Sales role
AI Pipeline Analyst
An AI pipeline analyst that reads your CRM export, finds the deals the forecast is fooling itself about, and gives a range, not a guess.
Give it your pipeline — a CRM export, a Google Sheet or the CRM itself — and it does the revenue-operations work most teams skip: coverage against the number that is left, stage-by-stage conversion, how long deals really sit in each stage, and a forecast split into commit, best case and upside. It segments before it concludes, flags the data it does not trust, and ends every review with the deals that need someone today.
Things you could ask for
Written the way you would actually say them. The agent plans the steps itself.
- “Here is this quarter’s pipeline export in Google Sheets. Calculate coverage against the $600,000 we still need, then the same figure with every deal untouched for 30 days taken out.”
- “Build a stage conversion table from last year’s closed deals: how many entered each stage, how many moved on, and the median days spent there.”
- “List every deal that has sat in its current stage for more than one and a half times the median, with the last activity date and what you would do about each.”
- “Give me a forecast for the quarter as commit, best case and upside, with the assumptions behind each — and tell me which numbers you would not trust because the CRM fields are empty.”
- “Every Monday at 8am, compare this week’s pipeline with last week’s and send me what moved, what stalled and what is new.”
What you get back
A pipeline health report: velocity, coverage and quality-adjusted coverage, a stage conversion funnel, the deals requiring intervention with a recommended action for each, and a forecast in commit, best case and upside with the assumptions stated. Delivered as a new tab in your Google Sheet with the formulas left in, a summary in chat or Google Slides, or as a weekly report on a schedule.
Tools it works with
A role is a job description; connectors are what let it do the job on your real work instead of on whatever you paste into a chat.
Where it is the wrong tool
- A remote agent has no shell, so it does not run a statistics package over your data. It works through the numbers and writes them into the sheet as formulas you can check; for a large dataset, ask for the formulas rather than trusting a figure quoted in chat.
- The forecast is only as good as the CRM. Missing close dates, stale stages and empty fields are flagged as gaps, not filled in — and an unreliable pipeline gets a wide range, not a confident number.
- It has no Salesforce or HubSpot analytics of its own. It reads records, exports and sheets; dashboards and forecasting features inside your CRM stay where they are.
- It measures the pipeline, not the people. Rep-level win rates are a reason to look closer, not a performance review; small samples get called out as small.
Starting an agent with this role
- 1Create a Burrak account — any plan works, and your first week is $1.
- 2In the Marketplace, open the Roles tab, find Pipeline Analyst and choose “Start an agent as Pipeline Analyst”. That creates a remote agent with the role’s instructions in its system prompt.
- 3Connect the tools it needs — Google Sheets, Attio, Airtable, Google Slides, Notion — from Connectors.
- 4Give it a brief. It keeps working in the cloud after you close the tab, and you are only charged while it is actually working.
The Pipeline Analyst role, in practice
- Which CRMs does it work with?
- Attio connects directly, and HubSpot and Salesforce connect through the same Connectors page. With any other CRM, export the pipeline to a Google Sheet or CSV and give it the file — the analysis is the same.
- Why a range instead of a single forecast number?
- Because a single number hides how sure you are. Commit, best case and upside, each with its assumptions, tell the people reading it which deals the number depends on — and that is the conversation a forecast call should be having.
- Can it run the pipeline review every week without me?
- Yes. Set it as a scheduled task and it reads the pipeline, compares it with last week and sends you the changes and the deals at risk. It runs whether or not your computer is on, and only reads unless your brief tells it to write.
Capability reference
Summarised from the role’s instructions, which are adapted from the agency-agents collection (AgentLand Contributors), used under the MIT licence.
- Pipeline velocity analysis across opportunities, deal size, win rate and cycle length
- Coverage and quality-adjusted coverage against remaining quota
- Deal health scoring on qualification depth, engagement and progression
- Layered forecasting reported as commit, best case and upside
- Explicit flags for stale pipeline and data quality gaps
Your AI assistant is waiting.
Put it to work today.
Put an agent on the work that never needed a human in the first place — the research, the reports, the follow-ups — and take the week back.
$1 for your first week · Cancel anytime · Works on Mac, Windows, iPhone, Android & Web