AI account research: a sourced brief on every target account
An agent runs deep research on each target account with Parallel and Exa, and hands you a one-page brief with priorities, news, people and a reason to reach out, every claim with its source.
- 7 days free
- No card required
- Nothing sends without your yes
One request. The agent does the rest.
Type it to Penny, or use it as the agent’s brief and let it run on its trigger.
Agent PennyRun with Custom agentFor each company on “Q1 target accounts”, research the last six months: priorities, news, leadership changes and hiring. Give me a one-page brief per account with sources and one reason to reach out now.
- Reads the accounts on your list
- Starts a deep research task for each one
- Searches recent news, company pages and announcements
- Pulls out priorities, changes and a reason to reach out
- Writes one brief per account, every claim with its source
- Accounts researched
- 20
- Briefs written
- 20
- Sources cited
- 140
- Reason to reach out
- One per account
For a high-value account, the first message has to show you did your homework. That homework used to take an afternoon per company: reading the website, the press page, the CEO’s posts, the job board and a few articles, then writing it up so the team could use it. An account research agent does that work for every account on your list and hands you a one-page brief, with every claim linked to where it came from.
What goes into a good brief
A useful account brief answers four questions. What does this company care about right now? What changed recently: funding, leadership, launches, expansion, hiring? Who are the people that matter for your deal? And what is the honest reason to reach out this month? The agent structures every brief around those questions, so your team can compare accounts at a glance and prioritise.
What the agent does
It reads the accounts on the list you name. For each company it starts a deep research task with Parallel Research, searches recent news and announcements with Exa, and reads the pages that matter with Parallel Search. It pulls out priorities, recent changes, key people and hiring signals, and writes a one-page brief with sources. When it cannot find something, it says so rather than guessing. The result appears in the run report as formatted briefs you can copy, and it can save a short summary to each account’s notes.
Set it up
- Connect the research apps you want to use. Parallel Search and Exa work without signing in; Parallel Research uses your API key.
- Create an agent from scratch with the brief above, and tick the research apps.
- Run it on your target account list, or schedule it monthly to refresh every brief.
- Set a spending cap for each run that fits the size of your list.
Turn research into meetings
- Give the briefs to the Writer, and every opening line can reference the account’s real priority.
- Use the leaders named in each brief to map the buying committee for account-based outreach.
- Save your best briefs to Knowledge, so the Reply Agent can use them when the account answers.
Why sources matter
An AI summary without sources is a liability in sales: one wrong fact in a first message and the conversation is over. Sourced briefs let a rep check the one claim they plan to use in ten seconds, and they make it easy to spot when a company has changed since the brief was written.
Connected in a click, used on your terms
Sign in once. You choose what it may do, and changes wait for your approval.
Parallel ResearchResearch/Web
Run deeper research tasks, then retrieve their progress and sourced results.
Parallel Research + Prosyo
ExaResearch/Web
Search the web and company news to research people before you write.
Exa + Prosyo
Parallel SearchResearch/Web
Search the public web and read pages for company and prospect research.
Parallel Search + Prosyo
Custom agent asks before it acts
You choose how much it does alone. Reading and research always run straight away; anything that sends, spends or changes another app waits as a card you approve, unless you allowed it.
- Ask before every stepIt asks before every step, unless you allowed that action to run on its own.
- Ask before any changeIt reads and researches freely, then brings every change back to you.
- Handle routine work aloneSaving people, tagging and preparing drafts run on their own. Sending, paid actions and other apps still wait.
Use cases that pair well with this one
ScheduledRevive closed-lost deals with a reason to reconnectEvery Monday an agent reads deals lost more than 90 days ago in your CRM, checks what changed at each company, and drafts a reason to reconnect for the ones worth another look.Custom agent
ScheduledFollow up on every demo request and form fill by morningEach morning an agent reads new submissions from Jotform or Typeform, researches each company and drafts a personal reply for every person, ready for your approval.Custom agentOn demandFind podcast and partnership opportunities that fit your audienceThe Research Agent searches the web with Exa and Parallel for podcasts, newsletters and partners that reach your buyers, and saves the hosts and owners with a reason each one fits.Research Agent Which research tools does it use?
The web research apps you tick, such as Parallel Research for deep multi-step research, Exa for company news, and Parallel Search for reading pages. Some work without an account; others use your API key.
How do I know the brief is accurate?
Every claim carries its source, so you can open the page it came from. The agent says when it could not find something instead of filling the gap.
Can it research people as well as companies?
Yes. Ask for the leaders on the account and what they have said in public, and use the result in the Writer’s first lines.
When does it run?
Whenever you like, or when you ask Penny in chat. If it becomes a weekly habit, put it on a schedule instead and it runs on its own.
Which apps does it use, and do I need accounts?
It uses Parallel Research, with your API key stored encrypted; Exa, no account needed; Parallel Search, no account needed. Each app works with your own access, so the agent only sees what your account can see, and anything an app charges for is billed by that app.
Can the agent act without my approval?
Reading and research run straight away. Routine work such as saving people to a list or tagging can run alone if you allow it. Sending, paid actions and changes in other apps wait for your yes unless you explicitly allowed that action.
What does a run cost?
Each step uses AI credits like a Penny question, and you set a spending cap for each run. If a run reaches its cap, it waits for you. Paid actions such as enrichment, and charges from connected apps, are separate.
Can I change what the agent does?
Yes, in plain words. Edit the brief, attach a skill such as your own playbook, tick or untick apps and change what needs approval. The next run follows the new instructions, and finished work stays saved.
Which agent runs it?
An agent you create from scratch with this use case as its brief. It takes a few minutes: name it, paste the request, pick the trigger and apps, and choose what needs approval.
How do I see what the agent did?
Every run comes with a plain-English report of what it did, what it cost and anything waiting for you. Results show as formatted answers, tables or diagrams, and you get a notification when a run finishes or needs your approval.
Which Prosyo plan do I need?
Agents come with the Penny Agents, Penny SDR and Penny SDR All-in plans: 5, 10 or 20 agents. Every workspace starts with a 7-day free trial, no card required.
Your next 10 meetings are already on LinkedIn.
7 days free, no card. Pay later and change your mind? Full refund within 7 days.
- 7 days free
- No card required
- Cancel in two clicks
