Industry
Part of Entity type selection: three decisions, not one
Entity type selection metrics worth tracking, and the ones to ignore
Entity type selection metrics: what can honestly be counted, what a bad result means, what no measure will tell you, and where the real numbers live.
Whether your business structure was the right choice comes down to whether you get sued, by whom, and for how much. The outcome is unmeasurable in advance, and anyone quoting you a number for it is selling something.
The behaviors that decide whether the structure holds can be counted, and almost nobody counts them. That is what follows.
No benchmarks appear below, because none are worth quoting. These are readings you take on your own business and compare with your own last reading. A fee, a deadline, a rate or a threshold attached to this subject comes from your state or your accountant.
What to take away
- The readings are not independent. They fail in a pattern.
- Annual review is worth having, but the changes that matter arrive between reviews.
- Any figure on this subject comes from an authority and changes.
Metrics worth tracking
| Metric | How to read it |
|---|---|
| Commingled transactions, last quarter | Business money in a personal account, or personal money in the business account. The only defensible number is zero. Anything else is evidence that there are not two separate parties. |
| Outstanding personal guarantees, count | Every one is a hole straight through the structure. If you cannot list them from memory, that is the finding. |
| Contracts signed in the wrong name, count | Agreements in your personal name that should be the entity's, or signed without stating the capacity you signed in. Each is an argument waiting to be made about who the counterparty dealt with. |
| Days since you last read your registry record | The state's public record is where good standing, agent and addresses live. If the answer is "never", start there. |
| Recurring obligations calendared, against obligations known | Count the ones you know about, then count the ones actually on a calendar. The gap is the exposure. |
| Owners without a signed governing document, count | Should be zero from the first day there is more than one owner. |
| States where you are registered, count | Every registration carries its own agent and its own recurring filings. People routinely undercount this. |
| Documented decisions in the last year, count | Minutes, resolutions, written consents. Zero is a common answer and a poor one. |
| Hours per year spent on entity administration | Track it once. It tells you whether the structure you chose is one you are actually maintaining. |
| Days between a change and every system knowing it | Address, agent, owners, name. Measured in weeks rather than in intentions. |
Take these once a year. Ten minutes, answered honestly, tells you more about whether your structure is doing its job than any amount of reading about entity types.
Entity Metrics to Count
- Wrong-account transactionsonly defensible number is zero
- Personal guarantees outstandinglist from memory
- Contracts signed in the wrong name
- Days since you read your registry record
- Recurring obligations with a calendar date
- Owners without a signed governing document
- States you appear inagent and filings
Metrics to ignore
These are easy to find, easy to count and popular in entity-choice conversations. Every one of them describes an economy, a population or a sales funnel rather than the separation between you and your business.
| Metric people quote | Why it says nothing about your entity |
|---|---|
| Census Bureau Business Formation Statistics | Monthly counts of new business applications, built from EIN applications, nationally and by state. A record month measures the economy, not your contracts. |
| Delaware's share of the Fortune 500 | The talking point is that about two-thirds of Fortune 500 companies incorporate in Delaware. That is a fact about public companies raising outside capital, filed through the Delaware Division of Corporations. Your single-member LLC in Ohio is answering a different question. |
| State fee tables | Delaware's annual LLC tax is $300 and Wyoming's annual report fee starts at $60, read from each state's fee page; record the date you read them. Both change by legislative session. A low fee is revenue policy, not protection. |
| Entity-choice quizzes and comparison charts from formation vendors | LegalZoom, ZenBusiness and Bizee publish comparison charts that route the reader into their own filing product. The quiz output is a sales path. |
| Business credit scores | Dun & Bradstreet's PAYDEX runs from 1 to 100 and scores 80 or above for payments made on time to suppliers. Experian's Intelliscore Plus rates supplier payment behavior on a 1 to 100 scale. Neither reads your bank accounts or your written consents. |
| Search interest in "S corporation election" | Google Trends indexes search interest from 0 to 100 against a peak. Popularity of the term among strangers is not a reading on your tax classification. |
| Veil-piercing case counts | Searching "piercing the corporate veil" on CourtListener or Google Scholar returns published opinions in your state. Each was decided on its own facts, after a loss. Counting them predicts nothing about yours. |
| Compliance solicitation mail | Private companies send "certificate of good standing" offers that look official. Most states let you check standing free on the registry site and charge a modest fee for a certified copy, typically $10 to $50. The volume of mail reflects mailing lists. |
One test separates a metric from noise. You must be able to take the reading from your own records, take it again next year, and act on the answer. Everything in the table above fails at least one of the three.
Reading a bad result
The readings are not independent. They fail in a pattern.
Bad Result Patterns
Combination
- Commingling + no decisions
- Separation hardest to defend
- Unlisted guarantees + wrong-name contracts
- Exposure not as structure suggests
- Known but uncalendared obligations
- Predicts loss of good standing
- High admin hours + low decisions
- Effort in wrong things
What it means
- Commingling + no decisions
- Entity exists on paper only
- Unlisted guarantees + wrong-name contracts
- Fix with an afternoon and a folder
- Known but uncalendared obligations
- Notices go to stale address
- High admin hours + low decisions
- Bookkeeping yes, governance no
Commingling plus no documented decisions is the classic combination, and the one that makes the separation hardest to defend. Neither is dramatic on its own. Together they describe a business where the entity exists on paper and nowhere else.
Start by opening a separate business account and moving every business transaction into it, then write one page of minutes for the decisions you can still date.
Guarantees you cannot list, plus contracts in the wrong name, means your actual exposure is not what your structure suggests. Fixing it is an afternoon with a folder rather than a restructuring.
Obligations you know about but have not calendared is the reading that predicts a loss of good standing. The notices go to whatever address is on file, which is the record most likely to be stale. See registered agents for why that particular failure is the expensive one.
High administration hours with low documented decisions suggests effort going into the wrong things: bookkeeping happening, governance not.
What no entity metric will tell you
A full set of good readings is not proof of safety. Five questions sit outside every metric on this page.
What No Metric Tells You
- Litigation risk for your business
- Whether a court respects separation
- Whether tax classification is optimal
- Whether you are adequately insured
- Whether structure fits in three years
- Litigation risk.Not forecastable for an individual business. Industry generalities do not tell you about yours.
- Whether a court would respect the separation.That depends on facts, conduct and jurisdiction, assessed after the event.
- Whether your tax classification is optimal.This is arithmetic on your actual numbers, done by someone who prepares returns. It changes as the numbers change, and it cannot be reduced to a threshold.
- Whether you are adequately insured.A broker's question, and one that matters more than most people expect: an entity determines who a claim lands on rather than who pays it.
- Whether the structure will still fit in three years.Nothing measures this. Reviewing on events does the job instead.
Review on events, not on a schedule
Annual review is a habit worth having, but the changes that actually matter arrive between reviews. Reopen the question when:
- A new member joins, a member leaves, or ownership transfers.
- You register in a new state.
- You sign a new personal guarantee.
- The business address changes.
- The registered agent changes.
- A new counterparty signs a contract with you.
Review on Events
- Owner joins or leavesReopen the question
- First employeeReopen the question
- Operating in another stateReopen the question
- Materially larger contractReopen the question
- Money out changesReopen the question
- Riskier activity lineReopen the question
Each of these can also make something on the record stale: post-formation changes covers the update chain and what breaks when a link in it is missed.
Where the real numbers live
These figures come from an authority and change without notice. Look them up rather than carrying them in your head:
Where to Verify Numbers
- State registryformation and recurring filings
- IRS small business pagesfederal tax
- SBA business guidegeneral orientation
- Federal identifier explainedEIN need and timing
- Formation and recurring filing requirementsyour own state's registry. State filing guides covers how to find and verify the official page.
- Federal tax obligations for businessesthe IRS's pages for small businesses.
- General orientation for new businessesthe SBA business guide.
- Whether your chosen form needs a federal identifier, and when to apply: the EIN application is IRS Form SS-4, and the federal identifier explained covers it.
Registered agent fees belong in the same check. Commercial agents typically charge $100 to $300 a year, and Northwest Registered Agent has long published $125 a year. The number you need is the one for each state where you are registered.
Whatever you copy out of any of them, note the date you read it and recheck before relying on it again.
Common questions
Is there a revenue level at which I should form an entity?
No such threshold exists as a general rule, and any number offered as one is somebody's rule of thumb. Exposure and ownership drive the decision far more than revenue does, and the tax arithmetic depends on your specific numbers.
How do I compare the cost of two structures?
Over several years, not at formation. Include recurring state filings, anything charged for the entity's existence, agent fees in each state you appear in, and the accounting work each structure requires. The one-time formation fee is the least significant term.
What is the single most useful thing to track?
Transactions in the wrong account. It is easy to count, the correct answer is unambiguous, and it is the behavior most directly connected to whether the separation holds.
How often should I run through this?
Once a year, plus on any of the events listed above. It is a short exercise, and its value is entirely in being repeated.
