How to Spot a 10b5-1 Plan in Form 4: The VWAP Fill Profile

Every quarter, retail investing media surfaces some version of 'CEO Dumps $X Million in Stock — Time to Worry?' The headline is usually built on a Form 4 filing showing an insider sale. Most of those headlines misread the cadence: the sale they're flagging as a discretionary downside signal is, in reality, the mechanical output of a pre-scheduled Rule 10b5-1 trading plan adopted months earlier. The data tells you which is which, but only if you read the fill profile rather than the dollar total.

This guide walks through the three structural giveaways of a 10b5-1 plan in Form 4 — VWAP fill profile, uniform daily share count, and multi-session continuation — and explains why each one matters for the analytical read. The examples come from filings published on our platform this quarter.

What a Rule 10b5-1 Plan Actually Is

Rule 10b5-1 is an SEC affirmative-defense rule that lets corporate insiders trade their company's stock without violating insider-trading prohibitions, provided the trades happen under a written plan adopted while the insider is not in possession of material non-public information. Once the plan is adopted, trades execute mechanically: the broker runs them on the dates and at the prices specified, regardless of whether the insider subsequently learns market-moving information or changes their view.

The plan's mechanical execution is the key analytical property. A 10b5-1 sale is, by structural definition, decoupled from the insider's contemporaneous view of the stock. A discretionary sale is not. Reading one as the other is the most common error in retail-investing insider coverage.

Giveaway 1: VWAP Fill Profile (The Strongest Signal)

A Volume-Weighted Average Price (VWAP) execution slices a parent order into many small child orders distributed across the trading session, in proportion to the realized volume profile. The broker's algorithm aims to minimize market impact by tracking the day's actual volume curve. This produces a characteristic Form 4 footprint: the same single day's sale shows up as 8-25 (sometimes more) micro-fills at slightly different prices, all within a tight intraday range.

From CrowdStrike CEO George Kurtz's April-May 2026 filings, a single trading session (May 4, 2026) shows 25 separate fills at prices between $467 and $475 totaling 9,451 shares. Twenty-five fills on a single calendar date is not human-driven trading — it's algorithmic execution against a parent order. A discretionary block sale by an insider talking to their personal broker would show 1-3 fills, not 25.

The fill-count threshold is the discriminator most readers miss. Look for:

• ≥5 fills on a single date: indicates algorithmic execution; could be a 10b5-1 plan or a large discretionary order routed VWAP-style
• ≥10 fills on a single date: almost always algorithmic; strong 10b5-1 plan signal
• 1-3 fills on a single date: discretionary or partial-session block

Giveaway 2: Uniform Daily Share Count

10b5-1 plans typically specify the daily quantity in shares: 'sell 2,500 shares per day' or 'sell 1% of position per session.' That produces a recognizable Form 4 footprint where the daily aggregate is the same number, repeated across multiple trading sessions, regardless of what the stock price is doing.

The Kurtz Form 4 tape this quarter shows exactly that pattern. April 28, 29, 30, May 1, May 5, and May 6 each show 2,500 shares sold. April 27 shows 1,948 shares (likely a partial-session catch-up), and May 4 shows 9,451 (likely a rollover from a missed session). The 2,500-per-day baseline is the plan parameter; the deviations are mechanical adjustments.

Compare this to a discretionary block: an insider deciding to sell into a price spike would typically execute one large block (5x to 20x their typical daily volume) on a single date and then go quiet for weeks. The uniform-daily pattern simply doesn't match that profile.

Giveaway 3: Multi-Session Continuation

10b5-1 plans run for predefined periods — usually 12 months from adoption, with re-adoption windows around year-end or post-earnings. The execution cadence persists across this period regardless of news flow. A discretionary view changes; a plan does not.

The Form 4 signature: same insider, same ticker, similar daily sizing, across multiple consecutive trading sessions spanning at least 2-3 weeks. Martine Rothblatt's UTHR exercise-and-sell sequence shows seven sessions of identical 9,500-share sizing across April 16 to May 7, 2026 — a textbook plan continuation.

What breaks the pattern (and would signal a regime change worth flagging):

• A sudden jump in daily sizing without a corresponding gap in filings
• A discretionary-looking single-block sale appearing inside the plan window
• An unexplained pause where the plan should have executed
• A new plan adoption with different parameters (you can sometimes infer this from a change in cadence post-blackout)

The 10b5-1 Footnote: Sometimes Present, Sometimes Not

SEC Form 4 includes optional footnote fields where the filer can indicate that the transaction was executed pursuant to a Rule 10b5-1 plan, and may also disclose the plan adoption date. When present, this is the definitive answer — no inference required. But footnote disclosure is voluntary, not mandatory, and many filers omit it.

When the footnote is absent, the three structural giveaways above are your inference path. The combination of VWAP fill profile + uniform daily share count + multi-session continuation is essentially diagnostic. A filing with all three features is, with very high probability, a plan execution. A filing with none of those features is, with very high probability, a discretionary trade.

Other Transaction Codes That Mimic 10b5-1 (And Aren't)

Form 4 carries other transaction codes that produce execution patterns superficially similar to a 10b5-1 plan but with different analytical meaning:

• Code F (Tax withholding at vest): Mechanical IRS-rate withholding when restricted stock units vest. Not a discretionary sale; not a directional view. Sized to cover the insider's tax liability at the vest moment.
• Code M (Option exercise): Converts an option into common stock. May or may not be followed by an S (sale) on the same day. The exercise itself is not a market trade.
• Code A (Award / grant): Compensation grants. Not a market signal at all.
• Code G (Gift / transfer): Charitable or trust transfers. Not a market trade.

An insider with weekly F codes alongside intermittent S codes is running a routine compensation-conversion cycle. The S codes here may still be plan-driven (10b5-1 plans often pair the S with the F or M to convert the vested shares to cash), but the underlying narrative is 'compensation flowing through to liquid form,' not 'insider losing faith.'

Reading the Pattern: A Diagnostic Checklist

When you see a Form 4 insider sale, run this mental check before drawing any sentiment inference:

1. Is the transaction code S (sale)? If not (M, F, A, G, C, D), it's structurally different — proceed to the relevant code's interpretation.
2. Does the filing show a 10b5-1 footnote or plan-adoption date? If yes, framing is plan-driven by SEC disclosure.
3. How many fills per session? ≥10 fills strongly indicates algorithmic execution. ≥5 is suggestive. 1-3 is discretionary or single-block.
4. Is the daily share count uniform across sessions? Same number repeated for 3+ days = plan signature.
5. Does the pattern continue for 2+ weeks? Multi-week continuation = plan.
6. What is sharesOwnedAfter on the most recent filing? If meaningfully large, the position is intact even with continuous selling.

Three or more 'plan' signals from this checklist means the sale is almost certainly mechanical liquidity, not a directional bet against the stock.

Why This Matters for the Headline-Reading Reader

Insider-trading coverage is dominated by absolute-dollar framings: 'CEO Sells $X.' That framing is correct on the math and meaningfully wrong on the implication for any plan-driven cadence. The headline reader who knows to scan the fill profile, the daily share count, and the multi-week continuation will catch the structural framing the headline missed.

The corollary is also true: when an insider's departure from a plan cadence is the story — a sudden discretionary block, a pause during a blackout, a plan refresh with notably different parameters — that is the conviction signal. The data noise in mechanical 10b5-1 execution becomes the analytical signal precisely because deviations from it stand out.

The smart-money signal feed applies plan-vs-discretionary classification upstream so that the surfaces — Smart Alerts, insider conviction scores — only fire on the discretionary deviations, not on the plan baseline. The market news coverage applies the same framework to individual insider Form 4 write-ups. The learn library covers the rest of the Form 4 transaction-code framework in detail.