DSF2

DSF2 stands for Delivery Sequence File, 2nd Generation. It is the USPS-licensed dataset that knows the order a letter carrier physically walks every route — which house comes first, which apartment door is next, which business is at the end of the loop — and the metadata that goes with each delivery point (residential or business, seasonal or year-round, single-family or multi-drop, vacant or occupied). DSF2 processing matches a mailing list against that file and appends those attributes to every record. Without DSF2, a mailing list is just a pile of addresses; with DSF2, it is a list sorted in the order USPS will actually deliver it, with every residential and business address correctly flagged.

Why this exists at all

USPS introduced the original Delivery Sequence File in the 1990s to give mailers visibility into the carrier’s actual route order, which was the prerequisite for two things: saturation mail (where a mailer wants to hit every address on a route or carrier walk) and walk-sequence presort discounts (the deepest automation tier USPS offers, because the mail arrives already sorted in delivery order). DSF2, the second-generation rebuild, added per-record attributes that the original file didn’t carry — residential vs. business, seasonal address indicators, drop-point counts for multifamily units, and operational flags for non-personal delivery points like P.O. Boxes and centralized mail rooms. The combination is what makes both deep-discount postage and granular targeting work on the same list.

How it actually works

A licensed DSF2 processor takes the CASS-standardized list and matches every record against the USPS DSF2 file. Each record comes back enriched with: a walk-sequence number (the position in the carrier’s actual route order), a delivery-type code (residential, business, mixed-use, P.O. Box, etc.), a drop-count for multifamily and multi-business addresses (how many separate deliveries happen at that address), seasonal/year-round flags, and a centralized-delivery indicator (whether mail goes to a cluster box or door-to-door). The enriched list can then be filtered (business addresses only, residential addresses only, single-family only) or sorted by walk sequence so the mail arrives at the local USPS facility already in the order the carrier will deliver it.

Two important details. First, DSF2 licensing is restricted by USPS — not every list-cleansing provider holds the license, and the file itself can’t be redistributed. The DSF2 match has to happen inside a USPS-certified processor’s system. Second, the walk-sequence numbers carry no value on their own; they only matter when the list is being prepared for a walk-sequence presort tier or for a saturation campaign. For a standard pre-sort drop, the residential/business flag and the drop-count are the load-bearing attributes.

What goes in, what comes out

Input: a CASS-standardized list. DSF2 needs the address in USPS canonical format to do the match, so it runs after CASS but typically in the same processing pass. Output: the same list with several new fields per record — walk-sequence number, delivery type (R/B/M/P), drop count, seasonal flag, centralized-delivery flag, and any operational suppressions USPS has flagged at that delivery point. Records that don’t match the DSF2 file are reported back with a no-match flag, which usually means the address is too new, too rural, or too non-standard to be in the file.

Common pitfalls

The biggest pitfall is assuming the residential/business flag from a list broker matches what DSF2 reports. Broker-assigned R/B flags are usually heuristic — the broker guesses based on the address pattern or the data source. DSF2’s flag is the USPS carrier’s actual classification. The two disagree often enough that a B2B campaign filtered on the wrong flag can lose 15-20% of its real target audience or mail to thousands of residential addresses the broker mislabeled. The second pitfall is running DSF2 once and assuming the enrichment stays valid. Walk sequences re-number when routes are restructured, business flags change when commercial space converts to residential, drop counts change when buildings add or remove units — the file refreshes regularly and stale DSF2 data quietly degrades the targeting it’s supposed to enable. The third pitfall is treating walk-sequence data as cosmetic. If the campaign is qualifying for walk-sequence pre-sort discounts, the sortation has to honor that order all the way to the postal manifest. Sorting on something else after the DSF2 enrichment defeats the discount.

How DirectMail.io runs it

DirectMail.io runs DSF2 enrichment alongside CASS and NCOA on every list, every drop. The residential/business flag, drop counts, and walk-sequence numbers append to each record automatically, the DSF2 certification stamps the manifest, and the dashboard surfaces the R/B breakdown so a B2B campaign can be filtered against DSF2’s actual classification rather than the broker’s guess. Walk-sequence-eligible drops route into the deepest pre-sort tier the volume qualifies for without manual intervention. Details on the DSF2 feature page.

When to use this

• Every drop that wants accurate R/B targeting. B2B and consumer-only campaigns both benefit. The DSF2 flag is more accurate than any broker-supplied indicator, and the cost of mistargeting is the entire campaign performance.
• Saturation campaigns. EDDM-style and walk-sequence saturation campaigns are impossible without DSF2 walk order. The carrier’s actual delivery sequence is the input to the sort.
• Multifamily-heavy geography. Drop counts matter when the campaign needs one piece per unit vs. one piece per building. DSF2 reports the actual delivery-point count per address.
• Walk-sequence pre-sort eligibility. Drops that qualify for walk-sequence pre-sort earn deeper postage discounts than standard automation rates. DSF2 is the precondition.

For the postal stack DSF2 sits inside, see the pre-sort glossary entry.