Telemarketing and B2B sales
Sales reps call with personalised call briefs, not guesswork
Danish telemarketing company
Format: One-day hackathon
In short
- Personalised call briefs are delivered automatically 30 minutes before every call
- Publicly available data is collected and enriched with AI, with no manual research
- Three times as many qualified calls per sales rep per day
- Consistent quality across new and experienced sales reps
During a one-day hackathon the team built an automated lead pipeline that gathers publicly available sources, enriches each lead with AI-generated insight on the industry, recent news and likely pain points, and combines it all into a personalised call brief that the rep receives 30 minutes before the call. The brief contains conversation openers based on real context, not guesswork. The reps now make three times as many qualified calls a day, not because they work faster, but because they spend their time talking instead of preparing.
Results in numbers
The challenge
A Danish telemarketing company with 25 sales reps spent an average of an hour and a half per person every day preparing manually for the companies they were about to call. The reps opened the Danish business register, went through LinkedIn, read press releases and tried to piece together a useful picture before each conversation. The result was inconsistent. Some calls opened with relevant context, others with a generic script. Conversion on cold calls was below the industry average, and the reps themselves felt they were spending their time on the wrong work.
The solution
During a one-day hackathon the team built an automated lead pipeline that gathers publicly available sources, enriches each lead with AI-generated insight on the industry, recent news and likely pain points, and combines it all into a personalised call brief that the rep receives 30 minutes before the call. The brief contains conversation openers based on real context, not guesswork. The reps now make three times as many qualified calls a day, not because they work faster, but because they spend their time talking instead of preparing.
How we built it
- 1
Mapping the preparation work
What does the best rep do in 90 seconds? What does a new rep do in 15 minutes? We modelled the ideal preparation and prioritised the data points that actually changed a conversation.
- 2
Collection and source structure
Built in n8n with modular collection from the Danish business register APIs, the companies' own websites and public LinkedIn company profiles. Every source is logged with a timestamp so it can be traced.
- 3
AI enrichment into insight
Claude reads the combined data and generates industry insight, recent news and hypotheses about pain points related to the client's offer. The content is structured so the rep can skim it in 60 seconds.
- 4
The call brief template
A tightly structured format with a conversation opener, three angles for the conversation and likely objections with suggested answers. The template can be adjusted per campaign without touching the workflow itself.
- 5
Delivery in the rep's calendar
The brief arrives as a calendar attachment 30 minutes before the call, so the rep can read up without opening a separate system.
What we learned
- →Never collect data in a way that breaks a website's terms of use. We stuck to publicly available information and still got 80 percent of the value.
- →AI-generated insight must be checked before it is used. The best reps catch mistakes and keep improving the prompts.
- →The format of the brief matters more than the amount of data. Reps have 90 seconds, not five minutes.
- →Live off the reps' feedback. We release a new version every 14 days based on what they catch and what they miss.
“Our sales reps now go into calls better prepared than ever. They spend the time they used to spend on preparation talking to more people. It is the best productivity improvement I have seen.”