LEADSPICKER - the AI sales prospecting machine

LEADSPICKER - the AI sales prospecting machine

TLDR: Prospecting for sales leads is a slog and nobody likes doing it. It burns a ton of sales costs, is time consuming and highly repetitive. Sales resources are tied up on mundane tasks when they could be engaging with customers. So, imagine a world where lead generation is automated. And meetings with potential clients are delivered straight into your inbox.  Meet Leadspicker, the “personal prospecting assistant” that exploits web-scraping, machine learning and business process automation to do just that!

The way it used to be done

Most of my corporate career was spent in sales roles. When I started out, which was sometime in the last century, I was selling into the B2B tech space. As a rookie sales rep in the 80’s, a large part of my time was spent on lead generation. For me, that meant sitting down with the Computer Users Year Book, trawling through the listings, looking for suitable candidates, then calling them up. That was a time before the Internet and the only source of information was the year book.

Nowadays, there is a wealth of information to support lead generation, much of it free to access, at the click of a button. However, in spite of all this additional information, the fundamental process is still the same. You take a long list of names, filter it down to a targeted list of suspects, then reduce it further to a qualified list of prospects. It is still a largely manual exercise, albeit with significantly more data at your fingertips.

Add the pandemic into the mix and the need for a digital solution to prospecting has never been more in demand. With no physical conferences or exhibitions in place, and the opportunity for face time reduced to a zoom call from your living room, sales organisations want a digital solution to keep their pipelines full and sales people busy. A recent study by data analyst Exasol found that 87% of US retailers are under pressure to make data driven decisions faster! And according to Hubspot’s Not Another State of Marketing Report, 61% of marketers say that generating leads is their biggest challenge.

So, I went in search of a 21st century approach to prospecting and lead generation and found Leadspicker.

Man and machine working together

Based in the Czech Republic, Leadspicker have built an automated lead generation and sales qualification operation. Using a combination of web scraping, machine learning and business process automation, Leadspicker take the leg work out of lead generation. So that humans working in business can focus on, well, business.

To find out more, I set up a zoom with Vlastimil Vodicka, the CEO and co-founder of Leadspicker. As always, I started out by asking him what problem they were trying to solve. “The thing about lead generation is that it is still manual for most organisations. It is also a set of repetitive tasks, that every does over and over again. The beauty of it is that we are in a sector (sales automation) where everyone has to do it (lead generation) and, if you do it right, it has the fastest impact on a company’s revenues and performance.”

Logo LP Sales Booster.png

Nobody likes doing it but someone has too!

I see this a lot when I coach sales managers and founders for SalesClub. A regular topic of conversation is about how they spend their sales budgets to get the best bang for their buck. There is no simple or straightforward answer to the question, but a good place to start is to look at the efficacy of their lead gen and prospecting efforts.

Time is as important as money here, because the opportunity cost of not spending time wisely will inevitably cost them more than all the hard cash they take spend.

Vlastimil sees this same problem too with his clients. He explained, “my clients are always telling tell me how they spent a year of their life in excel spreadsheets doing lead generation, and they’re still doing the same thing again and again.

“Which was my own experience and how I came to form Leadspicker. I was working in Vienna and doing lead gen for an investment fund. I realised that 70% of my daily task was repetitive and mundane, cutting, pasting and searching, moving between excel and google. I was spending a lot of time doing the same thing over and again, with the only difference being the key words.”

It was during this time that Vlastimil explained all of this to an old tech friend of his from his school days. Vlastimil was curious to know if it was possible to code these processes, instead of doing them all manually. His friend introduced Vlastimil to machine learning and, using examples of what he was working with, his friend coded machine learning algorithms to replicate the human process he was doing.

VB Insights

It sounds simple enough, but there is still a need for human oversight of the automated process. Vlastimil explained, “even though we now automate the whole process, we still have a human quality control process on the output.

“For example, let’s say that our client is looking into the energy sector. Sounds simple enough, but if you just download a list of companies that have something to do with energy, you might also get anyone connected with energy drinks. You still have to add a human layer over the top to check the output.”

The solution is only as good as the data

Any solution can only be as good as the data that feed it, right? So, I asked Vlastimil where he got the data from. “The first thing we do is to ask each of our clients how they doing it (lead gen) now? And the answer is usually the same. They mainly use search on Google to find prospects. Or they look at publicly available sources. Or they use LinkedIn.

“For example, our client has a sales territory of all the restaurants in Germany. Typically, what they will do is a google map search for all restaurants in Germany. They then copy and paste the search results into a spreadsheet. They then manually work through the list of names and addresses. This is their source data. It’s a long-winded way of doing it. The secret is to enhance the data to turn into something meaningful.

“We had a client that was manufacturing solar panels for commercial property. They wanted to find firms with buildings with a roof over 100m2 in size. This was very specific and we enriched the data with pictures from google maps. We then used our machine learning algorithms to process the enriched data and identify if they already had solar panels on the roof.”

The solution at Leadspicker is to take the traditional approach to prospecting,  automate it and then enhance it by overlaying other data sources and enriching the list of prospects. They can do this because there is a ton of freely and commercially available 3rd party data that Leadspicker can access.

Using GPT-3 from openAI

To make a sale, you need more than just a list of “suspects”. The list needs to be qualified and converted into “prospects”. This requires insight on a prospect hat is more layered than you get from simply doing a google search or looking at a directory of companies. Timing (for making an approach) is the most important part of this insight.

Vlastimil explained, “What we do is to monitor both the companies and the sector they are in. We track all media and look for hooks and triggers. In the manual way, the right approach is to investigate a suspect and look for news about them that you can use as a hook or a trigger for when you approach them. But when you do that manually, it is a slow and time consuming process.

“What we do is automate the process to give us scale. We have a system that uses the GPT-3 API from openAI. Leadspicker were one of the first of companies to have beta access to the GPT3 natural language processing technology to auto read text, identify hooks and triggers and auto write the summaries.”

17966.jpeg

Hooks and Triggers

The point is that Leadspicker are creating insights that you don’t get from (just) search engine results. Leadspicker uses insights to create a personalised message and improve the changces of getting a positive response. This is what they mean by hooks and triggers.

Once they have an enriched list of prospects, Leadspicker initiate the outreach process, making the first approach on behalf of their clients. They do this using their client’s email address, which means that any interest the prospect has will go directly back to the client’s inbox.

The net result is that the client’s sales team receive qualified leads direct into their inbox without having to spend their time and effort to create them. This is sweet music to my sales manager’s ears. Because when I was leading teams of sales people, I didn’t want to see them with their heads in a spreadsheet, I wanted to hear them engaging with customers.

What next for Leadspicker?

The tech firm is five years old now, and has around 350 customers operating across 40 countries. Having bootstrapped the firm, Leadspicker raised €2m last year in their first seed round. This is funding their own sales growth as they expand into new territories and increase the number of languages they can operate in.

Longer term, it seems obvious to me that Leadspicker are a perfect complement to a larger CRM or sales automation platform. From my experience, many of them promise but don’t deliver when it comes to generating meaningful customer insights and lead generation. Leadspicker could very well solve that for them!

Svott case study with Leadspicker Sales Booster

Analysis of Exasol report on data driven decision making for US retailers from Inside Big Data

What is GPT-3, the new openAI language model, blog by Ryan Taylor

30+ Lead Generation Stats, a 2021 blog from FunnelOverload

56 Sales podcasts every sales professional should follow, by Sales Hacker

Great! You’ve successfully signed up.
Welcome back! You've successfully signed in.
You've successfully subscribed to The Wiser! Newsletter.
Your link has expired.
Success! Check your email for magic link to sign-in.
Success! Your billing info has been updated.
Your billing was not updated.