Fight AI Outreach With AI (it's not a joke)

I don’t know about you, but these days my LinkedIn DMs are drowning in automated outreach.

It has never been easier to send a message that looks personalized, but on the other end most inboxes are cluttered and people are either ignoring it completely or wasting a lot of time sorting through messages.

This week, we’re using AI to fight back.

My favorite things to automate are those that require lots of tiny decisions to be made (if you have been here a while, you have probably picked up on this). The task switching cost of these types of tasks drives me crazy - so hopefully you find this one helpful in making you day to day less cluttered. 

LinkedIn is pretty specific about how automation can be used on the platform. It doesn’t allow third-party tools that scrape LinkedIn or automate activity like sending messages, so we’re not building a bot that takes over your inbox (unfortunately). But there is still a lot AI can do to help us spend less time managing it.

The Challenge:

Pull your last 20–30 LinkedIn messages and have AI help you figure out a triage and response system. Ask it to group the messages into useful categories like sales outreach, recruiters, requests for advice, networking, etc. You can also ask it to suggest categories based on what it finds.

Then decide how you want to handle each category and draft templated responses for your most common types of requests or get suggestions on longer form pieces you can link people to ie. career advice. 

By the end, you should have a simple system for triaging your inbox and a few responses you can reuse instead of spending a lot of time on this or consistently pushing it to the back of your list.

The prompt to get you started

Drop this into your AI of choice to get started:

Here are my last 20–30 LinkedIn messages: [Paste messages]

Group them into useful categories based on what the sender wants from me. Show me roughly what percentage of my inbox falls into each category. Then help me decide how I should generally handle each type: respond, decline, answer async, take a call or ignore.

For the categories where I’m writing similar responses repeatedly, draft a short reusable response. I want my responses to sound [describe your tone].

Flag any messages that don’t fit the normal rules and are probably worth looking at individually.

Too easy?

Automate the triage. Build a workflow that sorts messages automatically. You might prioritize certain names, companies or keywords. Ie. phrases like “quick demo,” “pick your brain” or “15 minutes of your time” trigger a different path.

Use simple rules for the obvious keywords and AI to analyze the messages that are harder to categorize.

Because LinkedIn restricts third-party tools that automate activity on its website, be careful about how you get messages into your workflow and keep the actual LinkedIn actions in your hands. I’ve included this in the prompt below so you can get suggestions on how to build it without getting yourself banned.

I want to build a system to automatically triage my LinkedIn messages.

Here are the types of messages I usually receive: [describe them].

Here are the people, topics and opportunities I care most about: [describe them].

Help me identify the keywords, phrases, names or other signals that could be used to sort these messages automatically. Create simple rules for what should be prioritized, deprioritized or sent to AI for a closer look.

Then help me turn those rules into a simple workflow I could build with an automation tool that is compliant with LinkedIn’s terms of service. I want the system to do the sorting and drafting of responses so that I only have to spend time on the messages that actually need me.

Why this matters?

AI has made it incredibly cheap to create more outreach. In general, the easier it becomes to produce things, the more crap we will all have to sort through.

Your LinkedIn inbox is a small example of that. Most of the work is sorting through everything to decide what deserves your time in the first place.

The circular nature of using AI to deal with a problem partly created by AI is not lost on me. But if AI is going to make it easier to create more noise, we should at least be using it to help us filter that noise.

That’s it for this week.

Hit reply and tell me what you thought of the challenge this week.

I read every response, and your experience helps me shape these challenges. If you get stuck or have questions, reach me at hello@tinyailab.ai