I had the mechanics of a tender offer wrong for years, and specifically I had the deadline wrong.
My mental model was a single date. The offer opens, it expires, shareholders either tendered enough shares or they did not. Reading an actual filing broke that model. The Schedule 14D-9 that Potbelly Corporation filed on 23 September 2025, in the offer from RaceTrac, sets out two clocks running at once, and the difference between them is where the risk sits.
The offer has an expiration time. That much I had right.
What I had missed is that it renews itself almost automatically. The filing allows the purchaser to extend "on one or more occasions if, at the then-scheduled Offer Expiration Time, any condition to the Offer has not been satisfied or waived, in consecutive periods of up to 10 business days each."
Ten business days, to be exact. Then another ten after that. The trigger is not a negotiation or a fresh agreement. It is simply the existence of an unsatisfied condition.
So the expiration date on the front page of an offer document is not a deadline in any meaningful sense. It is a checkpoint that recurs.
The extensions are bounded, and this is the number that matters to a seller. The purchaser may not extend "beyond the earlier to occur of (i) March 9, 2026, or, in certain circumstances, July 9, 2026, and (ii) the valid termination of the Merger Agreement."
That is the outside date. From a merger agreement signed at the start of September 2025, the first bound sits roughly six months out, with a conditional extension to roughly ten.
So there are two clocks, then. A ten-day one that keeps resetting and a six-month one that does not.
For anyone selling, the second clock is the one to negotiate hard. The first only decides how often everyone refreshes the paperwork.
The conditions are the substance. The main one is arithmetic: the shares tendered must represent "one more Share than 50% of the total number of Shares outstanding as of the expiration of the Offer at the Offer Expiration Time (the 'Minimum Condition')."
One share more than half. It is not a supermajority.
The rest of the list is what you would expect and I will not pretend it is exciting: "the absence, since the date of the Merger Agreement, of a Company Material Adverse Effect", "the accuracy of the Company's representations and warranties (subject to customary materiality qualifiers)", and the company's "compliance with and performance in all material respects of all obligations, covenants and conditions".
Payment is fast once the condition is met. The purchaser must "promptly after (and in any event, no later than the second business day after the Offer Acceptance Time) pay for such Shares."
The structure is a speed decision, and the filing says so plainly. The buyer's lawyers "indicated that RaceTrac preferred to utilize a two-step transaction structure comprised of a tender offer and a subsequent back-end merger pursuant to DGCL 251(h)."
The payoff of that section is one clause: the back-end merger is done under "251(h) of the DGCL and will be effected without a vote of the Company's stockholders."
There is no shareholder meeting. No proxy solicitation period. That is the entire appeal, and it is why a buyer in a hurry pays for structure rather than for time.
Nothing to do with the clocks. Skip ahead if you like.
The break fee was its own negotiation. The buyer's side "would require a customary non-solicitation provision and a termination fee of 3.5% of the transaction's equity value in the definitive agreement". The board pushed back, arguing a figure "would be approximately 3% of the transaction's equity value" and asking to run a post-signing market check instead.
Half a percentage point of equity value, argued over by telephone in late July. I have thought about that trade more than is reasonable, because the same conversation decides whether any rival can realistically appear later.
That was the aside. Back to the timetable.
Worth remembering, because the offer mechanics get all the attention and none of the money is decided there.
The sequence in the file: "$13.50 per share in cash" in mid-May, "$16.35 per share in cash" on 17 July, a board counter at "$18.00 per share in cash" on 29 July, and then a close on 30 July at "$17.12 per share in cash", which the buyer's bank called its "best and final" offer. The buyer's representative "conveyed that she did not believe that RaceTrac would be able to increase its offer above $17.12 per Share."
By the time the offer document existed, that argument was over. The clocks only govern delivery.
Whether the offer was extended at all, and how many times. This filing is dated 23 September 2025 and I did not follow the docket forward, so I cannot tell you how the two clocks actually ran in practice.
I also cannot tell you what "in certain circumstances" covers in the July 2026 outside date without reading the merger agreement itself, which is a separate document. The 14-D9 references the mechanism and does not restate the trigger.
And I would not generalise the 3% versus 3.5% argument into a rule. One board, one industry, one moment in the market.
Two clocks, not one. The visible date on an offer is soft and repeats in ten-business-day increments, while the outside date is the hard one and it is set months earlier in a document most people never open.
I still find it odd that the renewing clock is the one that gets printed in bold on the cover.