# Account Discovery

> Describe your ideal customer in plain English. Nimble agents search the live web for companies that fit, check every criterion against a source, and return a list you can export to your CRM.

Interactive demo: https://www.nimbleway.com/demos/account-discovery

## Find every company that fits your ICP

Describe your ideal customer. Nimble agents search the live web for companies that match and check every criterion against a source.

## Examples to try

- "Find Series A fintechs in the US hiring SDRs"
- Series A–B US fintechs hiring their first SDRs, Vertical SaaS for dental clinics in North America, European logistics companies with 200–1,000 employees, AI startups that launched a public API in the last 6 months
- Ramp, Vanta, Attio, Deel

## Set it up in your agent

The Nimble skill for this use case: https://docs.nimbleway.com/skills/account-discovery/SKILL.md

Prompt to paste into your agent:

```text
Be my prospecting assistant using Nimble. Follow https://docs.nimbleway.com/skills/account-discovery/SKILL.md.
```

Get an API key: https://online.nimbleway.com/signup

## The most accurate & efficient web search for GTM data

We asked Opus 5.5 hundreds of GTM questions about companies, people and industry news. It ran once with each web search provider, and every answer was checked against a hand-verified ground truth. We tracked accuracy and the total cost of getting there, including search fees and tokens.

Benchmark code and data: https://github.com/Nimbleway/gtm-enrichment-benchmark

## What you get

A list of companies that match an ideal customer profile (ICP), researched on the live web. Each row has the company name, website, why it fits, employees, HQ and a recent dated signal with its source. Every row is checked against each criterion: met, not confirmed or not met, with the evidence, a source URL and a date where the page gives one.

## Modes

| Mode | You give | Criteria it checks |
|---|---|---|
| Describe your ICP | One sentence, for example "Vertical SaaS for dental clinics in North America" | Category, plus funding stage, company size, region and any extra clause (such as "hiring their first SDRs") pulled out of the sentence |
| Find lookalikes | One company, for example Ramp | Same category as that company, sells to the same buyers |

In both modes you can add must-haves of your own and set filters for region, company size and funding stage. The demo asks for 10 or 20 results and searches three angles in parallel: directories and lists, news and funding, and hiring and product signals.

## Inputs

| Input | Required | Example |
|---|---|---|
| Ideal customer | Yes | Industry or product, geography and size, in one sentence |
| Hard rules | No | For example "raised Series A or later since 2024" or "not already a customer" |

## Run it yourself

The Nimble skill for this use case runs the Nimble agent template `account-discovery`: https://docs.nimbleway.com/skills/account-discovery/SKILL.md

- Create the agent once: `nimble_agents_create(template="account-discovery", agent_name="account-discovery")`
- Run it: `nimble_agents_run(agent_name="account-discovery", input="Find <type of company> in <geography> with <size>, that <hard rules>. Companies like <examples>.", effort="medium")`
- Effort: the skill runs the template at medium effort (about 1-3 minutes), even when the template's own default is different. Ask for low, medium, high or x-high to run another.

Turn each criterion into a clear test: "mid-market" is not a test, "200-2,000 employees" is. For a broad profile (a whole country or a big vertical), split it into slices by region or sub-vertical, one run per slice in parallel, then merge and dedupe by website. Don't ask for a number of rows; the agent returns every company it can verify. Headcount for private companies is often a range, so accept a range. Fewer rows than expected is a signal to run another slice, not a finished list.

## When to use it, and when not to

Use it for "find companies like", "build a prospect list", "ICP list", "find accounts that match", "target account list" or "companies that fit our ICP" requests.

- To research one known company, use Account Brief: https://www.nimbleway.com/demos/account-brief.md
- To find investment targets, use Deal Sourcing: https://www.nimbleway.com/demos/deal-sourcing.md
- To fill firmographics and tech stack for the companies you found, use Account Enrichment: https://www.nimbleway.com/demos/account-enrichment.md
- To find the people to contact at them, use Decision-Maker Identification: https://www.nimbleway.com/demos/decision-maker-identification.md

## Sample result

From the demo's recorded run for "Vertical SaaS for dental clinics in North America", Describe your ICP mode, 10 results requested. The run checked 9 companies against two criteria (vertical SaaS for dental clinics, HQ in North America), and all 9 met both. Five of the rows:

| Company | Website | HQ | Employees |
|---|---|---|---|
| Archy | archy.com | San Jose, California, USA | 138 (LinkedIn, 51-200 range); 130 per company blog |
| Curve Dental | curvedental.com | Alpharetta, Georgia, USA | 214 (LinkedIn, 201-500 range) |
| tab32 | tab32.com | Sacramento, California, USA | 50+ (company About page); 55 (LinkedIn) |
| Planet DDS | planetdds.com | Irvine, California, USA | 440 |
| ClearDent | cleardent.com | Burnaby, British Columbia, Canada | Not given |

Sources include https://www.curvedental.com/, https://www.linkedin.com/company/archydental/ and https://www.archy.com/blog/weve-raised-50-million-to-bring-dental-practices-into-the-ai-era

Each row also carries a one-line reason it fits and a dated signal, such as a funding round or a product launch, with its source.
