This traffic arrives late and warm
Someone hears you described on a podcast during a commute, or watches a review halfway through a long video, and does nothing at the time. They arrive two days later, having already heard several minutes of context about what you do, from someone who is not you.
That combination is unusual. Search traffic arrives with a question and no context. Ad traffic arrives with context you wrote and very little trust. This audience arrives with a lot of context you did not write, borrowed credibility from whoever mentioned you, and a specific claim in mind that they want to check.
They are frequently your best-converting visitors and your most frustrating, because the thing they want to confirm is often something the episode said and your website does not mention anywhere.
You will often have no referrer
The practical problem is that this traffic is close to invisible in the usual reports. A listener hears a spoken URL and types it. A viewer taps a link in a mobile app that strips the referrer. Someone reads a description, searches your brand name, and lands via organic search, where the number gets attributed to SEO.
So the honest starting position is that you cannot expect the channel to identify itself. Your live visitor list will show a lot of direct arrivals and brand searches, and no amount of staring at referrers will separate them from everyone else.
Two things work better than referrer analysis. The first is asking, once, in the conversation itself, in a way that does not feel like a survey. The second is watching the shape of the traffic rather than its source: this channel arrives in a distinctive wave that starts on publish day and decays slowly over weeks, quite unlike the sharp spike of a link from a busy site.
Answer the thing the episode promised
The single highest-value habit is to know what was actually said about you.
If a host described you as the cheap option, expect pricing questions and expect them to be sceptical. If a reviewer praised one specific capability, expect people arriving to check that capability exists and works the way it was described. If someone mentioned a limitation, expect that to be the first thing asked, and expect the asker to already believe it.
Listen to or watch the segment yourself and write down the three claims a reasonable person would take away. Then make sure someone on chat can answer each of them in a sentence. This takes twenty minutes and it is the difference between a warm visitor being confirmed and a warm visitor discovering that the company does not recognise its own description.
Be careful with claims you did not make. If a host overstated something, the chat is where that gets corrected, and correcting it plainly costs you far less than letting someone buy on a misunderstanding. Say what is true, say what the episode got slightly wrong, and do not pretend the difference does not exist.
Make the landing page carry the context
Most of this audience lands on your homepage, because that is what gets said out loud. Your homepage is written for people who arrived from a search, which means it explains what you are, which is the one thing this visitor already knows.
You do not need a dedicated landing page for every mention. What helps is making sure the specific thing likely to be discussed is reachable in one click from the homepage, and that the words used in the episode appear somewhere on the page. A visitor scanning for the capability they heard about will give up quickly if your site calls it something else entirely.
Where a mention is large enough to justify it, a page that acknowledges it directly works well, and it gives you a clean way to see the traffic, because a page nobody links to from your own navigation is easy to isolate in reports.
Staffing around a publish date
Unlike most traffic sources, this one has a schedule you can know in advance.
If you have been interviewed, or you know a review is coming, ask when it goes out. Then treat the publish date the way you would treat a campaign: make sure chat is staffed that day and the following two, and make sure whoever is on knows what the segment said. A wave of unusually well-informed visitors hitting an unstaffed widget is a bad trade for whatever the appearance cost you.
The decay is long. Podcast back-catalogues get listened to for months and video recommendations resurface unpredictably, so a mention keeps producing arrivals long after everyone internally has forgotten about it. Keep the notes about what was said somewhere the team can find them, rather than in the head of whoever did the interview.
How MyLiveChat fits
The live visitor list shows the page a visitor is on, how long they have been there, the path they took through the site and their referring URL where the browser provides one. For this channel the referrer is often empty, which is itself information: a rise in direct arrivals landing on the homepage and moving straight to one specific feature page is the signature of a mention.
Because proactive invitations can be triggered by the URL a visitor is on, you can put a specific greeting on the page that a mention drives people to, without changing the greeting on the rest of the site. That is the practical way to say something relevant to this audience while it lasts, and to take it down again when the wave has passed.
What to measure
Do not try to build attribution for this. You will spend more effort than the answer is worth, and the answer will still be wrong.
Measure two simpler things instead. First, the baseline: know your normal daily direct-and-brand-search volume so that a wave is visible as a departure from it, dated against the publish day. Second, the content of the conversations: if the questions arriving this week reference something you did not write, you are seeing the channel, whatever the referrer says.
Over a longer horizon, the useful comparison is not volume but quality. Count how many conversations from a mention week turn into something real, and compare it with an ordinary week. This audience usually converts better, and knowing by how much is what tells you whether the next appearance is worth the time.