The request is already written
A cold list is assembled from what a company can see about other companies and people: the sector they work in, their size, their job title, where they are. Everyone on it fits the description of a likely buyer, and almost nobody on it has said anything about wanting to buy. The outreach that follows has to create the interest before it can ask for a meeting, which is why it takes so many messages and why so much of it goes unanswered.
A public request starts at the other end. Somebody posts in a forum, on Reddit, in a Facebook group or in a community group asking who can supply a product or take on a piece of work, and the post usually says what they need, roughly when they need it and sometimes what they have already tried. That person has done the part of the work a cold campaign spends its budget on. They know they want to buy, they have said so in writing, and they are asking strangers to help them choose.
That is why we treat these posts as some of the strongest leads a company can find outside its own database, and also as some of the most perishable. A request tends to be answered by whoever sees it first, and the thread fills up with suggestions while the company that sells exactly what was asked for is busy with its own week and never knows the post was there.
Why a cold list costs more than it appears to
The price of a list is the smaller part of what it costs. Every name on it has to be researched or enriched, written to several times and kept off the next campaign once it has declined, and each of those messages goes out from a sending domain whose reputation suffers when people mark them as unwanted. The company pays for attention it does not yet have, and the bill arrives in sender reputation as well as in hours.
The conversations are harder too. A contact reached from a list starts the call deciding whether they have a problem at all, and the salesperson spends the meeting making the case for the category before they can make the case for the company. A person who posted a request starts the conversation by comparing suppliers. The questions are about scope, availability and price, which are the questions a salesperson wants to be answering.
None of this makes prospect research useless. Some markets never ask in public, and for them a carefully researched list with a named decision maker at each company is the right tool. The point is about order. Where buyers do ask in public, those requests are worth catching before any budget goes on reaching people who have not asked.
Where the requests sit, and why reading them by hand fails
The requests are spread across a lot of places. A single market can have a few large forums, several subreddits, a number of Facebook groups with overlapping members and community groups run by neighbourhoods, associations or professional networks. Each one has its own rules, its own way of tagging posts and its own regulars who answer first.
The wording varies as much as the places do. One person asks for a recommendation, another asks whether anyone has dealt with a particular company, and a third describes a problem without naming the service that would solve it. A keyword search catches the first and misses the other two, and it also catches every post where the same words turn up in a complaint, a joke or an advert from a competitor.
Most teams that try to do this by hand end up with one person checking a handful of groups when they remember to. It works for a while, and then a busy period comes, the checking stops and nobody notices, because nothing visible breaks. The requests keep arriving and somebody else answers them.
What we build
We agree the sources with the client first: the forums, subreddits, groups and community boards its buyers actually use, and which of those allow a company to take part. The automation watches those sources and reads each new post.
Every post is checked against criteria agreed with the client before anyone sees it. The checks establish whether the post is a request at all, whether what is being asked for is something the client sells, whether the person is somewhere the client can serve and whether the request is still open. Posts that fail any of them are dropped, and the criteria are written down so the client can change them when its offer changes.
A post that passes goes to a named person on the client's team with the original post linked, the part of it that matched and the reason it qualified, so that person can judge it in the time it takes to read it. The same lead is logged in the client's CRM with its source, the post and a stage, and where the same request can be recognised in several groups the records are merged, so the team answers once.
The reply itself comes from the client's team, written by a person in the company's own name. Where the request leads to an email address or a form enquiry, the client's follow-up sequence picks it up where one is in place, and a reply from the contact stops the sequence and goes to a person.
Answering in public
A reply in a public thread is read by the person who asked and by everyone else who reads the thread, then and later. It is a piece of public writing from the company, and we recommend it is treated like one. The replies that work answer the question that was asked, say plainly who is writing and which company they work for, and leave the person room to choose.
A pasted sales message does the opposite. Many groups have rules about promotion, some forbid it outright, and moderators can remove posts and members who ignore them. The team reads the rules of each group before anyone posts, and where a group does not allow businesses to answer, the right decision is to stay out of it.
Watching public posts also raises questions about what may be collected and stored. Platforms set their own terms for automated access, and privacy rules may apply to keeping what people post alongside their names. Each client should check which rules apply to its business and to each source before the work starts.
