---
name: event-target-accounts
description: Builds a tiered target account and contact list for a trade show or conference from the public exhibitor list, sponsor and speaker lists, an organizer attendee list or a registrant list. Normalizes company names to domains, matches them against the CRM (customers, open opportunities, target accounts, owners), scores ICP fit and assigns account tiers with a play per tier. Use for conference prep, pre-show target lists, ABM at events or matching an exhibitor list to accounts.
license: MIT
metadata:
  author: eventmarketingsoftware.com
  version: "1.0"
  category: trade-shows-and-field
  source: https://eventmarketingsoftware.com/skills/event-target-accounts
---

# Event target accounts

Turn a show's exhibitor, sponsor, speaker or attendee list into a tiered target list: who's there, how they relate to
you in the CRM, how well they fit your ICP, who owns them and what to do about each one before and during the show.

Account tiers here are **Tier 1, Tier 2, Tier 3** (accounts). Don't confuse them with **lead tier A/B/C**, which the
lead scoring and follow-up skills use for individual leads after the show.

## Intake

If the user has an `event-context.md` file, read it first and only ask for what's missing. Otherwise ask for the
missing essentials in ONE message:

1. Event name, dates and city, and whether you're exhibiting, sponsoring or attending without a booth.
2. The event list: exhibitors, sponsors, speakers, attendees or registrants (paste, CSV/XLSX, or a copied web page).
3. ICP: industries, employee or revenue range, regions, and any must-have or disqualifying traits.
4. CRM data: a connector, or an export of accounts with domain, type (customer, prospect, partner, competitor),
   owner, open opportunity stage and amount, and any named target account list.
5. Capacity: roughly how many meetings, dinner seats or booth conversations the team can handle (sets tier sizes).

If the user has no CRM export, still run the ICP scoring and mark every row `crm_status = unknown`.

## Connectors (optional)

- **CRM connector (HubSpot, Salesforce or similar):** read accounts, owners and open opportunities by domain. Read only.
- **Google Sheets connector:** read the event list from a sheet, or write the result to a new tab after approval.
- **No connector:** work from pasted text or uploaded CSV/XLSX and return a downloadable CSV plus a Markdown summary.

Never call a website or API at run time to scrape or enrich. Use only data the user gives you or their connected tools.

## Method

Copy this checklist into your reply and tick it off:

```
- [ ] 1. Load and count input rows
- [ ] 2. Normalize company names and domains
- [ ] 3. Dedupe to one row per account
- [ ] 4. Match to CRM accounts
- [ ] 5. Score ICP fit
- [ ] 6. Assign tier and play
- [ ] 7. Validate counts and list unmatched rows
```

### 1. Load
Identify the list type (exhibitor, sponsor, speaker, attendee, registrant). Record the input row count. Keep the
original company string in `company_raw` so nothing is lost.

### 2. Normalize
- Strip legal suffixes (Inc, LLC, Ltd, GmbH, SA, BV, Corp, Co, plc) and punctuation; lowercase for matching.
- Treat "The", "&" vs "and", and trailing words like "Group", "Holdings", "Technologies" as soft differences.
- If the list has URLs or emails, take the domain (drop `www.`, paths and personal email domains such as gmail.com,
  outlook.com, yahoo.com).
- Don't invent a domain. If none is given, leave `domain` blank and match by normalized name only.

### 3. Dedupe
One row per account. Keep the richest row and merge contact names (speakers, attendees) into a `contacts` column.
Exhibitors listed under several booths or brands become one account with all booth numbers.

### 4. Match to CRM
Match in this order and record the method in `match_method`:
1. exact domain, 2. exact normalized name, 3. close name (one token different, abbreviations, parent vs subsidiary).

Close-name matches are `match_confidence = review`; never treat them as confirmed. Pull `crm_status`
(customer, open opp, target account, prospect, partner, competitor, unknown), `owner`, `opp_stage`, `opp_amount`.
If two CRM accounts match one event row, flag both and don't pick.

You can run `scripts/match_accounts.py` for steps 2 to 4 when code execution is on (see "Script" below).

### 5. Score ICP fit
Score 0 to 3 on each criterion the user gave (industry, size, region, any must-have). Use the rubric in
[references/tiering-rubric.md](references/tiering-rubric.md). Unknown values score 1, not 0, and get noted in
`fit_notes`. Don't guess employee counts or revenue; say "not in data".

### 6. Assign tier and play
Default rules (the user can override):

| Tier | Rule | Default play |
|---|---|---|
| Tier 1 | Open opp, named target account, or ICP fit score in the top band | Book a meeting before the show; consider dinner invite; AE and exec outreach |
| Tier 2 | Customer (expansion or advocacy), or ICP fit in the middle band | Invite to booth or session; customer touch from CSM or AE |
| Tier 3 | Fits loosely or unknown fit | Recognize at booth; add to post-show nurture only with consent |
| Exclude | Competitor, partner with no sales motion, or disqualified | Don't contact; note why |

Size check: if Tier 1 is bigger than the team's meeting capacity, raise the cutoff and say so. Customers with an open
expansion opp go to Tier 1. Competitor exhibitors are useful for intel; list them separately.

For speakers and attendees, add `contact_name`, `title`, and `persona` (decision maker, influencer, user, unknown) from
the title. Seniority from title only; never guess emails.

### 7. Validate
- Rows in = rows out + rows merged as duplicates + rows excluded. Show the math.
- List every unmatched or `review` row in a separate table.
- State every assumption (suffix rules, ICP thresholds, tier cutoffs).

## Output

1. **Summary:** counts by tier, by crm_status, and number of review rows. One line on Tier 1 versus capacity.
2. **Target list** (Markdown table and CSV) with these columns in this order:

`tier, account, domain, company_raw, list_source, booth_or_session, crm_status, owner, opp_stage, opp_amount,
icp_score, fit_notes, contacts, match_method, match_confidence, play, next_step`

3. **Review table:** close-name matches, multi-matches and rows with no domain.
4. **Competitors and excluded:** account and reason.
5. **Owner view (optional):** Tier 1 and 2 grouped by owner, ready to paste to each AE.

Worked example: [references/examples.md](references/examples.md).

## Approval gates

- Never write to the CRM (new accounts, campaign members, tasks, fields) without showing the exact rows and getting a
  clear yes. Default to a CSV the user imports.
- If the user asks to add contacts to a campaign, use their existing campaign member statuses (for example "Targeted"
  or "Invited"). Preview first.
- Don't post owner lists to Slack or email without a preview and a yes.

## Privacy and consent

- Exhibitor and sponsor lists are public company data. Attendee and registrant lists are personal data the organizer
  shared under its own terms; use them only as those terms allow (many restrict them to opted-in attendees).
- Flag contacts in the EU or UK (from country, phone prefix or domain suffix) with `consent_check = needed` unless
  the data shows opt-in. Don't add them to marketing lists from this step.
- Keep personal data inside the user's own tools. Don't paste attendee data into public places.

## Script (optional)

`scripts/match_accounts.py` normalizes names, dedupes the event list and matches it to a CRM export by domain and
name. Python standard library only:

```
python scripts/match_accounts.py event_list.csv crm_accounts.csv matched.csv
```

It expects a `company` column (and optional `domain` or `website`) in the event file, and `account_name`, `domain`
in the CRM file. If code execution isn't available, do steps 2 to 4 in-chat with the same rules.

Sample inputs to try: `assets/sample-exhibitor-list.csv` and `assets/sample-crm-accounts.csv`.

## Pushback

- "Just give me everyone": still tier them; an untiered list of 400 exhibitors doesn't book meetings.
- "Can you find their emails?": no. Suggest the CRM, the rep, or a tool the user already licenses.

## Related skills

pre-show-meeting-outreach (writes the outreach for Tier 1 and 2), exec-dinner-planner (picks dinner guests from
Tier 1), booth-staff-briefing (shares Tier 1 names with booth staff).
