I think we’ve got this whole AI thing backwards.
I’ve been sitting in on a few discovery calls lately, mostly as a fly on the wall, and I have noticed something slightly alarming. We are currently living through the Great Generic Peak. Every agency owner is using the same three prompts to generate the same five "personalised" outreach emails, which are then being sent to the same fifty thousand weary prospects.
The result? The "delete" button has never seen this much action.
Between us, most of the people selling AI-powered sales strategies are just selling a faster way to be annoying. They are automating the noise rather than the signal.
The real money right now isn't in sending a thousand emails. It is in using AI to find the one reason why you should only send ten.
How to build an effective AI-powered sales strategy
If you want to actually win business as a consultant or agency, you have to stop thinking about AI as a megaphone. Think of it as a private investigator instead.
The smarter move right now is to use these tools to build a "context engine" rather than a content engine. Most people use LLMs to write the message. You should use them to justify the message.
Here is the secret sauce that the high-performing agencies are currently using to get ahead:
- Deep Intent Mapping: Instead of scraping a list of "CEOs in London," use AI to monitor niche signals. Look for companies moving offices, hiring for specific roles, or mentioning specific pain points in their quarterly reports.
- The "Gap" Analysis: Take a prospect’s current website or LinkedIn activity and ask an LLM to identify exactly where they are lagging behind their competitors in terms of technical implementation.
- Human-in-the-loop Synthesis: Use AI to draft the research, but spend five minutes writing the first two sentences yourself. That "uncanny valley" feeling of AI text is a deal-killer.
If you can prove you’ve done the homework, the prospect will forgive the fact that you used a tool to help you grade it. If you want to dive deeper into how this looks in practice, you can find more articles on AI here.
Improving B2B lead generation with AI without the spam
The biggest mistake I see is the "spray and pray" 2.0. People think that because the cost of sending an email is near zero, the volume should be near infinite.
Your prospects are smarter than that. They can smell a ChatGPT-generated compliment from three tabs away. "I was impressed by your recent post about vertical integration" is the new "I hope this email finds you well." It’s a red flag.
To improve your lead generation, you need to use AI to find "Triggers" rather than just "Targets."
A trigger is a specific event. A target is just a person with a job title. If you feed an AI your successful case studies and ask it to find companies currently experiencing the exact problems you just solved, your conversion rates will climb. It’s not about finding more leads; it’s about finding better reasons to talk to the leads you have.
Streamlining the sales process for AI agencies
Let's talk about what happens after the lead says hello. This is where most consultants drop the ball. They spend all their time on top-of-funnel automation and then revert to 1998 for the actual sales process.
You should be using AI to shorten the distance between "Hello" and "Deposit paid."
- Automated Briefing Docs: Immediately after a call, use a tool to summarise the transcript not just into notes, but into a draft Statement of Work.
- Objection Handling Databases: Build a local RAG (Retrieval-Augmented Generation) system of all your previous successful pitches. When a client asks a tricky question about ROI, you shouldn't be starting from scratch.
- Dynamic Pricing Models: Use AI to look at the scope of a project and compare it against your past data to predict how many hours it will actually take, not just what you hope it takes.
If you're finding that your own internal systems are a bit of a mess while you're trying to sell "efficiency" to others, it might be time to book a consultation so we can look at the plumbing.
Actionable lead generation techniques for consultants
If you want to change your results by Monday, try this specific workflow. It’s what I’ve been seeing work for those few people who aren't currently screaming into the void.
- Step 1: Find a company you really want to work with.
- Step 2: Feed their last three months of public output (blogs, social, news) into an LLM.
- Step 3: Ask the AI: "Based on this, what is the one thing they are clearly struggling with that they aren't admitting?"
- Step 4: Write a short, manual email mentioning that one thing.
- Step 5: Use AI to generate a 1-page "Roadmap to Solved" specifically for them.
This isn't about being "cutting-edge" (god, I hate that phrase). It’s about being useful. Most of your competitors are sending generic AI junk. By using AI to be highly specific and human, you stand out like a lighthouse in a fog of bot-shit.
Closing the gap between AI hype and sales reality
We’ve all seen the LinkedIn posts. Someone claims they’ve automated their entire agency and now they spend their days drinking coconut water on a beach while the bots print money.
In reality, they are usually one API update away from total collapse.
The "between us" truth is that AI isn't going to replace the salesperson. It is going to replace the boring parts of sales. If you are still spending four hours a day manually cleaning spreadsheets or writing "just checking in" emails, you are doing it wrong.
The goal is to use technology to buy back your time so you can actually talk to people. People still buy from people they like and trust. They don't buy from the smartest algorithm in the room; they buy from the person who used the algorithm to understand their problem better than anyone else.
If you're curious about how to put this into practice without needing a PhD in prompt engineering, it’s worth thinking about the tools you use every day. For those looking to get ahead of the curve, you should definitely get the free book which covers the fundamentals of making this stuff actually work in a business environment.
Let’s stop trying to out-spam each other. The first person to use AI to become more human is the one who’s going to win.
If you're looking for a smarter way to manage your sales pipeline, check out SalesM8 — it's the tool we built to make AI-powered selling actually practical.
About the Author
Steven Tann is an AI consultant, author of "You're Selling AI Wrong", and founder of SalesM8. He writes about AI, sales, and running a business from a narrowboat on the English canals. Connect with him at steventann.com.
