Finding contacts and developing relationships are increasingly essential parts of the sales process since buying groups are expanding. Gartner researchers found that the average size of a B2B buying group is now six to 10 stakeholders. Researching and identifying potential buyers, building buying groups, initiating cold contact with strangers, asking for referrals… These are just a few of the many steps sellers take during the prospecting phase of a deal.
Unfortunately, your ability as a seller to find new contacts is often limited to who you know and who your immediate network knows. It can take hours of research to find and connect with a new contact. And once you’ve found that new contact, it can be difficult to know if they’re the right person to help move a deal forward.
Fortunately, AI for sales teams is ushering in a new era of precision prospecting. If the old way of prospecting is casting an increasingly wider net and inspecting every single contact that you come across, data-driven precision prospecting uses a laser. AI can help sellers identify exactly the right person to help close a deal at the right moment - even people outside your direct network. And real-time, data-driven insights give sellers deep context on the deal, contact, company, and more - everything you need to develop meaningful and valuable relationships.
Successful AI-Powered Prospecting Starts with Great Data
AI is most useful when it is used to speed up manual and time-consuming tasks. It has the ability to process data and make connections in seconds that would take a person hours or days to do. Prospecting is an ideal use case for AI-powered automation since 40% of salespeople say that finding the right buyers is one of the most challenging parts of the sales process.
But applying AI to improve a sales process is easier said than done (and if a vendor is telling you differently, run). The biggest and most important differentiator in any AI platform is access to comprehensive and accurate data relevant to your business and use-cases. It’s the (other) golden rule: the better the data, the better the AI outputs.
The first successful AI workflows were built for consumer businesses with data from hundreds of thousands, if not millions of customers. For example, the AI that powers your custom Facebook feed was trained on all interactions of all Facebook users, ever.
AI tools for B2B enterprises rely on a much smaller data pool since enterprises prefer to keep their data segregated from each other. As a result, B2B AI applications only recently hit mainstream because their training dataset was always too small. This is where the ability to combine datasets across customers (who opt-in, very similar to what an individual does when they sign up for a Facebook account) becomes useful. Suddenly, B2B AI apps can be trained on much larger B2B datasets.
When a GTM AI platform has access to data about millions of contacts, and their history of purchase behavior across thousands of prospect accounts, it can use all that aggregated data to develop lookalike prospects, aka “buyer benchmarks”. Those buyer benchmarks can help answer questions like:
- Am I actually talking to the right people?
- Will I close this deal with the relationships I have and people I’m talking to?
- How much buying power does this person really have?
- How much can they influence the success of this deal?
This is when AI for sales prospecting gets really valuable, really quickly.
Getting the Complete Story with Buyer Benchmarks
Any seasoned seller knows that persona-level data like a person’s name, company, and email address doesn’t tell the whole story. It’s important to combine those three things with additional activity-related context about them as a buyer. What size deals were people like this person involved in? How involved were they and how much influence did they have on getting the deal closed?
A high-quality AI platform will help organizations collect all of their sales activity data over years and years of deals and then use AI to analyze all that data to develop buyer benchmarks. When you combine persona and activity data with the network effect of that person’s interactions across organizations, then you can infer if someone actually has the influence to get a deal closed.
Here are two examples of insights we share with our customers that are far more valuable than surface level information like name and email address.
- Inferred win rate. Based on past deals this person was engaged in, what influence do they have on the success of the deal?"
- Predicted buying power. This is the predicted range that the person can approve.
We also provide insights into the insights so sellers know how to use the information. For instance, someone can have a high inferred win rate but a low buying power and that’s not an ideal combination. This is how you move from casting that wide net to honing in on the right prospects and cultivating the right relationships to close deals.
Data Security and Privacy: Getting Insights Without Personal Information
As powerful as AI can be with access to the right data, it’s equally important to make sure you’re protecting your company data and your customer’s data. A good AI company will be transparent about how they’re protecting all that data to make sure it’s A. not shared with another company and B. not used to train any public AI models.
Here’s how we’re able to help customers prospect with precision using buyer benchmarks without compromising data security. Our always-growing network of cross-customer data currently contains over 170 million contacts and 2.5 billion unique sales activities. And absolutely none of that data is shared with other customers.
Instead, we’ve designed a “lookalike” system so a user can benefit from other company’s data without learning any personal information. Similar to Netflix, who captures data on your viewing preferences and recommends shows similar to those you've liked, we capture and combine millions of points of data from millions of contacts in our database to create lookalike personas that “look” like a person or people in a specific role, company, and industry. We call these Buyer Benchmarks. We are able to do this in an aggregated and anonymized way to ensure we meet the Privacy standards of enterprises. That is why ranges are used vs specific numbers.
If your buyer is VP level at a telecommunications company, we’ll look at hundreds of other VPs at similar type companies to see what kinds of deals these people are typically involved in and how they behave. To generate inferred buying power insights, for instance, we take win rates of all the people who were active in opportunities like yours and meaningfully aggregate them and distribute them as high, medium, low. We combine that with what we know about that specific person. That way, we can say with confidence that VP Mrs. Smith at Telecommunications Company X should have high-level buying power for this specific deal.
Not All AI for Sales is Created Equal
There are many different AI for sales vendors out there who will say they can provide these kinds of insights. Most of them simply can’t because they haven’t spent years and years building a data foundation. It’s this huge, cross-customer database combined with secure, anonymizing AI tactics that allow for the types of reality-based insights that sellers need to work more effectively and close more deals.
Learn how to select the right AI vendor for your business using our comprehensive guide and free RFP template download.