, letters that stay unopened, a landing page people leave before the first scroll: these are first-threshold signs. Unsubscribes, a muted channel, a blocked bot, replies asking you to stop writing, complaints to the platform: these are second-threshold signs. Visits that read to the end, replies that agree, meetings that end with “we will think about it” and a pipeline that never moves: these are third-threshold signs. Three patterns, three remedies, and no amount of rewriting substitutes for choosing the right one.

The same diagnosis applies to any repeated conversation that produces nothing. A request that is heard and not carried out. Advice taken with a nod and forgotten. A warning that is noticed and ignored. In every case the question to ask is not “what should I say differently” but “at which threshold is this stuck”, and only then to pick a remedy for that specific barrier instead of for everything at once.

What a company selling into Russia from abroad should do

A short route that follows from the three thresholds:

1. Separate “not seen” from “seen and ignored” before touching the copy. Pull the behaviour signals from Yandex Direct, Yandex Metrica, your mailing tool and your messenger channels, and sort them by threshold. Most teams skip this step and start rewriting, which treats three problems with one remedy that fits at most one of them.
2. Build a reserve into the signal for the first threshold. Assume the message loses strength in transit: through translation, through a platform with different conventions, through a buyer who meets your name for the first time. Contrast has to be set against what surrounds the ad on a Yandex results page, and the channel has to be the one the buyer actually watches. This is the part of the job that Yandex Ads setup and the landing page share.
3. Set a tolerance budget per channel. Decide how many touches a week a buyer gets by email, by Telegram, by retargeting, and treat the messenger budget as much smaller than the email one. When the receiver signals fatigue, stop, instead of adding one more reminder.
4. Find the sensitive point with local research, not by translation. The argument that moved buyers at home is a hypothesis in Russia, nothing more. Talking to Russian buyers before the campaign is what customer research is for, and it is cheaper than discovering the wrong sensitive point from a quarter of polite meetings.
5. Fix one threshold at a time. Raising contrast, lowering frequency and sharpening the argument at once makes it impossible to see which change worked. Diagnose, change the one thing the diagnosis points to, and read the behaviour again.

The three filters are rarely named out loud, yet they silently decide whether anything changes after a conversation. Knowing where exactly the message got stuck is worth more than any new attempt to say the same thing again, only louder.

Frequently asked questions

Sources

• Russian version of this article on belokrylovo.ru: Три порога восприятия: почему сообщение не срабатывает