How to research a company for sales in ten minutes: tier the account, read five public sources, and write one note you can actually use.

What company research for sales actually means

How to research a company for sales is a question about time, not about sources. The work itself is reading public information about a business before you contact anyone there, so that the first line you write or say is about them and not about you.

It is not a background check and it is not a full market study. The output is small: a handful of facts, one reason this company might care right now, and the name of the person who owns that problem.

Most sellers either skip it entirely and send something generic, or fall into a forty minute reading session that produces nothing they can use in a sentence. Both fail for the same reason: no time limit and no written output.

The routine on this page fixes both. You decide how many minutes an account earns, you read a fixed list of sources in a fixed order, and you stop when the note is full.

How to research a company for sales in ten minutes

This is the routine. It assumes the account already passed a fit check, so you are not deciding whether to contact them, only what to say. Set a timer, because the timer is the method.

1. Set the budget before you open a tab

   Decide how many minutes this account gets, based on its tier. A named strategic account earns twenty minutes. A row in a list of three hundred earns two, and most of that should come from enrichment, not reading.

2. Read the company site for what they sell and to whom

   Home page, one product page, the about page, and the customers page if there is one. You are looking for the words they use for their own buyers, because those words go straight into your first line.

3. Open the careers page and read the open roles

   Job ads are the cheapest signal on the internet. What a company is hiring for tells you which team is growing, which tools they run, and which problem got big enough to fund a headcount.

4. Look for what changed in the last ninety days

   News, press releases, filings, the product changelog and the blog. One recent change beats five permanent facts, because a change is a reason to be writing to them this week rather than any week.

5. Find the person who owns the problem

   Use the LinkedIn Page to see the team, then pick the one role that owns the outcome you affect. Check how long they have been in the seat and what they have posted about lately.

6. Write the note, then stop

   Fill the ten fields in the template below. If a field is still empty at minute ten, write "unknown" and move on. An incomplete note is finished research; an open tab is not.

Tier the account before you spend a minute on it

The most expensive mistake in account research is giving every company the same treatment. A four hundred row list and a named target account need different amounts of reading, and the difference is not small.

Write the tier into the row before the research starts. If you cannot say which tier a company belongs in, that is a targeting problem, and it is cheaper to fix in the prospect list than in the inbox.

Public sources that pay off, in the order to read them

Nine sources cover almost everything a seller needs, and they are all free. The order matters more than the list, because the early ones tell you which of the later ones are worth opening at all.

Everything in that table is public and free. Paid sales intelligence tools mostly save you the collecting, not the reading, and the reading is where the usable line comes from.

The company site: read the pages sellers usually skip

The home page is written for investors and analysts as much as for buyers, so it is the least useful page on the site. The pages below are better, and they take about two minutes together.

• One product page: the vocabulary they use for their own features tells you which words will sound native in your email and which will sound like a vendor.
• The customers or case studies page: who they sell to, at what size, in which industries. This is their ideal customer profile written out loud.
• Pricing or plans: whether they sell self-serve, sales-led or both, which tells you how their own revenue team is probably organized.
• The about or leadership page: founding story, office locations, and the names that will appear again in filings and press.
• The blog, sorted by newest: the last three posts show what the marketing team was told to care about this quarter.

Skip the resources library and the long-form guides. They are written to rank, not to describe the business, and they will eat ten minutes without giving you one specific sentence.

Careers pages and job ads, the most underused source

A job ad is a company telling the public exactly what it decided to spend money on. Nothing else on the open internet is that direct, and it is updated constantly.

• Which team is growing: six open roles in one function and none in the others tells you where this year's budget went.
• Which tools they run: required experience with named platforms is free technographic data, straight from the people who will use them.
• What is broken: responsibilities phrased as "build out", "clean up" or "stand up" describe a gap, not a routine.
• Seniority of the hire: a first head of a function means the function is being formalized, which is a buying moment.
• Location of the role: a role posted in a new city usually appears before the press release about the new market does.

A job ad also gives you a safe, checkable opening line. You are referencing something the company published on purpose, which is the difference between specific and salesy.

The data points worth collecting, and the ones that are not

Sales research goes wrong when it collects information that is true forever. The data points below are the ones that help, with an honest note on how long each one stays useful.

The first four are mechanical data, and a tool should fill them. The last four are the insights a person has to find, and they are where the outreach actually comes from.

SEC filings and EDGAR full text search

If the company is a United States public filer, or has raised money through an exempt offering, it has told the Securities and Exchange Commission things it has told nobody else in plain language. Those documents are public and searchable.

The tool is EDGAR full text search, at sec.gov/edgar/search. According to the SEC's own FAQ for the tool, it searches the full text of all EDGAR filings submitted electronically since 2001, including all data in the filing itself as well as all attachments such as exhibits.

What you can search by

The SEC says you can search by keyword, ticker, company name, CIK number and reporter's last name, individually or in combination. Natural language search is not supported, so a full question typed into the box will not behave the way a search engine does.

The operators that matter for sellers

Search is case insensitive, so capitalization does not change relevance. Under "More Search Options" you can limit by filing type and by date, with ranges the SEC lists as All since 2001, Last 5 Years, Last Year, Last 30 Days, or a custom range typed as yyyy-mm-dd.

Three searches worth running

• The company name plus your category term: shows whether they have ever described this spend, this risk or this vendor category in a filing.
• A competitor's name across all filers: exhibits and risk factors name vendors and partners, which surfaces companies already buying in your category.
• Your own product category, last thirty days: a standing search that turns new filings into a prospecting queue rather than a research errand.

This is a research source, not a shortcut around the rules. Filings are historical documents about a moment in time, and nothing in one entitles you to claim you know a company's current plans.

What each filing actually tells a seller

You do not need to read a whole annual report. Each form has named items, and three or four of them carry everything a seller can use.

Item 1A, Risk Factors, is the section most worth a seller's time. It is the company describing, in its own words and for a regulator, the things that could go wrong. If your product addresses one of those, you have a sentence nobody else in their inbox has.

Item 5.02 of an 8-K covers departure of directors or certain officers and appointment of certain officers. A new officer in a seat you sell into is a trigger, and the filing gives you the date it happened.

Form D is the one that works on private companies. The form asks for the issuer's identity, its principal place of business, and related persons, which it defines as each executive officer and director of the issuer and persons performing similar functions.

News, press releases and podcasts

News answers one question: what changed, and when. Sort by date, read the last ninety days, and ignore anything older unless it explains something in the recent items.

Press releases are written to be quoted, so quote them carefully. A claim on a company blog is the company's claim, not a fact, and repeating it as established truth is how a first line stops sounding credible.

The LinkedIn Page and the person behind the problem

The company Page gives you structure: headcount range, locations, what the marketing team publishes, and which functions are visible. The people tab tells you roughly how the org is shaped.

• Tenure in the seat: someone six months into a role is still deciding what to change. Someone six years in has already made the decision you are questioning.
• Where they worked before: a prior employer that already uses your category means the concept needs no explaining.
• What they post and comment on: their own words about the problem, which are safer to reference than anything you inferred.
• Who else is in the buying group: peers and the level above, because one contact is rarely the decision.
• Recent hiring in their team: visible new joiners confirm what the job ads suggested.

Our guide to LinkedIn prospecting covers the search side of this, including how to find the right titles when the org chart is not obvious from the Page alone.

Reviews, marketplaces and product changelogs

Review sites tell you what a company already pays for and what annoys them about it. App marketplaces and integration directories do the same job, and they are usually more current.

A changelog or release notes page tells you the shipping pace and the integrations they chose to build. For a technical sale, the public docs and the status page are better evidence than anything the sales page claims.

Read the complaints rather than the praise. A specific complaint about a tool in your category is the closest thing to a stated need that a public source ever gives you, and it pairs well with intent data if you have it.

Finding the challenges a company actually has

The hardest part of sales research is moving from information to a judgment about the prospect's challenges. Nobody publishes their pain points, so you infer them from signals, and then you say so honestly.

A pain point you inferred is a hypothesis. A pain point the company wrote down somewhere public is evidence. Keeping those two apart is the whole difference between research that helps and research that embarrasses you on the call.

Three kinds of signals help. Stated signals are the company saying something itself. Structural signals come from how the business is built. Behavioral signals come from what the company does rather than says.

Borrowed signals are the most tempting and the most dangerous. Pattern data from your own customers is useful for choosing what to ask about, and it is not evidence about this prospect until they confirm it.

The honest sentence is short: here is what we see, here is why we think it applies to you, tell us if it does not. That sentence gets more useful replies than a confident claim, and it costs nothing when the guess is wrong.

Turning the research into outreach

Research becomes sales outreach in one move: pick the single strongest item in the note and build the first line around it. Everything else in the note is for the conversation, not for the message.

1. Choose one item, usually field 3 or field 10, and drop the rest.
2. Write the line as an observation with a source, not as flattery.
3. Connect it to a problem in one clause, and mark it as a hypothesis.
4. Ask something the prospect can answer in a word.
5. Keep the note handy for the reply, because that is where the other nine fields help.

The same note feeds the whole sequence. A sales cadence that reuses one insight across five touches reads as persistence; one that invents a new hook each time reads as a machine working through a lead list.

If the insight will not survive being written down in one sentence, it was not an insight. That test kills most of the research that feels productive and produces nothing, and it is the fastest quality check a sales team can run.

Industry research: the context the company will not explain

A company page tells you what one business does. Industry research tells you what is happening to every business like it, and that is the context a prospect will not spell out for you.

Sellers who know an industry ask better questions. You can name the pressure the whole industry is under, and let the prospect tell you whether it applies to them, instead of guessing from their website alone.

• Revenue and profit trends: an industry where margins are tightening buys differently from one that is expanding, and the same pitch does not fit both.
• Industry outlook: whether the next few years are expected to grow or contract shapes how a prospect thinks about a multi-year commitment.
• Where the businesses are: market geography tells you whether your prospects cluster in a few regions or spread thin, which changes territory planning.
• Regulation and reporting: rules that apply to an entire industry create deadlines that no single company chose.
• Seasonality and buying cycles: knowing when an industry has money and attention keeps you from calling in its worst month.
• Who their customers are: if you understand the prospect's own buyers, you can talk about their revenue instead of your features.

Industry research is reusable in a way company research is not. Read one industry properly and it pays off across every prospect in that segment, which is why it belongs in team enablement rather than in each rep's ten minutes.

How prospect research fits next to company research

Prospect research and company research get used as the same phrase, and they are not. Company research is about the business. Prospect research is about the individual prospect you are about to contact, and about whether that prospect should be contacted at all.

In practice the order runs: qualify the account, research the company, then research the prospect. Skipping the first step means researching prospects who were never going to buy, which is the most common way sales teams waste research hours.

• Is this prospect in the buying group? Many prospects in a CRM are users, not owners of the outcome you affect.
• How long have they been there? A prospect in their first quarter is still auditing what they inherited.
• What have they said in public? A prospect's own posts and talks are the safest thing to reference.
• Have we spoken before? Prior conversations with this prospect or their colleagues sit in the CRM and outrank anything you can find online.
• Who else should hear this? One prospect is rarely the decision, so note the two other names while you are there.

Our guide to qualifying sales leads covers the gate that should come first, and prospecting covers where these prospects come from in the first place.

Account research and contact research are two different jobs

Account research is about the company: what it sells, how it makes money, what changed, and which pressures it is under. Contact research is about one person: their role, their tenure, their words and their likely mandate.

Do the account research once and store it where the whole team can see it. Redoing it for every contact at the same company is the single biggest waste of time in outbound, and it is invisible because it feels like work.

What to write down: the account research note

Research that lives in browser tabs is not research. The note below was written for this page and has ten fields, because ten is what fits on one screen and what a colleague can read before a call.

Company: {{companyName}}
Tier: {{tier}}    Researched: {{date}}    By: {{yourName}}

1. What they sell, in their words: {{oneLine}}
2. Who they sell it to: {{theirBuyer}}
3. What changed in the last 90 days: {{change}}  Source: {{url}}
4. What they are hiring for: {{openRoles}}
5. Tools we know they run: {{stack}}
6. The problem we think they have: {{hypothesis}}
7. Evidence for it: {{evidence}}
8. Who owns that outcome: {{name}}, {{title}}, in seat since {{startDate}}
9. Others in the buying group: {{names}}
10. Reason to write this week: {{reason}}

Unknowns I chose not to chase: {{unknowns}}

Two rules keep the note honest. Every claim carries the link it came from, and every field you could not fill says "unknown" rather than staying blank, so the next person knows it was looked for.

Store the note on the account record in the CRM, not in a personal document. That is what makes the research an asset instead of a chore, and it is what lets lead enrichment fill the mechanical fields for you next time.

Research before a call versus research before an email

The two need different depth, and confusing them is why some sellers over-research a cold email and under-research a first call.

For an email, one fact used well beats five facts stacked up. For a call, the research is not material to recite; it is what lets you ask a better question. Our guide to making a sales call covers what to do with it once the call starts.

If you cannot turn the note into one natural opening sentence, the research did not work. Rewriting the fact as a question is usually the fix, and how to write a cold email covers the rest of the message.

When company research is wasted

Research is not free, and there are cases where more of it makes the outcome worse rather than better. Recognizing them is how you get the ten minutes back.

• The account does not fit. No amount of reading fixes a company that cannot buy what you sell. Fix the targeting instead.
• The facts never reach the message. If the email would be identical without the research, the research was a private hobby.
• Everything you found is permanent. Headquarters, founding year and employee count explain nothing about why this week.
• It is already in the CRM. Somebody at your company researched this account last quarter. Read that note first.
• The signal is a month past its window. A funding round from last year or a role change from eighteen months ago dates your list, not your insight.
• You are researching to avoid sending. The most common one, and the hardest to admit. A timer is the only reliable cure.

There is also a limit to what any public source can tell you. Budget, internal politics and the real reason the last project stalled are not published anywhere, and you learn them by asking.

Doing this across a list, not one account

Ten minutes per account is fine for twenty accounts and impossible for four hundred. At list scale the research has to become columns, filled once, rather than reading repeated per row.

1. Decide the fields every row must have, and keep the list short enough that every row can actually have them.
2. Fill the mechanical fields by enrichment or by tool, not by hand: industry, size, location, stack.
3. Sort the list by signal, so the accounts with a recent change get the human minutes.
4. Hand-research only the top slice, and write a real note for each one.
5. Send the rest with a message that is honest about being a fit-based message, not a fake personal one.

The failure here is a half-personalized email: one researched detail bolted onto a template that does not follow from it. Readers notice the seam. A clean fit-based message outperforms a clumsy imitation of a personal one.

Who on the sales team does the research

Account research keeps getting redone because nobody owns it. In most sales teams the work splits cleanly once you decide which part is reusable and which part is per message.

The rule that keeps it working: research done for one message goes in the contact record, research about the business goes on the account record, and nothing important lives only in somebody's head.

Tools, as categories

This page does not rank vendors or quote prices. These are the categories that speed up company research for sales, and every one of them is optional.

• Sales intelligence and databases: firmographics, contacts and signals in one place, so you read rather than collect.
• Enrichment: fills the mechanical columns on a list automatically, so the human minutes go to the interpretive data instead.
• Job ad and hiring trackers: turn careers pages into a feed instead of a tab you remember to check.
• News and alert tools: standing alerts for named accounts and for your category.
• The CRM: where the note lives, which is the only tool on this list that is not optional.

Automated research summaries are useful for the mechanical half and unreliable for the interpretive half. Verify any claim you intend to put in a message, because a confident wrong sentence costs more than a generic right one.

How to tell whether the research is paying off

Common mistakes in account research

• No time limit, so one account absorbs an afternoon.
• Collecting facts that are true forever and explain nothing about now.
• Repeating the same account research for every contact at that account.
• Treating a company blog claim or a press release as verified fact.
• Writing a hypothesis with no evidence, then stating it with confidence.
• Leaving the note in a personal file instead of the account record.
• Reciting research on a call instead of turning it into questions.
• Researching accounts that failed the fit check in the first place.
• Using a signal long after its window closed, which dates the list.

Turning the note into an opening line

The note is only worth the ten minutes if a sentence comes out of it. The template below was written for this page: it takes fields 3, 6 and 10 and turns them into an opener that could not have been sent to anyone else.

Subject: {{change}} at {{companyName}}

Hi {{firstName}},

{{change}}, which I saw in {{sourceName}}. Teams that do that usually hit {{problem}} next, because {{reason}}.

That may not be true for you. If it is, {{yourName}} at {{yourCompany}} has a short write-up of what the {{role}} teams we work with changed first, and I can send it without a call attached.

If someone else owns this now, tell me who and I will stop writing to you.

{{yourName}}

Frequently asked questions