US 5621812

I'll search for authoritative records on this patent number and any 2026 CAFC docket activity.

Let me search for the actual claim text and any 2026 Federal Circuit activity.

Let me try to retrieve the actual claim text for this specific patent.

US Patent 5,621,812 — Analyst Summary

Bottom line up front: Bibliographic data and full specification text for this patent are well-corroborated. However, I was unable to retrieve the verbatim claim set of US 5,621,812 in this session (the authoritative Google Patents copy I have is truncated in the description and does not include the claims). My plain-language claim overview below is therefore inferred from the specification and from closely related family members and must be treated as provisional, not authoritative.

1. Bibliographic data (high confidence — from Google Patents, USPTO assignment records)

Assignment chain of record: Credit Verification Corp. → INTER*ACT SYSTEMS, Inc. (1998) → Inter-Act Electronic Marketing / Catalina Marketing International, Inc. (2000) → Catalina Marketing Corporation (2006 merger). Later security interests (Morgan Stanley 2007; JPMorgan Chase and Bank of America 2014) and releases (2013, 2019) are recorded against the patent but do not change assignee of title.

Prosecution ancestry (from the family record of US 6,684,195 and US 6,307,958): Appl. 08/063,413 (now 5,621,812) is a continuation of Appl. 07/886,383, filed 1992-05-19, abandoned, which is a continuation-in-part of Appl. 07/826,255. Note the priority chain does not run solely from the 1993 filing date, which is why the 1989-05-01 date appears.

2. Abstract

Uncertain — I do not have the literal abstract text in my captured copy, and I will not fabricate it. The substantive gist, taken from the fetched specification's opening "Definitions"/summary section, is:

A check-transaction processing method and system in which a customer's checking account number (automatically read from the check's MICR line) serves as a unique customer ID. A local customer database of customer records is built and maintained at the retail store, each record identified by that ID and holding verification status (POSITIVE / NEGATIVE / CAUTION) plus transactional data such as transaction frequency and dollar volume over specified intervals (Day/Week/Month/Total). Records are updated in real time during each check transaction and via operator status input, and the database is used both for check verification/risk management and for targeted marketing and customer-profile development.

3. Plain-language overview of the independent claims

Caveat first. The claim set for 5,621,812 is not present in the authoritative text I was given, and every web search I ran for "5621812 claims" returned either (a) the '812 patents of unrelated owners (e.g., 7,690,812; 10,639,812; 10,869,812; 8,408,109 family) or (b) claim text belonging to other members of this Deaton/Gabriel family (notably US 6,684,195 claim 9, which reads: "A method for performing customer promotion with a computer having a processor and a memory … entering into the memory a customer's identification code, along with customer transaction data, at the point-of-sale; storing into the memory a database of previously entered customer identification codes and transaction data; generating with the processor a signal representative of a customer's shopping history, wherein incentives are issued to customers in dependence upon said signal; monitoring whether or not said incentives have been effective; and subsequently issuing incentives having values depending upon the results of said monitoring step." — this is claim 9 of the '195 patent, NOT of the '812 patent, and must not be attributed to '812).

Conditional overview (inferred, unverified). Based on the '812 title, its specification, and the claim architecture of sibling family members, the independent claims are most likely directed to:

1. A method of building a customer database (the lead independent claim, matching the title). Plainly: automatically reading a customer's unique identification (the checking-account/MICR-derived ID) at the point of sale; storing a database of customer records each indexed by that ID; populating each record with transactional data accumulated over time, particularly transaction frequency and dollar volume over specified intervals; and updating those records on each subsequent transaction so that the database reflects a current shopping history usable for selective incentive marketing.

2. A corresponding system/apparatus claim covering the same subject matter structurally — point-of-sale terminals, a transaction processor, a storage medium holding the customer database, and logic for retrieving/updating records and generating an incentive or marketing response from the stored shopping history.

3. Possibly one or more independent claims framed around generating a signal representative of a customer's shopping history and issuing incentives in dependence thereon, and/or around identifying new versus existing customers from the database.

I cannot tell you, without the claim text, how many independent claims there are, whether they are method-only or method-plus-apparatus, or what specific limitations define the point of novelty. Anyone needing to rely on the claims should pull the granted claim set directly from USPTO Patent Center or the patent PDF.

4. Litigation / CAFC 2026 docket check

• I found no 2026 Federal Circuit docket, appeal, or IPR entry naming US 5,621,812. Searches for "5,621,812 Federal Circuit 2026" and similar returned nothing on point; results that did mention an "'812 patent" were about unrelated patents (e.g., Provisur/Weber litigation over US 10,639,812; IPRs on 7,690,812, 10,869,812; a wellbore '812 patent).
• The only litigation signal on the authoritative record is the Darts-IP "First worldwide family litigation filed" indicator on the Google Patents page — a family-level flag with no case identifiers or dates. It does not establish any 2026 activity.
• Consistent with this, the patent is recorded as Expired – Lifetime, anticipated expiration 2014-04-15, after which new infringement assertions on post-expiration conduct are generally unavailable. A 2026 Federal Circuit matter on this patent is therefore unlikely but not impossible (e.g., a legacy case or a post-grant proceeding).

5. Confidence and uncertainty summary

If you want the definitive claim analysis, the two authoritative sources are the USPTO Patent Center record for application 08/063,413 and the granted patent PDF at https://patents.google.com/patent/[US5621812A](/patent/US5621812A)/en (which includes the claims section my captured copy omitted).

I'll search for litigation records involving US 5,621,812 specifically.

Let me drill into the specific cases the first search surfaced, to check which patents were actually asserted.

The cases so far involve other patents in the family. Let me search the NPE litigation database directly for this patent number.

Litigation Search: US Patent 5,621,812

Bottom line: I found no litigation in which US 5,621,812 is identified as an asserted (or adjudicated) patent. Every Catalina-family infringement suit I could surface turns out, on inspection of the actual court records, to assert a different patent. Below I separate what I could confirm from what I could not.

Method note / limitation: My search session hit a step limit before I could exhaustively query PACER, Darts-IP, or the Unified Patents case list. This is an absence-of-evidence finding from the sources I was able to reach (Google Patents family record, Stanford NPE Litigation Database, CourtListener/RECAP, Justia Dockets, casetext, and secondary reporting), not proof that no case exists. Treat the negative result as provisional.

1. Confirmed: no case specifically asserting '812

Important caution on the Google Patents flag. The "family has litigation" indicator attaches to the patent family, not to '812 individually. Because the Deaton/Gabriel family is large (two dozen-plus US members: 5,307,196; 5,231,620; 5,201,010; 5,327,508; 5,388,165; 5,430,644; 5,448,471; 5,592,560; 5,621,812; 5,638,457; 5,675,662; 5,687,322; 6,307,958; 6,334,108; 6,351,735; 6,377,935; 6,424,949; 6,684,195, etc.), a family-level litigation flag is very likely driven by a suit asserting a different member. It is not evidence that '812 was ever asserted.

2. Cases I found — and why none of them are '812 cases

I am listing these so you can see the near-misses explicitly rejected, which matters given your instruction not to return similar-but-wrong numbers.

Net result of the table: of the seven Catalina-side matters, not one was confirmed as asserting '812. The three where the patent is identified point to 4,674,041 and 6,240,394 — different patents, different families in the '041 case.

Also not '812:

• EPO opposition T 0506/06 (EP application 95906202.7, "Method and system for selective incentive point-of-sale marketing in response to customer shopping histories"; applicant Catalina Marketing Corp.; appeal dismissed 2009-09-22). This is a European post-grant proceeding against the EP family member (the EP counterpart of US 5,638,457 / EP 0711434), not US litigation and not '812.
• A PTAB/USPTO petition document I retrieved (ptacts.uspto.gov/ptacts/.../1524341) contains "5621812" inside a long enumerated list of patent and publication numbers. I could not determine what that list represents (a prior-art set, a challenged-patent set, or a citation list), so I am not treating it as an IPR or other proceeding naming '812. It should be pulled and read directly before anyone relies on it.

3. 2026 Federal Circuit / CAFC check

Consistent with the prior section of this analysis: no 2026 Federal Circuit docket activity naming US 5,621,812 was found. The '812 patent is recorded as Expired – Lifetime with an anticipated expiration of 2014-04-15, which makes a fresh 2026 infringement appeal on this patent highly unlikely (post-expiration conduct generally cannot be enjoined; damages are limited to pre-expiration infringement).

4. Where to verify definitively

Because my negative result is provisional, the four sources that would settle this are:

1. PACER / CourtListener RECAP — search the patent number across all district courts (CourtListener indexes "Patent Number" as a docket field, which is how I found the LDM and Blueprint cases).
2. Lex Machina or Darts-IP — the Darts-IP family link on the Google Patents page (patents.darts-ip.com/?family=27407690) is the source of the "family has litigation" flag; the family record there should identify the specific patent and case.
3. Unified Patents litigation portal (portal.unifiedpatents.com/litigation/caselist) and its patent pages — I reached a Unified patent page referencing US-5621812-A as a family/priority entry under another patent, but not as a case-listing.
4. USPTO Patent Center for application 08/063,413 — the "Proceedings" tab will show any ex parte reexam, IPR, or other post-grant proceeding; the "Patent Litigation" section surfaces district-court suits reported under 35 U.S.C. § 290.

5. Confidence summary

If you need a litigation opinion on this patent, I would not certify "no litigation exists" on the basis of my searches alone. The honest statement is: no publicly indexed case identifying US 5,621,812 as an asserted patent was found, and the Catalina-family suits that dominate search results assert other patents — but a Darts-IP family litigation flag exists that I could not resolve to a specific case, and that unresolved flag is the principal open item.

I'll verify the structured "no proceedings" default against live sources before finalizing.

The default is "no PTAB activity," but let me test it against case-specific and family-level searches before I commit.

No PTAB hits so far. Two more probes — one for reexamination/reexam certificates, one for the underlying litigation that generated the Darts-IP flag.

Proceedings overview

Total AIA trial proceedings on US 5,621,812: zero (0). Breakdown: 0 active, 0 claims invalidated, 0 claims sustained, 0 settled, 0 institution-denied — the structured USPTO Open Data Portal "PTAB proceedings on file" block returns no AIA trial proceedings for this patent, and targeted web searches surfaced no IPR, PGR, or CBM naming it. The bottom-line defensive posture is therefore unusual but favorable for a defendant: the '812 claims have never been tested at the PTAB, so there is no IPR record to cite, no FWD to lean on, and no § 315(e)(2) estoppel narrowing anyone's prior-art options — but also no cancelled claims to point to. Your primary defense is temporal (expiration), not a PTAB kill shot.

There is consequently no "most-impactful proceeding" to list first. I will not manufacture one.

No proceedings to enumerate — and why that matters more than it looks

Because the required output format is per-proceeding, the honest answer is that the loop is empty. Below is what I actually verified, so you can see the absence is not an artifact of not looking.

What I checked

Why the absence is structurally expected (three independent reasons)

1. Expiration predates the practical IPR window for this art. Google Patents records the patent as Expired – Lifetime, anticipated expiration 2014-04-15 (consistent with a pre-URAA 17-years-from-issue term running from the 1997-04-15 grant). Most post-issuance attacks on this family cluster either never happened or happened as pre-AIA reexamination-era / district-court matters; the AIA trial regime only became available on 2012-09-16, by which time the patent had ~19 months of term left.
2. CBM sunset. Even had someone wanted a Covered Business Method review — and this patent's classification profile (G06Q30/0224, G06Q30/0238, G06Q30/0255, G06Q30/0268, G06Q30/0269; G06Q20/387) is textbook CBM-eligible subject matter — the transitional CBM program ended 2020-09-16 and could not reach an expired patent in any commercially meaningful way.
3. The enforcement was pre-AIA and largely non-PTAB. The Google Patents / Darts-IP "Family has litigation" flag is a family-level indicator with no case identifiers, dates, or patent-level attribution. It does not establish an IPR and does not establish any 2026 activity. The Catalina-family enforcement I could surface in this session is claim-construction and administrative litigation (the Coolsavings appeal; Catalina Marketing Corp. v. Kappos, E.D. Va. 1:11cv659, patent-term-adjustment challenge), not a PTAB trial on '812.

Flag — contradiction check against prior sections: none found. The previously generated summary's PTAB finding ("no AIA trial proceedings; absence of evidence, not proof of absence") and this section agree. The prior section's open item — verbatim claim text not retrieved — does not hurt us here, because with zero proceedings there are no claim-level dispositions to map. (It would matter if you later need to compare the '812 claim set against sibling claims that were litigated.)

Defensive value: with no FWD, no cancellation, and no estoppel, the '812 patent is neither hardened nor dead at the PTAB — it is simply untested there. For a defendant today, the operative question is not "which claims died in IPR" but "can this expired patent support any live claim at all."

Strategic summary

Claim status. There are no PTAB-cancelled claims and no PTAB-sustained claims in US 5,621,812 — every claim is, in PTAB terms, untested. I cannot give you a cancelled/sustained/untested table by claim number because (a) no proceeding exists to have produced such dispositions and (b) the verbatim claim set was not retrieved in the prior section. Anyone asserting the '812 patent is asserting claims that no administrative tribunal has construed or invalidated. Note the practical limit on that untested status: the patent is recorded expired as of 2014-04-15, so the claims cannot support prospective relief and any damages theory faces the § 286 six-year lookback measured backward from a complaint filed today — a window that opens well after the patent's expiration. That is a dispositive timing problem for a plaintiff, independent of validity.

Estoppel landscape. There is no § 315(e)(2) estoppel running against anyone with respect to the '812 patent, because no petitioner has ever obtained an IPR/PGR FWD on it. Every § 102/§ 103 ground based on patents and printed publications remains available to a defendant. Two caveats worth stating precisely: (i) § 315(e)(2) estoppel is patent-specific ("with respect to that patent"), so an IPR on a sibling family member (e.g., the '322, '473, '857, '662, '732, '114, '485, '195, '949, '735, '7935, '7958 members) would not estop grounds against '812 — though a petitioner on a sibling should still be careful about inconsistent positions and about which document's claims are being construed; and (ii) if you do file an IPR-style challenge and lose, you inherit estoppel — but note that IPR of an already-expired patent is generally of little value, since the practical benefit (prospective freedom) is gone.

Pattern signals. No repeat petitioner, because there is no petitioner at all. No PTAB appeals by the patent owner on '812. No evidence of a defensive aggregator (Unified Patents) challenge — Unified's portal surfaces '812 only as a cited/family reference in other patents' pages, not as a challenged patent. Property title passed Credit Verification → INTER*ACT SYSTEMS (1998) → Inter-Act Electronic Marketing / Catalina Marketing International (2000) → Catalina Marketing Corporation (2006 merger), with later secured-party liens (Morgan Stanley 2007; JPMorgan Chase and Bank of America 2014) and releases (2013, 2019) that do not change title. The current owner-facing entity is Catalina Marketing; note the unrelated Catalina Marketing Chapter 11 activity appearing in the records I surfaced (e.g., Pitta v. Catalina Marketing Corporation, Bankr. D.N.J. adv. proc. 3:25-ap-01585, filed 2025) — that is a preference/fraudulent-transfer adversary proceeding, not a patent matter, but it is a standing/ownership and collectability consideration if someone is asserting the '812 patent and pointing at Catalina.

Recommended next steps

1. There is no FWD to link to and I will not invent one. If you were expecting a claims-invalidated proceeding on this patent, the expectation is not supported: the ODP record and web checks both come up empty. State plainly, in any invalidity opinion or demand-letter response, that US 5,621,812 has no PTAB history — absence of PTAB activity on a once-heavily-asserted family is itself a signal that the family's enforcement window closed before the AIA trial era matured.
2. Lead with expiration, not invalidity. Confirm the 2014-04-15 expiration independently — the authoritative sources are the USPTO Patent Center record for application 08/063,413 (https://patentcenter.uspto.gov) and the granted PDF at https://patents.google.com/patent/[US5621812A](/patent/US5621812A)/en — and check the maintenance-fee history and any terminal disclaimer effect on the family. If your client's accused conduct postdates 2014-04-15, the § 286 damages window may be empty.
3. Close the claim-text gap before doing any claim-level work. The verbatim claim set was never retrieved. Pull it from Patent Center or the patent PDF; a claim-level validity or infringement analysis on an expired patent is only worth doing to defeat a pre-expiration damages theory.
4. Verify the Darts-IP family litigation flag at the patent level. The Google Patents flag (https://patents.google.com/patent/US5621812A/en, "Family has litigation") identifies no case. Pull the Darts-IP family record (family 27407690) or run the patent number through PACER/CourtListener (https://www.courtlistener.com) and Docket Navigator to identify which family member was asserted, by whom, and when — do not import sibling-patent outcomes onto '812.
5. Confirm the continued absence of PTAB activity via PTAB E2E (https://ptacts.uspto.gov/ptacts/) and the USPTO Patent Trial and Appeal Board decision search; my conclusion rests on the ODP block plus bounded web checks, and I hit a tool-call ceiling before exhausting every query permutation. Absence of evidence remains absence of evidence.

Confidence: High that no AIA trial proceeding exists on US 5,621,812 as of the ODP ingest and my search coverage. Medium-high on the reasons (expiration timing, CBM sunset, pre-AIA enforcement era) — these are inferences from the expiration date and the classification/enforcement record, not documents establishing motive. Not verified in this session: the specific patent(s) at issue in Catalina Mktg. Int'l v. Coolsavings.com, 289 F.3d 801 (Fed. Cir. 2002) — I believe the '812 and/or its siblings were involved, but I could not confirm which, so treat that attribution as unverified.

I'll search for the assignment records and ownership-chain details for this patent.

Let me try to pin down reel/frame numbers and correspondents from the assignment records.

Interesting — I found a lawsuit reference. Let me dig into that and the specific assignment records.

US 5,621,812 — Ownership-Chain Reconstruction

Scope note before the analysis. I was able to reconstruct the events in the chain from the Google Patents legal-events record (which mirrors the USPTO assignment database) and from press/litigation sources. I was not able to retrieve the reel/frame numbers or the correspondents of record for this patent in this session — no tool call I made returned a reel/frame pair keyed to 5,621,812, and every "REEL/FRAME" hit I got belonged to sibling Catalina patents, not to '812. Accordingly, the per-event "Reel" and "Correspondent" fields below are marked [NOT RETRIEVED] rather than guessed. Anyone needing those fields should query https://assignmentcenter.uspto.gov/ by patent number 5621812 directly.

Inventors

Neither inventor is a "departing inventor" in the classic fire-sale sense — the patent issued to their own employer, Credit Verification Corp., in 1997, and the two inventors are listed as assignors on corporate assignments, not as free agents.

Unusual patterns worth flagging (evidence-based):

1. Eighteen-year-later inventor assignments. Both named inventors re-appear as assignors in the assignment record on 2007-09-25 (Deaton) and 2007-09-27 (Gabriel), conveying their interest back to Credit Verification Corporation — the very entity that had already assigned the portfolio away in 1998. This is the signature of a chain-of-title curative / confirmatory assignment, not a new sale. It is recorded roughly three weeks before Catalina granted Morgan Stanley a portfolio security interest (2007-10-15), which is exactly the factual sequence you'd expect if a lender's IP diligence surfaced a title gap. I flag the timing correlation as an inference; I have not seen the assignment instruments and cannot confirm the curative purpose.
2. A non-inventor appears as assignor. The 2007-09-25 record names "Deaton, David W.; Wood, Robert Scott." Robert Scott Wood is not a named inventor on US 5,621,812. He may be a co-inventor on a sibling family member, an heir/successor-in-interest, or an officer holding rights. Unresolved — flagging rather than guessing.

Original assignee

Credit Verification Corporation (Abilene, TX) — named as assignee on the face of the issued patent.

• Primary line of business: check verification / returned-check recovery services for retail grocery — i.e., the exact subject matter of the patent (local customer database, MICR-derived customer ID, POSITIVE/CAUTION/NEGATIVE status, DWT frequency-and-dollar thresholds). This is a service business that commercialized the claims, not a holding vehicle.
• Product embodying the claims: the patent describes a deployed check-verification system (POS terminals, MICR check readers, in-store transaction processor, host/remote global update). The specification's release-level engineering detail (Western Digital WD286 board, Seagate ST225, Sealevel SIO-485, MS-DOS 3.3, C-Tree) is strong evidence of an actual shipped implementation.
• Current status: Not determinable from sources in this session. Credit Verification Corporation is not a currently operating brand I can evidence; it survives in the record only as an assignment counterparty (1998 assignor; 2007 assignee of the curative assignments). It is not on any public NPE asserter list I checked. I will not assert "dissolved" without a corporate record.

Assignment timeline

Dates below are as carried in the Google Patents legal-events (assignment) record. Google Patents does not always distinguish execution date from recording date; I have not independently verified which is which, so I label them as recorded events.

1998–2002 — Original seller and the Inter-Act acquisitions

• 1998-05-27 — recorded 1998-05-27 — Reel [NOT RETRIEVED]

  • Conveyance: Assignment (Assignment of Assignors' Interest)
  • Assignor: Credit Verification Corporation
  • Assignee: INTER*ACT SYSTEMS, INCORPORATED
  • Correspondent: [NOT RETRIEVED]
  • Context: Acquisition / portfolio transfer — the original operating assignee sells the Deaton check-verification portfolio to Inter-Act Systems.

• 1999-06-08 — recorded 1999-06-08 — Reel [NOT RETRIEVED]

  • Conveyance: Collateral Assignment of Security Interest
  • Assignor: IQ. COMMERCE CORPORATION
  • Assignee: COMDISCO, INC.
  • Correspondent: [NOT RETRIEVED]
  • Context: Securitization — venture-debt collateral. Note the assignor is IQ.Commerce Corporation, not Inter-Act Systems; the corporate-name tangle in this chain is discussed below.

• 2000-08-01 — recorded 2000-08-01 — Reel [NOT RETRIEVED]

  • Conveyance: Security Agreement
  • Assignor: INTER ACT ELECTRONIC MARKETING, INC.
  • Assignee: FIRST GREENWICH CAPITAL, LLC
  • Correspondent: [NOT RETRIEVED]
  • Context: Securitization — second lienholder layered onto the same assets.

• 2000-08-28 — recorded 2000-08-28 — Reel [NOT RETRIEVED]

  • Conveyance: Release
  • Assignor: FIRST GREENWICH CAPITAL, LLC
  • Assignee: INTER ACT ELECTRONIC MARKETING, INC.
  • Correspondent: [NOT RETRIEVED]
  • Context: Release of the 2000-08-01 security interest — 27 days later. A lender taking a lien and releasing it within a month on the eve of an acquisition is consistent with a bridge/consent financing, not a restructuring.

• 2000-11-20 — recorded 2000-11-20 — Reel [NOT RETRIEVED]

  • Conveyance: Release of Lien
  • Assignor: COMDISCO, INC.
  • Assignee: INTER-ACT ELECTRONIC MARKETING
  • Correspondent: [NOT RETRIEVED]
  • Context: Release of the 1999-06-08 Comdisco collateral assignment, recorded the same day the portfolio is sold to Catalina — i.e., lien clean-up as a closing condition.

• 2000-11-20 — recorded 2000-11-20 — Reel [NOT RETRIEVED]

  • Conveyance: Assignment (Assignment of Assignors' Interest)
  • Assignor: INTER-ACT ELECTRONIC MARKETING, INC. (previously named Interact Systems, Inc.)
  • Assignee: CATALINA MARKETING INTERNATIONAL, INC.
  • Correspondent: [NOT RETRIEVED]
  • Context: Acquisition (transfer-to-acquirer). The record's parenthetical "(previously named Interact Systems, Inc.)" is important: Inter-Act Electronic Marketing is a name change, not a separate shell — it is the same 1998 assignee.

• 2002-09-26 — recorded 2002-09-26 — Reel [NOT RETRIEVED]

  • Conveyance: Release
  • Assignor: SILICON VALLEY BANK
  • Assignee: IQ.COMMERCE CORPORATION
  • Correspondent: [NOT RETRIEVED]
  • Context: Release of a security interest recorded against the '812 record even though the surviving party is IQ.Commerce, not Catalina — a likely portfolio-wide release carried onto this patent's record.

2006–2019 — Catalina era: merger, curative assignments, and three rounds of liens

• 2006-06-27 — recorded 2006-06-27 — Reel [NOT RETRIEVED]

  • Conveyance: Merger
  • Assignor: CATALINA MARKETING INTERNATIONAL, INC.
  • Assignee: CATALINA MARKETING CORPORATION
  • Correspondent: [NOT RETRIEVED]
  • Context: Internal reorganization / change of name only — the parent corporation absorbs the international subsidiary.

• 2007-09-25 — recorded 2007-09-25 — Reel [NOT RETRIEVED]

  • Conveyance: Assignment (Assignment of Assignors' Interest)
  • Assignors: DEATON, DAVID W.; WOOD, ROBERT SCOTT
  • Assignee: CREDIT VERIFICATION CORPORATION
  • Correspondent: [NOT RETRIEVED]
  • Context: Chain-of-title curative assignment — individual rights-holders confirming title back to the original assignee's name, 9 years after that entity had assigned the patent away.

• 2007-09-27 — recorded 2007-09-27 — Reel [NOT RETRIEVED]

  • Conveyance: Assignment (Assignment of Assignors' Interest)
  • Assignor: GABRIEL, RODNEY G.
  • Assignee: CREDIT VERIFICATION CORPORATION
  • Correspondent: [NOT RETRIEVED]
  • Context: Second curative assignment, capturing the other named inventor two days later — the two-step sequencing strongly suggests missing-source-document clean-up.

• 2007-10-15 — recorded 2007-10-15 — Reel [NOT RETRIEVED]

  • Conveyance: Security Agreement
  • Assignors: CATALINA HEALTH RESOURCE, LLC; CATALINA MARKETING PROCUREMENT, LLC; CATALINA MARKETING WORLDWIDE, LLC; CATALINA-PACIFIC MEDIA, LLC; CHECKOUT ACQUISITION CORP.; CHECKOUT HOLDING CORP.; CMJ INVESTMENTS LLC
  • Assignee: MORGAN STANLEY & CO. INCORPORATED
  • Correspondent: [NOT RETRIEVED]
  • Context: Securitization — a whole-portfolio lien granted three weeks after the inventors' curative assignments. This is the single strongest sequencing tell in the chain.

• 2013-10-24 — recorded 2013-10-24 — Reel [NOT RETRIEVED]

  • Conveyance: Release by Secured Party
  • Assignor: MORGAN STANLEY & CO. LLC (fka Morgan Stanley & Co. Incorporated)
  • Assignees: CATALINA HEALTH RESOURCE, LLC; CATALINA MARKETING PROCUREMENT, LLC; CATALINA MARKETING WORLDWIDE, LLC; CATALINA-PACIFIC MEDIA, LLC; CMJ INVESTMENTS, LLC
  • Correspondent: [NOT RETRIEVED]
  • Context: Release, ahead of the 2014 refinancing.

• 2014-04-11 — recorded 2014-04-11 — Reel [NOT RETRIEVED]

  • Conveyance: First Lien Patent Security Agreement
  • Assignors: CATALINA MARKETING CORPORATION; MODIV MEDIA, INC.
  • Assignee: JPMORGAN CHASE BANK, N.A., as administrative agent
  • Correspondent: [NOT RETRIEVED]
  • Context: Securitization — refinancing of the Morgan Stanley facility.

• 2014-04-14 — recorded 2014-04-14 — Reel [NOT RETRIEVED]

  • Conveyance: Second Lien Patent Security Agreement
  • Assignors: CATALINA MARKETING CORPORATION; MODIV MEDIA, INC.
  • Assignee: BANK OF AMERICA, N.A., as administrative agent
  • Correspondent: [NOT RETRIEVED]
  • Context: Securitization — second-lien tranche of the same refinancing. Note the patent's statutory expiration is 2014-04-15, one day after this filing.

• 2019-02-20 — recorded 2019-02-20 — Reel [NOT RETRIEVED]

  • Conveyance: Release of Security Interest in Patents
  • Assignor: JPMORGAN CHASE BANK, N.A., as agent
  • Assignees: CATALINA MARKETING CORPORATION; MODIV MEDIA, INC.
  • Correspondent: [NOT RETRIEVED]
  • Context: Release — final lien discharge. No assignment of title has been recorded after this date.

Reel/frame leads I could not tie to '812 (do not rely on these as '812 records)

The same portfolio-level security instruments recur on sibling Catalina patents at these reel/frame pairs, which is circumstantial evidence that '812 appears on the same reels — but I could not confirm that:

Source for these pairs: legal-events tables on sibling Catalina publications (e.g., https://patents.google.com/patent/US20070103993A1/en#legal-events). Treat as leads only.

Timeline diagram

timeline
    title Ownership of US 5621812
    1997 : Issued to Credit Verification Corp
    1998 : Assigned to Inter Act Systems
    1999 : Comdisco collateral assignment
    2000 : Catalina acquires Deaton portfolio
    2006 : Merger into Catalina Marketing
    2007 : Inventor curative assignments
         : Morgan Stanley security agreement
    2013 : Morgan Stanley lien released
    2014 : JPMorgan and BofA security agreements
         : Patent expires
    2019 : JPMorgan lien released

NPE / troll-pattern signals

1. Shell-entity transfer — NOT PRESENT.
The chain runs Credit Verification Corp. → Inter*Act Systems, Inc. → Inter-Act Electronic Marketing, Inc. → Catalina Marketing International → Catalina Marketing Corporation. Every named entity is a corporate operating business, and the record itself tells us the middle link is a name change ("previously named Interact Systems, Inc.", recorded 2000-11-20), not a new shell. No "IP / Holdings / Licensing / Ventures" entity appears in the chain of title; the only LLCs in the record (Catalina Health Resource, Catalina Marketing Procurement, Catalina Marketing Worldwide, Catalina-Pacific Media, CMJ Investments) are grantors under a security agreement, not assignees of title.

2. Known asserter in the chain — NOT PRESENT (as to public NPE lists).
I could not match any assignee in this chain to Acacia, Marathon, IV, IPNav, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation Corp, or Spangenberg entities. Catalina Marketing Corporation is an operating company — it operates the in-store checkout-coupon network referenced in contemporaneous trade press (16,000+ supermarkets). Being a frequent patent enforcer is not the same as being a listed NPE; I am not overreading this.

3. Repeat correspondent across the chain — UNCLEAR / INSUFFICIENT DATA.
This is the field the task flags as the highest-value tell, and it is precisely the field I could not populate. Correspondents of record were not retrievable for any of the 15 recorded events. I will not invent attorney names. Recommendation: pull correspondents from Assignment Center for all 15 reels and diff them against Unified Patents / RPX asserter directories.

4. Cascading transfers — PRESENT but BENIGN.
There are five recorded events between 1998-05-27 and 2000-11-20 (two assignments, two lien actions, one lien release), which fits the <24-month pattern on its face. However, the substance rebuts the NPE reading: the 1999 Comdisco collateral assignment and the 2000 First Greenwich security agreement are financing instruments, both released (2000-08-28 and 2000-11-20), and the terminal event is a sale to a strategic acquirer with a real business. This is a venture-financed start-up's cap-table paper trail, not shell-name rotation.

5. Pre-litigation transfer — UNCLEAR.
There is documented litigation: Inter-Act Systems, Inc. (Norwalk, CT) sued Catalina Marketing Corp. in the District of Connecticut — an initial 1996 suit over an in-store, paper-based couponing patent, plus a later suit asserting eight U.S. patents from the acquired "Deaton" portfolio (https://www.chiefmarketer.com/the-patent-chase-continues/). Separately, Catalina publicly announced US 6,424,949 in 2002 and warned the trade that infringers should seek licenses (https://www.supermarketnews.com/grocery-operations/catalina-s-latest-patent-evokes-surprise-and-concern). But the direction of the Inter-Act/Catalina dispute cuts against a clean pre-litigation-transfer read — the plaintiff appears to be the prior assignee, and Catalina the current owner, which is chronologically awkward (Inter-Act assigned the portfolio to Catalina in 2000). I cannot reconcile the entity identities from the sources available and will not force a conclusion. Unclear.

6. Bankruptcy fire-sale — NOT PRESENT (no evidence).
No Chapter 7/11 is evidenced for Credit Verification Corporation in any source I retrieved. The 2014 first- and second-lien security agreements and their 2019 release are consistent with a leveraged capitalization. I am aware of unverified public reporting about later Catalina restructuring but could not confirm a filing in this session, so I do not score it.

7. Privateering — NOT PRESENT.
No operating company is shown transferring to an NPE to assert against its own competitors. Catalina asserted its portfolio in its own name, as an operating company does.

8. Defensive aggregator — NOT PRESENT.
The chain does not terminate at RPX, AST, LOT, Unified Patents, or OIN. It terminates at Catalina Marketing Corporation (with Modiv Media, Inc. appearing as co-grantor on the 2014/2019 records). The patent is Expired – Lifetime, anticipated expiration 2014-04-15, so it is neutralized by expiry, not by defensive aggregation.

Verdict

Operating-company assertion.

The '812 chain is an operating-company chain end to end: Credit Verification Corp. (the entity that built the POS check-verification system the claims describe) sold to Inter*Act Systems (1998-05-27), which was renamed and then acquired by Catalina Marketing International (2000-11-20), which merged into Catalina Marketing Corporation (2006-06-27) — the current assignee of record, an operating company that ran the checkout-coupon network the patent family covers. The heavy lien traffic (Morgan Stanley 2007-10-15; JPMorgan first lien 2014-04-11; Bank of America second lien 2014-04-14; releases 2013-10-24 and 2019-02-20) is securitization of a real business's IP collateral, not NPE shell rotation, and no assignment of title is recorded after 2019-02-20.

The one genuinely notable artifact is the 2007-09-25 / 2007-09-27 pair of individual assignments back to Credit Verification Corporation — Deaton, Wood, and Gabriel conveying rights to an entity that had assigned the patent away nine years earlier, three weeks before the Morgan Stanley lien. That is a title-clean-up pattern, and it is the item I would most want the reel/frame documents for. It is consistent with prosecution/financing diligence, and not sufficient on its own to support an NPE characterization.

Correction / consistency note relative to the earlier-generated section: the prior summary's assignment chain ("Credit Verification Corp. → INTER*ACT SYSTEMS (1998) → Inter-Act Electronic Marketing / Catalina Marketing International (2000) → Catalina Marketing Corporation (2006 merger)," with security interests in 2007, 2014 and releases in 2013, 2019) matches this reconstruction exactly — no contradiction. The prior summary's caveat that the verbatim claim set was not retrieved also stands; the sibling-patent claim language quoted in my searches (e.g., US 6,684,195; US 6,351,735) belongs to other family members and should not be attributed to '812.

Primary verification link: https://assignmentcenter.uspto.gov/ — search patent number 5621812. What to look for there that I could not get: (a) the 15 reel/frame pairs, (b) the correspondent of record on each, and (c) executed-vs-recorded dates. If the same correspondent appears across the 1998, 2000, 2007 and 2014 entries, that would be the one finding capable of changing this verdict.

I'll search for the authoritative prior-art citations for US 5,621,812.

Let me pull the front-page citation list specifically.

I have the family/prosecution data but not the '812 front-page citation list. Let me go directly to the patent PDF and to FreePatentsOnline/Justia.

Prior-Art Analysis — US 5,621,812 (Appl. 08/063,413)

0. Scope, and an important disclosure about this search

I searched for the patent number 5621812 specifically. Two things must be stated up front, because they materially limit what I can give you:

1. I could not retrieve the front-page "(56) References Cited" list of US 5,621,812. The authoritative full text I was given stops in the middle of the specification (Section 3.7, the status-roll subroutine) and contains no front page, no (56) section, and no claims. Every search I ran returned (a) the same truncated Google Patents description, (b) forward citations (later patents citing '812), or (c) same-family Deaton/Gabriel patents. None returned the '812's own examiner/applicant citation list.
2. I therefore will not fabricate a (56) list. Below I give (i) the references that are literally cited inside the '812 specification text — these are verifiable from the authoritative document and are "patent citations" appearing in '812 — and (ii) an explicit list of look-alike references that are not prior art against '812 and are commonly mis-cited as such.

Per my operating rules I have not auto-corrected any identifier. Where search results show probable OCR corruption (e.g., a front page reading "6,307,938 B1 ... Deaton et al." alongside the correct "6,307,958"), I flag it rather than silently fixing it.

1. Grounded references — patents cited within the '812 specification (high confidence as to citation; dates partly inferred)

The '812 specification states:

"Automatic check readers have been heretofore known, and the descriptions of such previously developed automatic check readers are found in U.S. Pat. Nos. 4,277,689; 4,143,355; 4,396,902 and 5,054,092, the subject matter of which is incorporated by reference herein."
— US 5,621,812, specification (automatic check reader section), https://patents.google.com/patent/[US5621812A](/patent/US5621812A)/en

Non-patent literature cited in the '812 text:

• The MICR Handbook, Rylla R. Goldberg, Heath Printers — expressly "incorporated by reference" in the '812 specification for MICR encoding. Relevant under § 102(b) only if published more than one year before the critical date; that date is not verified here.
• Copending application Ser. No. 07/826,255 — expressly identified in '812 as disclosing use of a checking-account number as a unique customer ID for credit verification and marketing. This is the priority ancestor, not prior art (the chain runs 07/345,475 (1989, abandoned) → 07/826,255 (1992, abandoned) → 07/886,383 → 08/063,413 = '812). It is an inventor's own earlier application and cannot be § 102 art against '812.

2. References that are not prior art against '812 (flagging a common error)

Many search results surface these numbers next to '812. They are same-family Deaton/Gabriel patents or forward citations, not § 102 art:

Family/prosecution provenance (authoritative, from the US 6,684,195 official gazette record):

"Application 08/096921 is a continuation in part of application No. 08/063413, filed on May 17, 1993, granted, now 5621812 ... Application 08/063413 is a continuation of application No. 07/886383, filed on May 19, 1992, abandoned. Application 07/886383 is a continuation in part of application No. 07/826255, filed on Jan. 24, 1992, abandoned. Application 07/826255 is a continuation of application No. 07/345475, filed on May 01, 1989, abandoned."
— https://webapp1.dlib.indiana.edu/virtual_disk_library/index.cgi/[5628977](/patent/5628977)/FID3294/OG/html/1278-4/us06684195-20040127.html

⚠️ Contradiction to flag: my earlier summary treated the priority date as "1989-05-01, medium confidence." The official gazette lineage above confirms the earliest claimed benefit runs to 07/345,475 filed 1989-05-01, but the '812 issues from a CIP-of-a-CIP chain, so different claim limitations carry different effective dates. That is a legal conclusion I cannot draw without the claim set.

3. Conditional § 102 mapping (with the caveat that governs the whole exercise)

Anticipation requires a single reference disclosing every limitation of a claim, arranged as in the claim. I cannot perform that test definitively because the '812 claim set is not in the authoritative text available to me (the same limitation noted in the prior generated section — it remains unresolved and is not cured by these searches).

Conditional conclusions, stated as analytical judgments rather than findings:

• The four check-reader patents (4,277,689; 4,143,355; 4,396,902; 5,054,092) cannot anticipate the independent claims of '812 as titled. Per its title and specification, '812 is directed to building a customer database for selective incentive marketing — storing records indexed by customer ID, accumulating frequency/dollar-volume data over intervals, and generating marketing responses. A check-reader hardware reference lacks those limitations. Expect these to matter, if at all, only against narrow dependent claims concerning the automatic reader/MICR parsing, and even then more likely under § 103 in combination than § 102.
• US 5,054,092 is the strongest § 102(e) candidate among the four because it issued in 1991, inside the window between the assumed 1989 priority date and the 1993 filing. Its status depends entirely on its own filing date versus the '812's effective date for the parsing subject matter.
• The most likely true § 102 art against '812 is not in the group I retrieved at all. For a 1993-filed application with 1989 lineage, the dangerous art is 1980s point-of-sale/coupon/loyalty systems — e.g., checkout-coupon dispensing and retail scanner-data marketing patents. I did not retrieve '812's (56) list, so I cannot state which of those the examiner actually cited, and I will not guess at numbers.

4. A lead worth pursuing (not a § 102 finding)

A Board of Patent Appeals and Interferences decision from 2000 captioned "DEATON et al." appeared in results — https://law.onecle.com/board-of-patent-appeals/2000/fd10460112-13-[2000001](/patent/2000001).html. An interference involving the Deaton/Gabriel applications would be a productive source: interference records identify the competing prior art/claims and the junior/senior party positions, which is frequently the best window into what art the examiner and the applicants actually fought over. I could not retrieve the decision's substance in this session, so I make no assertion about what it holds.

5. What to pull to complete this analysis (authoritative sources)

1. USPTO Patent Center, application 08/063,413 — the complete file wrapper, including the examiner's PTO-892/1449 forms (the true (56) list) and the claims as allowed.
2. Granted patent PDF, https://patents.google.com/patent/US5621812A/en (the PDF, not the HTML description) — front page "(56) References Cited" plus the 1997 claim set.
3. USPTO Assignment/Patent Full-Text (PatFT) for the four cited reader patents to verify their filing/issue dates, which is what the § 102(e) analysis turns on.
4. Darts-IP family link at https://patents.darts-ip.com/?family=27407690 (Google Patents flags "Family has litigation" with no case data) — to identify the actual asserted art.

6. Confidence summary

Bottom line: I can give you four grounded patent citations and one NPL citation that appear in the '812 document itself, plus a cleared-out list of numbers that are not '812 prior art. I cannot give you '812's examiner-cited (56) list or a limitation-by-limitation § 102 mapping, because neither the front page nor the claims of '812 were retrievable from the sources available to me in this session. I would rather tell you that than hand you an invented citation list.

I'll ground the obviousness analysis in actual prior-art records. Let me pull the reference lists and related family art.

Let me pull the face-of-patent references and the actual claim text.

Now let me try to obtain the actual claim set and the cited references.

I hit the tool-call limit mid-research. Here is the analysis based on what I was able to verify, with explicit flags where the record is thin.

Obviousness Analysis — US 5,621,812 under 35 U.S.C. § 103

0. Threshold caveat (read first)

Two evidentiary gaps constrain everything below, and I will not paper over them:

1. I still could not retrieve the verbatim claim set of US 5,621,812. The authoritative copy of https://patents.google.com/patent/[US5621812A](/patent/US5621812A)/en supplied to me is truncated in the description (it cuts off at "...status is rolled to an allowable status value, or is not rolled, in accord…") and contains no claims. My limitation-by-limitation mapping is therefore built from (a) the title, (b) the specification's own summary of "Important aspects of the present invention," and (c) the claim architecture of the family member US 6,684,195 (38 claims; claim 9 quoted in the Official Gazette record). Any § 103 conclusion is provisional until the granted claims are pulled from USPTO Patent Center for application 08/063,413.

2. The "Prior Art section" of the Google Patents page contains no references. Structurally it contains only:

   • Prior art keywords: customer, check, data, status, store
   • Prior art date: 1989-05-01

   That keyword string is itself analytically useful — it is Google's (and effectively the classification system's) own characterization of the claimed subject matter as the combination of five known retail data elements. There is no examiner citation list, no IDS, and no "References Cited" block in my captured copy. So the § 103 case below necessarily rests on (i) the prior art admitted on the face of the specification, and (ii) candidate secondary references that I identify with confidence flags.

1. The presumptive prior-art date and what art is available

The Google Patents field states a prior art date of 1989-05-01, which corresponds to the filing of parent application 07/345,475 (1989-05-01), confirmed by the § 120 chain recited in the Official Gazette entry for US 6,684,195:

"Application 08/063413 is a continuation of application No. 07/886383, filed on May 19, 1992… Application 07/886383 is a continuation in part of application No. 07/826255… Application 07/826255 is a continuation of application No. 07/345475, filed on May 01, 1989, abandoned."
— https://webapp1.dlib.indiana.edu/virtual_disk_library/index.cgi/[5628977](/patent/5628977)/FID3294/OG/html/1278-4/us06684195-20040127.html

Consequences for the § 103 posture:

• This is a pre-AIA patent (filed 1993-05-17). § 102/§ 103 and pre-AIA § 103(c) (common ownership) govern.
• Art before 1989-05-01 is clean § 102(a)/(b)/102(e) art. Art in the window 1989-05-01 → 1992-05-17 is available only as § 102(a) (prior invention by another) or § 102(e) (earlier-filed, later-issued U.S. patents/applications), not as a § 102(b) statutory bar — the applicant can swear behind with the 1989 priority chain.
• Critical trap: the large Deaton/Gabriel sibling family — US 5,201,010 ("Method and system for building a database and performing marketing based upon prior shopping history," 1993-04-06, per https://www.freepatentsonline.com/[7853630](/patent/7853630).html), 5,237,620, 5,305,196, 5,327,508, 5,388,165, 5,430,644, 5,448,471, 5,592,560, 5,638,457, 5,642,485, 5,644,723, 5,659,469, 5,675,662, 5,687,322, and 5,649,114 — is NOT usable as prior art against '812. These share inventors, share the same § 120 priority chain, and were commonly owned by Credit Verification Corp./Catalina (the WO family lists confirm the shared ancestry: https://patentimages.storage.googleapis.com/52/89/c6/89e769487772f8/WO1999066450A1.pdf). They are disqualified by § 102(e)/§ 103(c) and by identity of inventive entity. Likewise, the specification's own reference to "Copending patent application Ser. No. 07/826,255" (which "discloses a system and technique wherein a customer's checking account number may be used as a unique customer identification number to provide credit verification and also to perform marketing functions") is a family member, not prior art — though it is devastating as an admission of what the inventors considered known.

2. The admitted prior art (highest-confidence § 103 foundation)

The specification's "Definitions" section expressly concedes the following as existing:

(A) Remote telephone-line check-verification systems. "Prior credit verification systems require connecting a point-of-sale terminal through telephone lines to a remote transaction processing system…"

(B) A PC-based returned-check system with POS response distribution, same store and across stores. "This PC stores information on that check (name, address, dollar amount of the check, reason for the return of the check, etc.) and this PC can be programmed to transfer that data to other processors controlling point-of-sale keypad terminals, both in the same and in other store-based operations. Responses displayed by one of these point-of-sale terminals may be altered pursuant to these transfers of data."

(C) A multi-tasking processor doing returned-check entry and POS keypad operation with inter-store telephone transfer. "…the same processor simultaneously manages the operations of returned check entry and point-of-sale keypad operation. This multiple tasking processor can be programmed to transfer data to other similar store-based operations by telephone communications."

(D) Automatic check readers. U.S. Pat. Nos. 4,277,689; 4,143,355; 4,396,902; 5,054,092, plus The MICR Handbook (Rylla R. Goldberg) — all incorporated by reference into '812.

(E) Account-number-as-customer-ID. Copending 07/826,255 (above).

(F) Internal multi-tasking software architecture. The spec concedes the System Kernal is "not critical to this invention, being a matter of routine design specification," and that it is implemented with the off-the-shelf program Time Slicer (Life Boat Systems); the database layer with C-Tree (Faircom Software); the OS is MS-DOS 3.3; the hardware is a Western Digital WD286-WDM2 / Intel 80286 board with a Seagate ST225 disk. All of this is admitted routine design choice.

Standardization of the hardware/OS software stack matters a great deal to § 103: a claim element reciting "a transaction processor coupled to a disk system storing a customer database" cannot carry patentable weight when the patent itself calls the implementation "a matter of routine design selection."

3. Candidate secondary prior art (verify before relying)

These are the references a defendant or petitioner would reach for. I am flagging confidence because I could not complete verification of publication dates within this session.

I will not assert specific column/line citations for any of these, because I did not retrieve their full texts. Do not quote the table above as verified. The right next step is to pull each from Google Patents and confirm number → inventor → date → subject.

4. The § 103 combinations

Combination 1 — Admitted art (A) + (B) + (C): check verification + distributed POS responses + multi-tasking/inter-store transfer

What it renders obvious: The core architectural claim — POS terminals, a transaction processor, a local customer database of check records, requests transmitted from the POS, a response returned and displayed at the POS, and periodic transfer of selected records to/from other stores.

Why a POSITA would combine: (B) and (C) are the same system the inventors distinguish from, differing only in that the record content is "returned check" data rather than a full transaction-history record. The specification's own stated need — "a need has thus arisen for an automated system for providing quick and efficient check verification and marketing follow-up" — is a stated design objective, which under KSR supplies the motivation. Where a reference teaches a POS system that alters displayed responses based on a PC-maintained check file, extending that file to accumulate frequency and dollar-volume counters is the application of a known technique to a known structure to obtain a predictable result.

Combination 2 — Admitted art (D)/(E) + Combination 1: automatic MICR capture of the account number as the customer ID

What it renders obvious: The use of the checking-account number as the unique customer key, captured automatically rather than keyed.

Motivation: (E) already discloses the concept of using the account number as a unique ID; (D) supplies the hardware to read MICR automatically. The only gap — teaching the reader which subfield of the On Us field is the account number — is squarely addressed in the field by the MICR Handbook's explanation of the field structure, and the specification itself concedes the "teach-the-reader" approach is an implementation matter (a stored transit code table plus a parsing table). Combining a stored transit-code lookup table with a known MICR reader is precisely the "arrangement of old elements yielding no more than expected" scenario.

Counter-argument the patent owner would press (and it has teeth): the spec asserts previous readers were "not satisfactory… because of their inability to uniformly detect desired account information on all checks in a consistent manner," and that account numbers are "interspersed with other data such as the check sequence number." If the '812 claims actually recite the transit-code-table/parsing limitation, the owner has a non-obviousness argument grounded in the art's failure to solve the branch-to-branch and bank-acquisition variability problem. This is the single most defensible limitation in the patent.

Combination 3 — Combination 1 + POS-coupon art (Mindrum '212 / Off '672) + prior loyalty programs

What it renders obvious: "Selective incentive marketing in response to customer shopping histories" — the second half of the title, and the likely substance of the independent "incentive" claims.

Motivation: Both art clusters are in the same field (retail POS checkout), solve overlapping problems (identifying and influencing shoppers at checkout), and the coupon-dispensing art already teaches the output side (dispense a coupon at the POS based on what the shopper did). The '812's contribution on this side is essentially to swap the input — use a historical, accumulated shopping record rather than the instant basket. Under KSR, "if a technique has been used to improve one device, and a person of ordinary skill in the art would recognize that it would improve similar devices in the same way, using the technique is obvious."

Combination 4 — Admitted art (F): the status-roll / DWT-interval bookkeeping

What it renders obvious: The Claim-limiting structural features described in the flow diagrams — Access Date/Time roll, Day/Week/Month/Total counters, purge limits, CAUTION→POSITIVE clearance interval, CALL MANAGER override. These are, on the patent's own description, database maintenance parameters ("the specific file specifications… are not critical to the invention, being a matter of design choice," and "the specific functions available… are a matter of routine design selection"). Time-windowed counters (day/week/month/total) and record-expiration/purge scheduling are elementary data-processing constructs; a POSITA implementing a rolling transaction-history record would reach them as a matter of ordinary engineering.

5. The strongest non-obviousness hook to expect (and where '812 may survive)

If the granted claims include a feedback loop of the type recited in sibling claim 9 of US 6,684,195 — "monitoring whether or not said incentives have been effective; and subsequently issuing incentives having values depending upon the results of said monitoring step" — that closed-loop incentive-valuation step is the limitation the POS-coupon art (Mindrum/Off) does not obviously supply; those references disclose issuing coupons, not measuring efficacy and re-setting the incentive value on that basis. A well-pleaded § 103 ground would need a separate reference or expert testimony on incentive-effectiveness measurement (direct-mail test/control methodology, split-run coupon testing) to close it. I have not identified such a reference, and I flag that as an open gap rather than assert one exists.

Second hook: the local-database-at-the-store limitation, contrasted with the admitted art's remote/telephone-line architecture. This is likely to fail as a distinction because the admitted art (B) and (C) is already local-PC-based, and local-vs-remote placement of a database is a classic design choice.

6. Procedural posture that affects the practical value of any § 103 ground

• Expired – Lifetime, anticipated expiration 2014-04-15 (Google Patents record). A § 103 invalidity holding now matters only for pre-expiration conduct still in suit.
• The subject matter is quintessentially covered business method (financial product/service; G06Q20/387, G06Q30/0224, G06Q30/0238, G06Q30/0255, G06Q30/0268, G06Q30/0269). However, the AIA § 18 CBM transitional program sunset on September 16, 2020 — [flag: this date is from my training data, not verified in this session] — so as of April 2026 the available vehicles are IPR (limited to § 102/§ 103 on patents/printed publications only), ex parte reexamination, or district-court § 282. IPR would be the natural forum, and the admitted-prior-art combinations above are all patents/printed publications, so they are IPR-eligible.
• I found no 2026 Federal Circuit or PTAB activity naming US 5,621,812 in this session. The only Deaton-citation hit I located was US 5,621,812 being used as prior art against other patents in a Jack Henry/data-treasury petition (https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1458229](/patent/1458229)/...) — i.e., '812 as a sword, not as a target.

7. Confidence summary

Bottom line: On the record I can verify, the '812's architecture, its local check-status database, its account-number-as-ID scheme, its frequency/dollar-interval bookkeeping, and its POS incentive delivery are each taught or suggested by the prior art the patent itself admits, in the same field, for the same stated purposes — a strong § 103 case under KSR for all but (i) any MICR transit-code-table/parsing limitation and (ii) any closed-loop "measure effectiveness, then re-set incentive value" limitation. Both of those need the literal claim text before anyone can opine responsibly.