Is cold outreach becoming less effective because of AI-generated messages?
Posted by Applied AI Commons team (community team) on
As writing tools make it easy to produce personalised-sounding messages at scale, inboxes are filling with outreach that looks tailored but often feels generic. That raises a fair question about whether recipients are becoming more sceptical, and what still earns their attention.
There are several ways to look at this. Some members may focus on what genuine personalisation looks like compared with surface-level details. Others may discuss trust, sender reputation, and whether lower volume with better relevance works better. It is also worth asking whether the channel itself is changing or simply the standard for quality.
What changes have you noticed in how prospects respond to outreach, and what still earns a thoughtful reply?
5 replies
Applied AI Commons team (starter reply from the community team),
Genuine personalisation usually connects something specific about the recipient to a reason the message is relevant now. Surface-level personalisation mentions a detail, such as a recent post or a university, without that connection. A quick test is to remove the personalised line and see whether the rest of the message changes. If it does not, the personalisation is decorative. Recipients increasingly seem to recognise that pattern, so one well-reasoned observation tends to land better than several shallow ones.
Applied AI Commons team (starter reply from the community team),
Volume has knock-on effects beyond reply rates. Sending large numbers of similar messages can harm sender reputation and deliverability, which then affects every message from that domain, including the good ones. Common practices include: - separating outreach from main business email domains - warming up new sending addresses gradually - keeping lists clean and removing addresses that bounce - monitoring spam complaints closely Lower volume with tighter targeting can protect the channel as well as improve the quality of conversations.
Applied AI Commons team (starter reply from the community team),
A counterpoint: it may not be AI that makes outreach less effective, but the volume it makes cheap. Poorly targeted outreach existed long before writing tools; AI simply lowered the cost of producing it. Messages that were relevant before are probably still relevant now. The real change may be that the bar has risen, because recipients now see so many messages that look personalised that they judge more harshly on relevance and on whether the sender clearly understands their situation.
Applied AI Commons team (starter reply from the community team),
One practical approach is to use AI for the research and drafting steps, but keep a person responsible for the final judgement on each message. For example: - the tool gathers context and suggests an angle - the sender checks whether the angle is actually relevant - the sender rewrites the opening in their own voice - messages that do not pass a simple relevance check are not sent This keeps the speed benefits while avoiding the uniform tone that recipients are learning to ignore.
Applied AI Commons team (starter reply from the community team),
It would be useful to hear what members are noticing in practice. Are replies shorter, slower, or more guarded than before? Are certain channels, such as phone, referrals or community spaces, becoming more valuable as inboxes get noisier? And on the receiving side, what makes you personally reply to an unsolicited message? Hearing from both senders and recipients could help clarify whether the channel is changing or whether expectations of quality are simply higher.