Dollar General remodel tracker: the national weekly rate

Dollar General remodels its stores in a rolling wave, and this page counts it. Over 2 complete weeks of first-party observation (7 September 2026 to 23 September 2026), Penny Tree recorded 785 remodel episodes at Dollar General stores across 37 states — an average of 371 newly-found remodels a week, with 219.5 a week starting their 70%-off markdown, 60 a week reaching their first $0.01 item, and 20.5 a week finishing and returning to regular prices. Right now 649 stores are mid-remodel in 32 states. The median gap from the first 70%-off merchandise to the first penny is 11.2 days. Every figure on this page is measured by this site, updates itself weekly, and is free to reuse with attribution.

How often does a Dollar General remodel start?

An average of 371 new remodels a week, measured over 2 complete weeks. The busiest single week in the record saw 650 found (week of 7 September 2026). Totals for the whole window: 785 found, 444 observed starting a 70%-off markdown, 121 observed reaching a penny, and 67 observed finishing.

Two things make that number smaller than the real one, and both are deliberate. It counts remodels this site found, not every remodel Dollar General runs: a store we have never swept cannot be counted. And it counts an episode — one store, one remodel — so a store that drops out of a sweep and returns is still one event, not two. The direction of the error is known and it only ever runs one way: the true national number is higher than the number on this page.

What are the four stages of a Dollar General remodel?

Four, and this site records the date it first observed each one:

1. Announced. Dollar General's own website begins showing a remodel banner for that store — "70% OFF | STORE REMODEL SALE | HUNDREDS OF ITEMS | THIS STORE ONLY". No discounted items are findable yet; the store has simply been flagged.
2. 70% off. Marked-down merchandise appears in that store's catalogue. This is when a trip is worth making for most shoppers.
3. Penny. The first $0.01 item appears. What did not sell at 70% off is flipped to a cent as an instruction to staff to pull it — the same mechanism as a Dollar Tree penny item.
4. Ended. The store stops showing remodel pricing and goes back to regular prices.

The median observed gap from the announcement to the first 70%-off merchandise is 5.8 days; from the first 70%-off merchandise to the first penny, 11.2 days; from the first penny to the end of the event, —. Those are medians over every episode where this site caught both ends.

Which states have the most Dollar General remodels?

TX leads the record with 115 remodel episodes, which is 5.9% of the Dollar General stores this site holds in that state. The table below is the full standing. The rate column is the honest way to read it — a big state has more remodels because it has more stores, so the percentage matters more than the count.

A remodel is not evenly spread and not seasonal in any pattern this site can yet defend. What the data does support: remodels cluster, several stores in one metro entering the cycle within days of each other, which is consistent with the crews that do the work travelling by region.

How do you know a store is remodeling before the pennies drop?

Three independent signals, in the order they usually arrive. First, Dollar General's own site targets a remodel banner at stores in the markdown window — one request per store answers "is this store remodeling?" with the company's own yes or no, and that is where the "announced" stage comes from. Second, building permits: a remodel needs a permit, and roughly three dozen municipal open-data portals publish them, so a commercial-remodel permit naming Dollar General is a warning weeks ahead of any price movement. Third, the prices themselves — a sweep of a store's catalogue showing a cluster of items at a fraction of their normal price is a remodel whether or not anything else said so.

The three disagree often enough to be worth keeping separate. A banner with no discounted merchandise is a store that has been flagged and not yet marked down; discounted merchandise with no banner is a remodel nobody announced. Both are on the board, labelled for what they are.

How is this measured? (and what the numbers do not say)

Every figure here is read from one table — a transition log with one row per stage change per store, written by the importers as they observe it. Counting rules, stated so they can be checked:

• The unit is an episode (one store, one remodel), never a row or a sweep.
• First observation only. A store that re-enters the same stage repeatedly counts once for that stage in that episode.
• "This week" is a rolling seven days, not the calendar week.
• Averages use complete weeks only. The week in progress is excluded — including it would make the pace read as a decline every Monday morning.
• Detection lag is inside the numbers. A gap of "11.2 days" is the gap between when this site saw 70%-off merchandise and when it saw the first penny, not between the two events in the store. That is the right quantity for a shopper deciding when to drive there, and it is the only one an outside observer can honestly report.

What these numbers are not: a Dollar General figure. The company does not publish remodel counts, and nothing here is sourced from it beyond the public store-level banner. Nor are they a complete census — the floor is what this site has swept, so every count below is a lower bound.

Can you use this data?

Yes — the counts, the weekly series and the per-state table on this page are published under CC BY 4.0. Quote them, chart them, or cite them in your own research; the only condition is attribution to Penny Tree with a link to this page. The page carries machine-readable Dataset metadata for the same reason, so an automated reader can tell what it is looking at.

If you are checking these figures against your own: the definitions in the method section above are the ones to match, and the observation-lag caveat is the one most likely to explain a difference.

Remodels by week, as observed

Remodel episodes by state

Browse the live penny database →