Summary
Alliance’s prediction for the future is that AI will become so abundant in cold outreach that those methods won’t work in the future. We also think a lot of AI-enabled cold outreach is missing the point: personal sounding emails are great until they all sound personal. So in order to understand how to truly use AI effectively in sales we asked ourselves a basic question: what can AI never do? Answer: human interactions; the handshakes; the building of the network.
Our mission is simple: build tools to enable referrals & better connections with clients. Here are four quick summaries of different ways we’re thinking about solving this issue:
Referral Finder – LinkedIn is widely used for networking, but most connections on the platform are not truly personal. We believe we’ve identified a way to use AI and key data points to uncover not just in-network contacts but genuine, close relationships. For example, if you have a great client, our approach would identify that they personally know another ideal prospect, allowing you to request a warm introduction through a trusted relationship.
ClientDepot – Advancements in AI now make it possible to quickly analyze client interactions across emails, calls, and meetings. This could help businesses easily answer questions like, “Which clients showed interest in this feature?” or but even personal insights such as being notified ahead of a meeting that they just got back from vacation – ensuring key insights are never lost.
We’re currently blueprinting this project out and would like to ask a few questions & hear your feedback on the below. If you’re in sales, please consider scheduling a quick 15 min call with our team.
Referral Finder – How AI should be used in sales
Real networking happens through real introductions – not through random connection requests or cold outreach. We believe AI should enhance human relationships, not replace them. While platforms like LinkedIn give access to massive professional networks, they fail at surfacing the trust-based connections that actually open doors. The real value isn’t just knowing who you’re connected to but understanding which of those connections are strong enough to get you a real introduction. By combining external data sources with LinkedIn, we can map relationships in a way that makes networking more intentional—helping users ask for introductions that are both meaningful and natural.
For this to work for you, we’d start by identifying your Ideal Client Profile (ICP) – the exact type of prospect you want to reach. Once that’s locked in, the next step is mapping your network by using the LinkedIn profile URL of someone you already have a strong relationship with. From there, AI would analyze that person’s connections and external data sources to surface the best possible introductions. The key isn’t just finding mutual connections—it’s identifying which of those connections are both a great fit and a realistic introduction that your contact would actually feel comfortable making. Instead of sending cold outreach or blindly asking for referrals, you’d see exactly who in your extended network aligns with your target market, allowing you to reach out with confidence and precision. This shifts networking from a game of chance to a strategic process—one that strengthens your relationships rather than exhausting them.
There are a few ways this could be implemented, but we don’t know yet what would actually be the most useful for you. Maybe it’s a tool that sits on top of LinkedIn and flags strong referral opportunities directly in your feed. Or it could be something that integrates with your CRM, surfacing warm introductions based on your existing relationships. Another approach could be an email or message generator that helps you craft the perfect ask when reaching out for an introduction. We have ideas, but we don’t want to assume – we want to hear from you. How do you currently handle referrals? Where does it work well, and where does it break down? Your input will help us figure out whether this is worth building and, if so, how to make it something you’d actually want to use.
ClientDepot – Maybe it’s a good thing your company has 7 years of email backups
Sales is more than just closing deals—it’s about maintaining relationships, understanding client needs, and knowing when to engage. But with so many conversations happening across emails, calls, and meetings, it’s easy for key details to get lost. AI now makes it possible to analyze these interactions at scale, helping sales teams stay on top of important client insights without relying on scattered notes or memory. Instead of just logging data, ClientDepot would act as a real-time assistant—surfacing the right information at the right time so you always have the context you need.
Here’s how this could work, from simple use cases to more advanced insights:
Pre-meeting reminders – Before a call, get a quick summary of past interactions, including key topics discussed and any follow-up items.
Client history – If an AE or key team member leaves, ensure their client interactions, notes, and insights don’t disappear – providing a seamless transition for the next person handling the relationship..
Marketing & sales alignment – Identify patterns in client conversations to determine which questions paying customers ask that prospects who don’t convert never bring up—helping refine messaging and objection handling.This may seem bizarre that AI has made this possible but we assure you this is
We’re exploring different ways to build this, but we don’t want to assume what’s most valuable to you. Would it be most useful as an inbox assistant, a CRM integration, or a search tool for past interactions? We’d love to hear how you currently track client conversations, where gaps exist, and what would make your process more effective. Your input will help shape whether and how we move forward with this.
Conclusion
Right now, AI in sales is mostly being used to automate cold outreach, crank out “personalized” emails, and flood inboxes with messages that all start sounding the same. We think that’s missing the point. AI shouldn’t be about sending more—it should be about making every interaction better. Whether it’s ReferralFinder helping you uncover warm introductions or ClientDepot making sure you never lose track of key client details, the goal is simple: use AI to support real, human connections, not automate them away. Instead of sending cold emails into the void or scrambling to remember what a client mentioned in your last call, you’d have the right information at the right time to make every interaction more natural and impactful.
But we don’t want to assume we’ve got it all figured out—that’s where you come in. If referrals are part of your business or you’ve ever wished you had a better way to track client conversations, we’d love to hear how you handle it today. What works? What’s a pain? Your input will help us make sure we’re building something that actually solves a real problem, not just another tool that sounds good on paper.
Calendly link to book 15 minutes with us here