A collection of representative B2B lead discovery scenarios, showing how AI identifies qualified sales opportunities from real-world business conversations.
The Spread Moved from 0.8% to 4%: A Market-Making BD Needs the Contract End Date, Not Another Complaint
Two Telegram messages report the same spread widening, but only one carries a contract end date and an invitation to compare proposals. Here is how a market-making BD lead tells them apart and what to verify before the first reply.
This is an illustrative scenario designed to explain the product’s judgement logic. It is not a real customer case, testimonial, contract, revenue result, or conversion claim.
01Situation
02Signal judgement
03Confidence vs priority
04Human next step
Signals considered
- The spread change includes a measurement basis
- The incumbent contract end date is mentioned
- The project invites an alternative proposal or quote
This is a simulated composite scenario. Its messages, numbers, deadlines and business circumstances are illustrative and do not represent a real customer, conversation, contract or result.
Two Messages, Two Different Follow-up Decisions
On an ordinary afternoon, a market-making business development lead might see several spread complaints across the Telegram groups they monitor and treat all of them the same way: a quick read, a bookmark, no reply. Complaints are cheap; replies are not. Each follow-up costs time, and most messages do not end in a conversation worth having.
Then two messages about the same market maker arrive hours apart. The first is a complaint and nothing else. The second is a complaint that adds a contract end date and an invitation to compare proposals. Your decision right now is not which quote to send or which project to rank first. It is which message earns a follow-up at all — and what you must verify before your first reply, because answering every complaint fills a day, while answering none can leave a genuine replacement process unnoticed.
The messages below are illustrative composites built for this article. The numbers, dates, and wording are not a transcript of any real group and are not evidence about any real market maker.
Message A (paraphrased): “Our market maker’s spread went from 0.8% to 4% this week. This is unacceptable.”
Message B (paraphrased): “The spread on ETH/USDT moved from about 0.8% to 4% over the past week. Our contract ends in six weeks and we are comparing options. Could you send a proposal we can place next to the current one?”
Both messages report the same movement. They are not the same lead.
A Spread Number Without a Window Says Little
First, the terms. A market maker is a firm that continuously quotes both a buy price and a sell price for a token, earning part of its revenue from the difference between them. That difference is the bid-ask spread: the gap between the highest price a buyer is willing to pay (the bid) and the lowest price a seller will accept (the ask), usually expressed as a percentage of the midpoint price.
A wider spread is observable. Why it widened is not. A spread can widen because volatility rose, because liquidity thinned, because the market maker was carrying inventory it needed to hedge, or because it deliberately changed its pricing policy. One message cannot tell you which cause applies, so it also cannot tell you that the incumbent provider failed. The same number could reflect market conditions rather than performance.
Message A contains one metric and one judgement, and nothing that anchors either. There is no trading pair (the pair of assets being traded, such as ETH against USDT), so the claim cannot be checked against market data. “This week” is not a measurement window you can verify after the fact. No decision owner is named: nothing in the post says the author can choose a provider, and a display name or posting time is not evidence of authority, severity, or buying intent. The author could be a community member, a junior operations person, or someone venting about a volatile week. The same text could be performance noise or the first step of a provider search — the message itself does not say which.
That is why Message A gets a bookmark, not a reply.
What the Second Message Adds: a Date and an Invitation
Message B reports the same movement and then adds four facts that change the reading:
- A trading pair (ETH/USDT), so the number can be checked against real market data.
- A measurement window (“over the past week”), which turns “4%” from a standalone number into a claim about a period.
- A contract end date (“six weeks”), which makes a replacement window concrete.
- An invitation to send a proposal, which suggests someone is collecting options now, not at some distant renewal.
The invitation is the strongest signal and the easiest to over-read. It is evidence that an evaluation is in progress, not that the project will switch, that the current market maker cannot fix the issue, or that you are the only provider being asked. Message B also leaves unknowns: whether the author is the decision owner or a relay for the treasury or trading desk; whether the contract has a notice period or an exclusivity clause that changes the timeline; whether other market makers have already been invited; and whether “4%” is an average over the week or a single peak print.
Verify These Five Facts Before the First Reply
Before writing anything, sort what the group message establishes from what it leaves open:
- Trading pair. Observable in Message B. Without it, the claim cannot be checked at all.
- Measurement window and basis. The post says “past week,” but is 4% an average or a peak, and compared against what benchmark? Unknown.
- Contract term. The end date is stated; the notice period, exclusivity, and what happens at renewal by default are not. Unknown.
- Decision owner. The author may be the decision maker or a messenger for someone else. Unknown from the post.
- Willingness to compare. The invitation is evidence that proposals are being collected; how many, and from whom, is unknown.
Each of these can be resolved with one or two short questions. None should be assumed from the tone of the message or the presence of a screenshot. The point of the list is not to qualify the poster — it is to decide whether your reply is worth writing.
A First Reply That Checks Facts, Not Commitment
A restrained first reply confirms what the message left open instead of pitching pricing or asking for a call. One example, built for this article:
“Thanks — the pair and the window help. Was 4% an average over the past week or a peak print, and over what hours? If you are collecting proposals, who should we address ours to?”
The reply does three things: it pins down the measurement basis, it asks for the decision owner, and it leaves the next move with the poster. It deliberately does not quote terms, because a number without a window and a decision without an owner cannot support a proposal yet. It also does not chase: after the question is asked, the natural next step is to wait for the answer.
Where the Evidence Stops
The useful habit is separating what a group message proves from what it only suggests. A number proves little by itself. A number plus a pair, a window, a contract date, and an invitation suggests a real evaluation — and even then, the decision owner and the term details stay unverified until someone answers.
This is where monitoring tooling earns a narrow place in the workflow. When the same complaint is forwarded across several authorized groups you intentionally connected, Top Prospect keeps the original message, its source, and its context together, and it deduplicates repeated forwards so your human review starts from distinct posts rather than copies. It does not automatically contact anyone in those groups, and it cannot certify that a project is actually switching. Separating the 4% complaint that means nothing from the one carrying a contract date is still a judgement call — and it belongs to you.