21 Days of AI for Sales Teams

The Concept

Generic outreach is not just ineffective. It actively damages your reputation with the people you most want to reach.

When a VP of Sales receives your email and nothing in it reflects any knowledge of their company, their current situation, or why your message is relevant to them right now, they do not just ignore it — they form a negative impression of you and your company. You have signalled that you are willing to waste their time. That impression persists. It makes a future conversation harder, not just this one.

The problem has never been that personalised outreach does not work. It is that doing it properly — reading the company website, scanning their LinkedIn, checking recent news, looking at job postings for clues about priorities — takes 20 to 40 minutes per account. At that rate, a rep with 50 target accounts would spend two full working days just on research before writing a single word of outreach. So most reps skip it, send something generic, and wonder why their reply rates are 2%.

AI changes that equation. Not by inventing information about your prospect — that is how you get caught in a lie before you have even started. But by taking everything you have found and organising it into a coherent picture of what this company is probably focused on, what pressures they are likely under, and where your offering has the best chance of being relevant.

The difference between a researched angle and a generic value prop

A generic value prop sounds like this: "We help sales teams close more deals faster with our AI-powered platform." Every rep at every competitor says a version of this. The prospect has read it hundreds of times. It creates no reaction because it is about you, not about them.

A researched angle sounds like this: "I noticed you have posted four new SDR roles in the past six weeks, all requiring experience with outbound sequencing tools. That kind of ramp usually creates a training gap between what new reps need to know and what they can actually execute on day one. That is the exact problem our platform was built to close." That message contains a specific observation, connects it to a plausible pressure, and ties it to an outcome. The prospect reads it and thinks: this person has actually looked at what we are doing.

The difference is not creativity. It is specificity. And specificity comes from research.

What AI is actually doing when you give it account information

When you paste a company's website, LinkedIn content, and recent news into a prompt and ask it to extract priorities and pain points, the model is doing something genuinely useful: it is reading across everything you have given it and looking for patterns, signals, and implied pressures that a human skimming the same material might miss.

A hiring page buried in a careers section might show twelve open engineering roles and two open sales roles — which suggests the company is in a product-build phase, not a go-to-market push, which has implications for how you frame your outreach. A CEO's LinkedIn post from three weeks ago might mention entering a new market, which is a trigger for infrastructure, compliance, or workflow tools. A press release about a funding round might specify that the capital is earmarked for international expansion — a detail that is easy to overlook when you are scanning quickly.

The AI surfaces these signals. Your job is to decide which ones are actually relevant to your product and which outreach angle to lead with. That judgment still requires a human. But the research that informs that judgment no longer has to consume 30 minutes per account.

Why this approach compounds over time

There is a practical benefit that goes beyond individual emails. When you build the habit of running this research process before every first touch, you start accumulating a body of knowledge about your target market. You begin to see which triggers consistently signal good-fit prospects. You learn which pain points resonate by tracking which angles actually generate replies. Over a quarter, you are not just doing better outreach — you are building a proprietary model of your market based on real evidence.

That is not something a generic email sequence can give you.

You are a senior B2B sales researcher helping me prepare outreach to a target account. Here is the information I have gathered about them: [PASTE WEBSITE COPY, LINKEDIN ABOUT SECTION, RECENT NEWS, JOB POSTINGS, OR ANY OTHER PUBLIC INFORMATION YOU HAVE FOUND].
My product or service is: [DESCRIBE WHAT YOU SELL IN 2–3 SENTENCES — what it does, who it helps, and the core outcome it delivers].
Based on the information I have provided, please do the following: 1. Identify 3–5 likely business priorities this company is focused on right now, based on what is visible in their public presence 2. List 2–3 potential pain points or pressures that a company in their position typically faces — and flag which of these are supported by evidence from what I pasted vs which are reasoned assumptions 3. Identify any growth triggers visible in the information (e.g. hiring patterns, new product launches, expansion signals, leadership changes, funding announcements) 4. Generate 3 distinct personalised outreach angles — each one connecting a specific thing I noticed about their business to a specific outcome my product or service helps create 5. For each angle, rate its strength on a scale of 1–3 and explain why in one sentence
Label assumptions clearly. Do not invent facts. If the information I provided is thin, tell me what additional research would strengthen the output.