How to Build a Startup Sales List from Public SEC Data

Why SEC data beats every other list source

Most funded startup lists are lagging indicators. Crunchbase is updated after the press release. LinkedIn company pages fill out when the hiring starts. Apollo and ZoomInfo pull from those same downstream sources. By the time a round shows up anywhere you are already prospecting, it is 3 to 6 weeks old and every other rep with a Google alert has already sent their outreach.

SEC Form D is the primary source. Under Regulation D, any US company that raises equity or debt in a private offering must file a Form D with the SEC within 15 days of the first sale of securities. No exemptions for stealth-mode companies. No grace period for small rounds. The filing is public record the day it hits EDGAR.

That 15-day window is the gap between the Form D filing date and the press announcement. During that gap, the company has fresh capital, no vendor list, and a founder or department head who is actively evaluating tools. The team is in planning mode, not execution mode. Budget decisions are being made, not deferred. That is when you want to be in their inbox, not after the TechCrunch post.

Every funding signal you buy from a data vendor is a derivative of this public source. The SEC data is free, daily, and earlier than anything else. The only cost is the time to work it.

What you need before you start

Before touching EDGAR, define your ideal company profile. Vague criteria produce garbage lists. Be specific on five dimensions:

1. Funding stage. Seed and Series A companies are typically 5 to 30 employees, founder-led, and buying on instinct. Series B and C companies have department heads and procurement processes. The buyer and the message are different at each stage.
2. Raise size. Match the raise to your deal size. If you sell a $20K/year SaaS tool, a $500K pre-seed is a real prospect. A $500K raise is not a real prospect for a $200K enterprise contract. Use $2M as the floor for most SMB tools, $10M for mid-market, $25M for enterprise.
3. Sector. Form D filings include an industry code field. Not all sectors buy what you sell. Fintech, SaaS, and healthcare tech are high-density target sectors for most B2B tools. Real estate syndications and oil-and-gas LLCs file Form D in high volume but are mostly irrelevant.
4. Geography.Form D lists the company's registered state. If you sell US-only or have a regional team, filter accordingly. Delaware registrations cover most VC-backed startups regardless of where the team actually sits.
5. Days since filing. This is the most important filter. Set a hard cap at 30 days. Filings older than 30 days have likely gone public. The buying window is mostly closed. You are competing against reps who acted weeks ago.

Write these criteria down before you open EDGAR. Every filter you skip adds noise you will spend time cleaning later.

The manual EDGAR workflow, step by step

This is the honest version of the workflow: it works, and it takes 3 to 4 hours per week when done manually.

1. Hit the full-text search endpoint. Go to efts.sec.gov/LATEST/search-index?q=%22form+D%22&dateRange=custom&startdt=YYYY-MM-DD&enddt=YYYY-MM-DD&forms=D. Swap in a 7-day date range. This returns raw JSON. Each result is a Form D filing with accession number, company name, filed date, and state of incorporation.
2. Download the XML for each filing. Each Form D filing has a structured XML file at https://www.sec.gov/Archives/edgar/data/[CIK]/[accession]/primary_doc.xml. The XML contains the full structured record: total offering amount, amount sold, type of securities (equity, debt, convertible), exemption type (506b or 506c), investor count, and signer names with titles.
3. Parse and filter.Write a Python script to parse the XML fields you defined in step one: amount raised, industry code, state. Filter out real estate LLCs, funds, and other non-operating companies. A basic heuristic: if the entity name ends in "Fund", "LP", "Partners", or "LLC" with no other product signal, skip it. This requires judgment calls on hundreds of rows.
4. Pull signer names. Every Form D lists at least one signer, the person who certified the filing. This is usually the CEO, CFO, or managing partner. Their name and title are in the XML. This is your starting contact for the prospecting sequence.
5. Cross-reference LinkedIn Sales Navigator. Search the company name in Sales Nav to find the current headcount and confirm the signer's current title. The Form D may list someone who has since left. Check. Also find the department head for your specific category: head of engineering, VP of marketing, CFO depending on what you sell.
6. Check for web presence. If the company has no website or a parked domain, the raise was likely a fund or a pre-product entity. Flag and skip. SimilarWeb and Semrush can show whether the company has real web traffic, which narrows to post-launch companies with active users.
7. Load into your CRM. Drop the cleaned list into Salesforce or HubSpot with the filing date as a field. You will use filing date to time your outreach sequences and track aging.

Expect to process 200 to 400 raw filings per week to produce 15 to 40 qualified prospects, depending on how narrow your ICP is. Each filing takes 3 to 8 minutes to review, filter, and cross-reference. That is where the 3 to 4 hours per week comes from.

How to score and prioritize

Not all Form D prospects are equal. Four signals separate high-priority leads from pipeline filler:

• Recency. File date within 14 days: top priority. 15 to 30 days: standard priority. Over 30 days: low priority, likely already announced. Sort your list by filing date descending before anything else.
• Raise size vs. your deal size. A company that raised $3M and your average contract is $50K is a real conversation. A company that raised $500K for the same deal size is a longer sales cycle at best. Build a ratio column: raise size divided by your ACV. A ratio above 20x is a safe buying signal.
• 506b vs. 506c. The exemption type tells you something about the investor base. 506c offerings are publicly solicited, meaning the company is more outward-facing and often more growth-oriented. 506b offerings are to pre-existing relationships, which skews toward traditional VC rounds. Neither is better universally, but 506c companies are often more open to cold outreach by disposition.
• Revenue range signals. Form D does not include revenue, but you can infer it. A seed company with 5 employees and no web traffic is pre-revenue. A Series B company with 50 employees and 100K monthly web visitors is in growth mode. Use headcount and traffic as revenue proxies when direct data is not available.

Score each row on recency, raise-to-ACV ratio, exemption type, and revenue signal. Weight recency highest. A top-quartile prospect is filed within 14 days, raised at least 20x your ACV, and has detectable web traffic. Work those first.

What to do with the list

The 30-day buying window is real. Funded companies spend the first month after close setting priorities, evaluating tools, and making initial vendor decisions. After that, the team goes heads-down building, the CFO gets more conservative, and new vendor consideration requires longer procurement cycles.

Time your outreach to the filing date, not the date you built the list. A prospect filed 5 days ago gets your first touch today. A prospect filed 22 days ago gets a compressed sequence.

Message framing: do not lead with the funding. Every rep in their inbox is leading with the funding. Lead with category timing. The angle that works:

Congrats on the round. This is usually when teams start locking in [your category] before the stack gets set. Worth 15 minutes to see if the timing makes sense?

Three sentences. No pitch. No capabilities deck. The goal is a reply. Keep the reference to the round brief: one word, "congrats," not a paragraph about how impressive the raise is.

Personalize with specifics from the filing: industry, amount raised, funding type. A $500K seed-stage SaaS company gets a different message than a $30M Series B fintech. The raise size tells you where they are in the scaling arc. Match your message to that.

For sequencing: three touches over 10 days. First touch on day one. Second touch on day four, one sentence adding context. Third touch on day ten, a short close or permission to opt out. No more than three touches in the 30-day window. After 30 days, move to a low-frequency nurture.

FlareSight: same list, enriched, in 20 minutes

The manual workflow above produces a real list. It also takes 3 to 4 hours every week, requires a Python script you have to maintain, depends on a LinkedIn Sales Navigator subscription for contact finding, and still leaves gaps in company descriptions and revenue signals.

FlareSight runs the same pipeline automatically. Every Form D filed with the SEC is indexed within hours. Each company is enriched with web traffic data, founder contacts, funding history, and a plain-English description of what the company does. The filters you would build manually (raise size, filing date, industry, state, funding type) are pre-built and instant.

The output is the same list you would build by hand, minus the 3 to 4 hours. You apply your ICP criteria in the filter panel, sort by filing date, and export. The workflow that takes a full work morning manually takes about 20 minutes with FlareSight.

The underlying data is the same: Form D filings from EDGAR, the same public source your competitors have access to. The difference is time. Reps who act on the filing within 14 days close more meetings than reps who act on the same data at 45 days. The tool does not change the signal. It changes how fast you can act on it.