What an account brief should cover before the meeting

A good account brief fits on one page and answers one question, why this company might need your equipment now. AI helps gather and summarise public information, but you decide the structure. If every sales rep asks for something different, every brief will look different and nobody can compare them.

These are the sections we suggest for a brief on a pharma company, a biotech or a CDMO. Every fact needs the link it comes from, and anything that cannot be verified becomes a question for the meeting.

  • What they make and in which modality, such as monoclonal antibodies, vaccines or cell and gene therapy.
  • Where the pipeline stands and which products are getting close to approval.
  • Facilities, expansions and investments announced in recent months.
  • Job postings that signal growth, such as MSAT, upstream or downstream roles.
  • Who is attending the meeting, with their job title and role in the purchase.
  • Open questions the meeting needs to answer.

Where to get the facts without relying on what the AI says

For a European pharma company, many of the most useful sources are public. The European Medicines Agency publishes a monthly list of medicines under evaluation by its human medicines committee. The list gives the international non-proprietary name of the substance and a short summary of the indication, and it is worth cross-checking against the company's press releases.

The EU clinical trials website has information on individual trials in the EU and the European Economic Area since 31 January 2022. A late-stage trial may mean the company is already thinking about scale-up and manufacturing. Annual reports, press releases about new sites and job postings fill in the rest of the picture.

Which AI tool works well for preparing sales calls?

Any assistant that can search the web and cite the sources behind each answer. For a company, a business plan with clear data protection terms is the safer choice. The template and the prompt matter more than the tool, because they are what prevent invented answers and a different brief for every sales rep.

How to prompt the AI so it does not make up facts

AI assistants can state false facts with great confidence, such as an approval that never happened or a plant in the wrong city. In front of a pharma customer, a mistake like that costs credibility straight away. So your prompt should force the tool to cite a link for every claim and to say clearly when it found nothing.

It helps to split the brief into two columns, sourced facts and the sales rep's hypotheses. Facts are checked by opening the link before the meeting, and anything without a source is removed or turned into a question. Someone on the team should review the brief before it is shared, especially figures and dates.

Want us to look at which sources and which prompt your sales team could use to prepare meetings? Leave your details and we will talk it through.

Prepare your meetings

From the brief to the questions you ask in the meeting

The brief is mainly there to prepare good questions. If one of the company's antibodies reaches late-stage trials, it makes sense to ask how they plan scale-up or tech transfer. If they are hiring upstream specialists, they may be expanding cell culture capacity. These are hypotheses the customer has to confirm.

With a pharma buyer, showing that you understand their context opens the conversation, and reading the brief back to them closes it. Use the facts to ask specific questions and let the customer explain their project, their scale and their timeline.

Can we look up the people who will attend the meeting?

Yes, within limits. Job title, company and public professional profile help you prepare the conversation, and private life has no place in it. If you store that data in the CRM, article 14 of the GDPR generally requires you to inform the person within one month or at first contact, and article 21 lets them object.

GDPR limits when you research people

The GDPR does not cover data about companies as legal persons, as its recital 14 explains, but it does cover data about the people attending the meeting. If you store data in the CRM about someone who did not give it to you, article 14(3) says that, as a rule, you must inform them within a reasonable period and within one month at the latest, or at the first communication if you use it to contact them.

In practice, stick to job title, company and public professional profile, and leave private life out. If you use the data for direct marketing, article 21 gives the person the right to object at any time, and you must then stop using it for that purpose. Before pasting personal data into an AI tool, check what the provider does with it.

How do we know whether a pipeline fact is correct?

Open the link that comes with it and check it against a primary source, such as the European Medicines Agency's list of medicines under evaluation, the EU clinical trials website or the company's own press releases. If there is no link, or the source says something different, the fact does not go into the brief.

How to roll it out across the sales team

Start with a brief template and a fixed prompt for the AI assistant that the whole team uses the same way. Store each brief in the CRM under the account, so the next person who visits does not start from scratch. After the meeting, the sales rep adds what was confirmed and what was ruled out.

After a few meetings, compare the briefs with what actually happened. If the hypotheses were far from reality, adjust the sources or the prompt. When we set this up with a sales team, we start from the meetings already in their calendar and the sources they already use.

Add crida studio as a preferred source on Google