Failure to Investigate: The Most Common Winning Ineffective-Assistance Claim

By Elizabeth Franklin-Best, Esq., and Christopher Zoukis, JD, MBA

What must you prove when counsel failed to investigate?

You must prove three things: that counsel failed to investigate something a reasonable lawyer would have looked into, what that investigation would have uncovered, and that there is a reasonable probability the result would have been different. The third fails without the second — so the supporting declaration must state the actual content of the missing evidence, not that it exists.

Key takeaways on failure-to-investigate claims

• Strickland states the duty in one sentence. Counsel “has a duty to make reasonable investigations or to make a reasonable decision that makes particular investigations unnecessary.” Strickland v. Washington, 466 U.S. 668, 691 (1984).
• What gets reviewed is the investigation, not the choice. Wiggins v. Smith, Warden, 539 U.S. 510 (2003), holds that “a reviewing court must consider the reasonableness of the investigation said to support that strategy” — the label “strategy” does not end the inquiry.
• An uncooperative client does not excuse the failure. Rompilla v. Beard, 545 U.S. 374 (2005), holds counsel “bound to make reasonable efforts to obtain and review material that counsel knows the prosecution will probably rely on as evidence of aggravation,” even where the defendant and his family said there was nothing to find.
• Conclusory allegations lose. The Fifth Circuit requires a movant to “name the witness, demonstrate that the witness was available to testify and would have done so, set out the content of the witness’s proposed testimony, and show that the testimony would have been favorable to a particular defense.” Day v. Quarterman, 566 F.3d 527, 538 (5th Cir. 2009).
• The affidavit is the claim. In the Sixth Circuit, a movant who has “not introduced affidavits or any other evidence establishing what [the witnesses] would have said” cannot show prejudice. Tinsley v. Million, 399 F.3d 796, 810 (6th Cir. 2005).
• This claim belongs in a §2255 motion, not a direct appeal. Massaro v. United States, 538 U.S. 500 (2003), holds an IAC claim may be brought under §2255 “whether or not the petitioner could have raised the claim on direct appeal,” because the district court is the forum that can build the record.

Is failure to investigate ineffective assistance of counsel?

It can be, and it is the category federal courts have granted relief on most often. But “can be” is doing real work in that sentence, and the distance between a grievance and a claim is where most of these motions die.

The Sixth Amendment does not require counsel to investigate everything. Strickland v. Washington, 466 U.S. 668 (1984) frames the duty as an obligation of reasonableness, not of exhaustiveness: “counsel has a duty to make reasonable investigations or to make a reasonable decision that makes particular investigations unnecessary.” Both halves of that sentence are operative. A lawyer who investigates and concludes further work would be pointless has satisfied the duty. A lawyer who never investigated and therefore could not have concluded anything has not.

Strickland also supplies the sliding scale that governs everything downstream: “strategic choices made after thorough investigation of law and facts relevant to plausible options are virtually unchallengeable; and strategic choices made after less than complete investigation are reasonable precisely to the extent that reasonable professional judgments support the limitations on investigation.” Read that as a ratio. The less counsel looked, the less protection the word “strategy” buys. And the Court instructed that “a particular decision not to investigate must be directly assessed for reasonableness in all the circumstances, applying a heavy measure of deference to counsel’s judgments.”

Two boundaries before going further. The two-prong framework itself — deficient performance plus prejudice, and every qualifier the Court built into each — is set out at the Strickland v. Washington two-prong test. The full map of attorney-conduct categories that federal courts have accepted, of which this is one, is at ineffective assistance of counsel. This page assumes both and goes straight to the proof problem that is specific to investigation claims.

What Wiggins v. Smith actually changed

Wiggins v. Smith, Warden, 539 U.S. 510 (2003) is the case to read if the Government’s response to your motion says the word “strategy,” because Wiggins is where the Supreme Court refused to accept that label at face value.

Counsel in Wiggins limited the mitigation investigation to a presentence report and Department of Social Services records, never commissioned a social history, and later characterized the omission as a tactical choice to argue residual doubt instead. The Court did not evaluate the tactic. It evaluated what the tactic was built on: “In assessing the reasonableness of an attorney’s investigation, however, a court must consider not only the quantum of evidence already known to counsel, but also whether the known evidence would lead a reasonable attorney to investigate further.” And then the sentence that decides most of these cases: “Strickland does not establish that a cursory investigation automatically justifies a tactical decision with respect to sentencing strategy. Rather, a reviewing court must consider the reasonableness of the investigation said to support that strategy.”

That is the doctrinal move worth understanding. The question is not “was the decision defensible?” — almost any decision can be made to sound defensible after the fact. The question is “was the investigation behind the decision reasonable?” A lawyer who knows nothing cannot make an informed choice, and an uninformed choice is not a strategic one.

Wiggins also confirms where the outside standard comes from. The Court treated “[p]revailing norms of practice as reflected in American Bar Association standards and the like” as “guides to determining what is reasonable” — not binding rules, but evidence of the professional baseline. In practice that means the prevailing standards for the offense type at the time of the representation are fair material for a §2255 motion, and a declaration from an experienced practitioner describing what a competent lawyer would have done in that district in that year is admissible and useful.

One structural note. Wiggins, Rompilla, and Williams all arrived at the Supreme Court as §2254 petitions from state convictions, so each was filtered through AEDPA’s deference standard on top of Strickland. That filter is what Harrington v. Richter, 562 U.S. 86 (2011) called “doubly” deferential review. On a federal §2255 motion there is no §2254(d) layer — the district court applies Strickland directly. The Strickland propositions those cases state therefore apply with full force to a federal movant, and if anything a federal §2255 movant is better positioned than the petitioners who won them.

Rompilla v. Beard: the file counsel knew the Government would use

Rompilla v. Beard, 545 U.S. 374 (2005) answers the hardest version of the government’s defense — that the client and his family said there was nothing to find.

Counsel in Rompilla interviewed the defendant and five family members, consulted three mental health experts, and got nothing usable. What counsel did not do was pull the Commonwealth’s file on the prior conviction the prosecution had announced it would use as aggravation, even though that file sat in the same courthouse. The Supreme Court held that “even when a capital defendant’s family members and the defendant himself have suggested that no mitigating evidence is available, his lawyer is bound to make reasonable efforts to obtain and review material that counsel knows the prosecution will probably rely on as evidence of aggravation at the sentencing phase of trial.”

The Court’s explanation of why is the part that generalizes. “Reasonable efforts certainly included obtaining the Commonwealth’s own readily available file on the prior conviction to learn what the Commonwealth knew about the crime, to discover any mitigating evidence the Commonwealth would downplay, and to anticipate the details of the aggravating evidence the Commonwealth would emphasize.” And, more bluntly: “It flouts prudence to deny that a defense lawyer should try to look at a file he knows the prosecution will cull for aggravating evidence, let alone when the file is sitting in the trial courthouse, open for the asking.”

Two limits keep the holding honest. Justice O’Connor’s concurrence states that the decision “imposes no such rule” of mandatory prior-conviction file review and “simply applies our longstanding case-by-case approach.” And the Court did not address counsel’s duty when the material is a warehouse of records rather than a single accessible file. The generalizable proposition is narrower than “counsel must review everything”: where counsel knows the Government will rely on a specific body of material, and that material is reasonably obtainable, not obtaining it is difficult to defend.

For a federal §2255 movant, the Rompilla fact pattern recurs constantly outside the capital context. Prior convictions used for a career-offender designation. Underlying records for a relevant-conduct computation. The state case file behind a §922(g) predicate. Co-defendant plea agreements and cooperation histories. Where the Government told counsel in advance what it intended to use, and counsel never looked, Rompilla is directly on point.

Williams v. Taylor: the mistake that is not strategy at all

Williams v. Taylor, 529 U.S. 362 (2000) is the cleanest illustration of a failure to investigate that no post-hoc rationalization can reach, because the omission was rooted in a legal error rather than a judgment call.

Counsel there “failed to conduct an investigation that would have uncovered extensive records graphically describing Williams’ nightmarish childhood, not because of any strategic calculation but because they incorrectly thought that state law barred access to such records.” The Court then cataloged what the records held — the parents’ imprisonment for criminal neglect, repeated beatings, two years in state custody including an abusive foster placement, borderline intellectual functioning, no schooling past sixth grade — along with prison records commending Williams and testimony from prison officials describing him as among the inmates “least likely to act in a violent, dangerous or provocative way.”

The lesson for a §2255 motion is about cause, not just omission. There is a meaningful difference between “counsel weighed this and chose not to pursue it” and “counsel did not know the material was available.” The second is what Williams describes, and it is the version courts reject most readily as a strategic justification. When you interview trial counsel or obtain the file, the question worth asking is not “why didn’t you call this witness?” It is “did you know this record existed, and if not, why not?”

Williams also settles a related point that the Government still argues. Citing Lockhart v. Fretwell and Nix v. Whiteside, the Court held those cases “do not justify a departure from a straightforward application of Strickland when the ineffectiveness of counsel does deprive the defendant of a substantive or procedural right to which the law entitles him.” In other words, the prejudice inquiry is the ordinary reasonable-probability inquiry, not some heightened “fundamental fairness” overlay.

The four-element uncalled-witness showing

The most common failure-to-investigate claim in federal post-conviction practice is the uncalled witness: counsel never interviewed the alibi witness, never contacted the co-worker who could have testified about the timeline, never spoke to the family member who would have corroborated the account.

The Fifth Circuit states the required showing in four discrete elements, and it is the clearest checklist any circuit has produced. In Day v. Quarterman, 566 F.3d 527, 538 (5th Cir. 2009), the court held that “to prevail on an ineffective assistance claim based on counsel’s failure to call a witness, the petitioner must name the witness, demonstrate that the witness was available to testify and would have done so, set out the content of the witness’s proposed testimony, and show that the testimony would have been favorable to a particular defense.” The court added that it has “required this showing for claims regarding uncalled lay and expert witnesses alike,” and rejected the argument that the requirement adds elements to Strickland: “Strickland itself requires this specific showing for uncalled witness claims.”

A circuit note, stated precisely. The four-element enumeration is the Fifth Circuit’s, traced to Alexander v. McCotter, 775 F.2d 595, 602 (5th Cir. 1985). Other circuits are not in conflict with it, but they do not phrase the rule as four elements. The Sixth Circuit requires the same substance through the prejudice prong; the Eleventh Circuit requires a proffer of favorable testimony; the D.C. Circuit requires a “comprehensive showing” of what the investigation would have produced. Treat the four elements as the best available drafting checklist nationwide, and cite your own circuit’s formulation in the motion.

The four-element uncalled-witness showing and what documents each element requires

#	Element	What it means	What must be attached	What fails
1	Name the witness	Identify the person specifically — full name, and enough identifying detail that the court and the Government can locate them.	The witness’s name in the motion body and in a sworn declaration; last known address or relationship to the movant.	“Several witnesses in my neighborhood.” “A co-worker.” Any unnamed category of person.
2	Availability	The witness existed, was locatable at the relevant time, and was not legally or practically unavailable (dead, out of the country, invoking the Fifth Amendment).	A statement in the witness’s own declaration that they were available during the trial or sentencing period; or, if the witness will not sign, records placing them locally and reachable.	Silence on availability. A witness who was a fugitive, in custody elsewhere, or would have invoked the Fifth.
3	Willingness to testify	The witness would actually have taken the stand — not merely that they knew something.	The witness’s declaration saying, in the first person, that they would have testified if asked. This element is what most claims omit.	“She would have helped.” “He knew the truth.” A declaration that establishes knowledge but never says the witness would have appeared.
4	Content, and that it was favorable	The specific substance of the testimony, and how it would have changed the case — an element negated, an alibi supplied, a witness impeached, a Guidelines fact defeated.	The witness’s declaration setting out the testimony in factual detail, plus a paragraph in the motion tying that testimony to a specific element, defense, or Guidelines computation.	Characterizations instead of facts. “He would have contradicted the informant” without saying how, on what point, or with what knowledge.

Sources: the four elements are stated in Day v. Quarterman, 566 F.3d 527, 538 (5th Cir. 2009) (citing Alexander v. McCotter, 775 F.2d 595, 602 (5th Cir. 1985)); the documentary requirements reflect the practice federal district courts apply under Rule 2(b) and Rule 7 of the Rules Governing Section 2255 Proceedings.

Do I need an affidavit from the witness?

Practically, yes — and understanding why explains what to do when you cannot get one.

The declaration is not a formality attached to the claim. In most cases it is the claim, because it is the only admissible proof of elements 3 and 4. The Sixth Circuit put it plainly in Tinsley v. Million, 399 F.3d 796, 810 (6th Cir. 2005): the petitioner “has not introduced affidavits or any other evidence establishing what they would have said,” and “[i]n the absence of any evidence showing that they would have offered specific favorable testimony, Tinsley cannot show prejudice.” Note the detail — Tinsley claimed to have affidavits that “simply indicated that they would have testified.” Willingness alone was not enough. Content was still missing.

The Eleventh Circuit reaches the same result on the other half: in Alexander v. Dugger, 841 F.2d 371, 375 (11th Cir. 1988), the petitioner “proffers no evidence to suggest that [the witness] would have testified favorably had his attorney questioned him,” and the claim failed for want of prejudice. Alexander also contains the warning every movant should absorb before naming a witness: the court found that a second uncalled witness had given a deposition contradicting the alibi, so “counsel’s failure to call [him] as a witness, far from prejudicing Alexander’s case, actually helped it.” Investigate your own witness before you plead them.

A declaration that works is signed under penalty of perjury under 28 U.S.C. §1746, is in the witness’s own voice, states how the witness knows what they know, sets out the testimony as facts rather than conclusions, and says expressly that the witness was available and would have testified if called. It does not need notarization. Rule 2(b)(5) of the Rules Governing Section 2255 Proceedings applies the same mechanism to the motion itself, which “must … be signed under penalty of perjury by the movant or by a person authorized to sign it for the movant.”

When the witness will not sign. This happens often, and it is not automatically fatal. The options, in descending order of strength:

1. A declaration from the person who spoke to the witness — an investigator, a family member, current counsel — stating the date of contact, what the witness said, and that the witness confirmed they would have testified. Hearsay concerns are real but Rule 7(b) of the §2255 Rules expressly contemplates that “[a]ffidavits also may be submitted and considered as part of the record.”
2. Contemporaneous documents that carry the content — a recorded interview, a police report of the witness’s statement, a text message thread, a written statement given at the time.
3. A Rule 6 discovery request. Rule 6(a) allows a judge, “for good cause,” to authorize discovery under the criminal or civil rules — including a deposition of a witness who will not sign voluntarily. Good cause means specific factual allegations that, if developed, would entitle the movant to relief, which is precisely why elements 1, 2, and 4 have to be pleaded even when element 3 is out of reach.
4. A request for a hearing with a subpoena. If the court orders an evidentiary hearing, Rule 8(c) requires appointment of counsel for a qualifying movant, and the witness can be compelled.

What does not work is a motion that says the witness exists and asks the court to find out what they would have said. That is the posture Tinsley rejected, and it is the posture district courts routinely reject.

Applied Insight — Christopher Zoukis, JD, MBA, Managing Director: When we open a file where the complaint is “my lawyer never talked to anyone,” the first working document is not a legal memorandum. It is a witness list — every name the client can produce, what each person would say, how the client knows they would say it, and whether anyone has actually spoken to them since trial. That list usually shrinks by half on contact, because some witnesses turn out not to know what the client believed they knew and some would have hurt the case. What survives that pass is the claim. Building the list first also protects the client: the worst outcome is a motion naming a witness whose actual account, once the Government reaches them, damages the record.

Records that were never obtained

The second large category is documentary rather than testimonial, and it is the one that is easiest to prove because the evidence still exists.

Rompilla is the anchor: counsel is bound to make reasonable efforts to obtain material counsel knows the Government will use. But the category is broader than aggravation files. In federal practice the recurring omissions are:

• Prior conviction records used to support a career-offender designation under the Guidelines, or a predicate under the Armed Career Criminal Act — where the charging document, plea colloquy, or judgment would have shown the prior did not qualify.
• Medical and mental-health records bearing on competency, on capacity at the time of the offense, or on a mitigation theory at sentencing.
• Institutional records — school, military, employment, child welfare — that supply the social history Wiggins and Williams were about.
• Financial and business records in fraud cases, where loss amount drives the Guidelines and the underlying documents were never analyzed.
• Co-defendant and cooperator materials the Government produced but counsel never reviewed against the trial theory.

The proof problem here is easier than with witnesses, and movants under-use that fact. Records do not change their minds, do not refuse to sign, and do not need to be persuaded to appear. If the record exists, obtain it, attach it, and state in the motion what it shows and which finding it would have defeated. That converts an abstract complaint into a documentary showing a judge can evaluate on the papers. Where the records are in the case file or the transcript rather than in the movant’s hands, they have to be obtained before the motion is drafted — see getting transcripts, the PSR, and your file for how. Where the missing material is prior statements of Government witnesses, the governing statute is the Jencks Act, and where it is exculpatory material the Government suppressed, the claim is a Brady violation rather than an investigation claim.

What if my lawyer never hired an expert?

Expert failures are the third category, and they divide along a line that decides most of them: whether counsel made a choice or simply did not know.

Harrington v. Richter is the losing side for the movant. Counsel there did not consult a blood-evidence expert, and the Court held a state court could reasonably conclude that fell within professional norms, because “[i]n many instances cross-examination will be sufficient to expose defects in an expert’s presentation.” Richter also supplies the prejudice language the Government will quote at you: “The likelihood of a different result must be substantial, not just conceivable.”

The winning version of the claim is not “a better expert existed.” It is that counsel did not know an expert was available, did not know what the discipline could establish, and never found out — the Williams pattern applied to expertise. And the proof requirement is the same as for witnesses. Day v. Quarterman expressly extends the four-element showing to “uncalled lay and expert witnesses alike,” and Tinsley rejected an expert claim because the petitioner had not “identified an expert who would have testified that these possibilities were more probable,” offering “other than conclusory allegations that such experts exist.”

So the attachment for an expert claim is a letter or declaration from a qualified expert, retained now, stating their qualifications, what analysis they performed or would have performed, what opinion they would have offered at the time of trial or sentencing, and the basis for it. A retrospective expert report that reaches a different conclusion than the Government’s expert is admissible support under Rule 7 and is what moves a forensic-reexamination claim from speculation to a factual dispute the court has to resolve. This matters most in cases turning on drug quantity and purity, firearms and toolmark comparison, digital forensics and device attribution, loss calculation, medical causation, and eyewitness identification — fields where the underlying science has moved and a competent examination today may reach a different result than the one the jury heard.

How do I show prejudice from a failure to investigate?

This is the question the page exists to answer, because it is where the claim is won or lost and where the competitor pages stop.

Prejudice under Strickland is “a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different,” and “[a] reasonable probability is a probability sufficient to undermine confidence in the outcome.” Applied to an investigation claim, that standard has a built-in sequencing problem: you cannot show a reasonable probability of a different result from evidence you have not produced. The prejudice showing is therefore derivative of the content showing, and a motion that omits the content has not merely weakened prejudice — it has made prejudice unprovable as a matter of logic.

The circuits say this in nearly identical terms. The D.C. Circuit in United States v. Askew, 88 F.3d 1065, 1073 (D.C. Cir. 1996), held that a defendant basing a claim on failure to investigate “must make a comprehensive showing as to what the investigation would have produced. The focus of the inquiry must be on what information would have been obtained from such an investigation and whether such information, assuming its admissibility in court, would have produced a different result.” The court explained why the rule is fair: “an appellate court should not overturn a conviction simply because the defendant has teasingly suggested that there may be facts out there that his trial counsel could have discovered and that would have helped his case. If any such facts exist, the defendant must identify them.”

Three practical points about the prejudice showing.

It is measured against the whole record, not in isolation. The court weighs the missing evidence against what the jury actually heard. Where the Government’s case rested on a single cooperating witness, one corroborated impeachment fact can undermine confidence in the outcome. Where the Government had a confession, surveillance, and forensic corroboration, an additional character witness will not.

The relevant proceeding may be sentencing rather than trial. If the uninvestigated material would have changed the Guidelines calculation or supplied a variance argument, the prejudice is measured in months of imprisonment, not in the verdict — a materially easier showing, treated separately at ineffective assistance at sentencing.

Admissibility counts. Askew qualifies the inquiry with “assuming its admissibility in court.” Evidence that would have been excluded produces no prejudice, so a claim built on inadmissible hearsay, on a privileged communication, or on a witness whose testimony would have opened the door to the movant’s prior convictions is not a prejudicial omission. That last risk is real — Tinsley held counsel’s decision not to call character witnesses defensible precisely because it would have “opened the door to cross-examination on the conviction.”

What if my lawyer says it was a strategic decision?

Expect this, because it is the Government’s standard response and it usually arrives in the form of an affidavit from trial counsel filed with the Government’s answer.

Three things defeat it, in order of strength.

First, the Wiggins answer. The label is not the test. “Strickland does not establish that a cursory investigation automatically justifies a tactical decision…. Rather, a reviewing court must consider the reasonableness of the investigation said to support that strategy.” If counsel never spoke to the witness, never pulled the record, and never consulted the field, then whatever counsel decided was not a strategic choice within the meaning of Strickland — it was a choice made in the absence of the information a strategic choice requires.

Second, the record. Harrington v. Richter holds that “Strickland … calls for an inquiry into the objective reasonableness of counsel’s performance, not counsel’s subjective state of mind.” That cuts both ways, and here it cuts for the movant: an after-the-fact affidavit reciting a strategy is not conclusive if the contemporaneous record contradicts it. Look for billing and CJA vouchers showing no investigative time, a case file with no interview memoranda, a continuance never sought, an investigator never appointed, a funding request under 18 U.S.C. §3006A never made, or a sentencing transcript in which counsel offers a rationale inconsistent with the one now advanced.

Third, internal inconsistency. A strategy that would have required knowing the very fact counsel says was not worth investigating is not a strategy. Williams is the model: counsel’s real reason was a mistaken belief that state law barred access, which is a legal error, not a judgment.

One caution that belongs on any honest page. Filing this claim generally waives attorney-client privilege as to the subject matter of the claim, which is how trial counsel’s affidavit gets into the record in the first place. The scope of that waiver, and how courts limit it, is covered at attorney-client privilege waiver in a 2255. Decide what you are willing to open before you plead.

Applied Insight — Elizabeth Franklin-Best, Esq., Principal Attorney: The most useful document in a failure-to-investigate case is often the one nobody thinks to request: trial counsel’s own file, including time records. A strategy defense is an assertion about a decision, and decisions leave traces — a memo, a retained investigator, an interview note, a funding motion, a continuance. When the file contains none of those for the period in question, the affidavit describing a considered choice has nothing behind it, and that gap is what we put in front of the court. It is also the reason we prefer to have the file in hand before the motion is drafted rather than after the Government answers.

Why this claim belongs in a §2255 motion, not a direct appeal

Because the record you need does not exist yet, and the Supreme Court has said so.

Massaro v. United States, 538 U.S. 500 (2003), holds that “an ineffective-assistance-of-counsel claim may be brought in a collateral proceeding under §2255, whether or not the petitioner could have raised the claim on direct appeal,” and that “failure to raise an ineffective-assistance-of-counsel claim on direct appeal does not bar the claim from being brought in a later, appropriate proceeding under §2255.” That resolves the procedural-default problem that would otherwise sink most of these claims.

The reasoning is specific to investigation claims and worth quoting, because it explains the entire architecture of this page. “When an ineffective-assistance claim is brought on direct appeal, appellate counsel and the court must proceed on a trial record not developed precisely for the object of litigating or preserving the claim and thus often incomplete or inadequate for this purpose…. The evidence introduced at trial, however, will be devoted to issues of guilt or innocence, and the resulting record in many cases will not disclose the facts necessary to decide either prong of the Strickland analysis.” The district court, by contrast, “may take testimony from witnesses for the defendant and the prosecution and from the counsel alleged to have rendered the deficient performance.”

A witness who was never called does not appear in a trial transcript. A record counsel never obtained is not in the appellate record. That is precisely why Massaro routes the claim to §2255 — and why the motion, not the appeal, is where the evidence has to be built. Massaro is careful to add that it does not hold IAC claims “must be reserved for collateral review,” since “[t]here may be cases in which trial counsel’s ineffectiveness is so apparent from the record” that appellate counsel raises it directly. The choice between the two vehicles is treated at direct appeal vs 2255, and the residual default rules at procedural default.

What the motion has to attach

Rule 2(b) of the Rules Governing Section 2255 Proceedings sets the pleading floor, and it is more demanding than notice pleading. The motion must “specify all the grounds for relief available to the moving party,” “state the facts supporting each ground,” and “state the relief requested,” and it must be “signed under penalty of perjury.”

That is why Rule 4(b) is the real gatekeeper. Under Rule 4(b), the judge “must promptly examine” the motion, and “[i]f it plainly appears from the motion, any attached exhibits, and the record of prior proceedings that the moving party is not entitled to relief, the judge must dismiss the motion.” Note what the judge examines: the motion, the attached exhibits, and the prior record. A failure-to-investigate claim with no exhibits is being screened on the pleading alone, against a record built to establish guilt.

A complete filing on this claim ordinarily includes:

Attachment	What it establishes	Rule it comes in under
Movant’s declaration under 28 U.S.C. §1746	What the movant told counsel, when, and what counsel said in response	Rule 2(b)(5)
Witness declarations, one per witness	Elements 1–4 of the uncalled-witness showing	Rule 7(b)
Investigator or third-party declaration	Substitute proof of witness content where the witness will not sign	Rule 7(b)
The records counsel never obtained	The documentary content the investigation would have produced	Rule 7(a)–(b)
Expert letter or report	What a qualified examination would have concluded, and on what basis	Rule 7(b)
Trial counsel’s file, notes, and time records	The absence of investigative work behind the claimed strategy	Rule 6(a) if counsel will not produce
Relevant transcript excerpts and the PSR	The record against which prejudice is measured	Rule 4(b) record
Proposed discovery requests	Preserves the Rule 6 request if the court wants development	Rule 6(b)

Two mechanical points. Rule 6(b) requires that a discovery request “provide reasons for the request” and “include any proposed interrogatories and requests for admission, and must specify any requested documents” — a bare request for discovery will be denied. And under Rule 7(c), the judge “must give the party against whom the additional materials are offered an opportunity to admit or deny their correctness,” so expect the Government to respond to every declaration you file.

Whether any of this produces a hearing is a separate question governed by Rule 8(a), and it is covered at 2255 evidentiary hearing. What matters here is the relationship between the two: courts order hearings on investigation claims when the attachments create a genuine factual dispute the record cannot resolve, and dismiss them under Rule 4(b) when the attachments do not exist.

Applied Insight — Elizabeth Franklin-Best, Esq., Principal Attorney: We build a failure-to-investigate motion from the exhibits inward. Before a word of argument is written, the exhibit list has to be able to answer four questions on its own: who, what they would have said, that they would have said it, and what difference it makes to a specific element or a specific Guidelines finding. If an exhibit cannot be tied to one of those four, it does not go in — length is not persuasion in a Rule 4(b) screening, and a thick motion with thin attachments reads exactly like what it is.

What this claim is not

Three boundaries, because misdirected claims consume the one year under §2255(f) and then fail.

It is not a complaint that counsel lost. Counsel who investigated, found nothing useful, and made a judgment has satisfied Strickland. The claim requires an omission, not a defeat.

It is not a substitute for a suppression or Brady claim. Where the Government withheld material favorable evidence, the constitutional violation belongs to the prosecution and is analyzed under a different framework at Brady violation — although counsel’s failure to pursue what the Government did disclose can be an investigation claim.

It is not a general re-argument of the facts. New evidence that shows innocence, but does not implicate counsel’s performance, is a different vehicle with a different deadline — see newly discovered evidence in federal court and actual innocence.

And the deadline governs all of it. A §2255 motion is subject to a one-year limitation period running from the latest of four trigger dates, and the clock does not pause while you gather declarations. Compute the date before you start collecting, at the 2255 deadline calculator, and confirm which of the four triggers applies at post-conviction deadlines. Where the failure concerned advice about a plea rather than an investigation, the governing analysis is at guilty plea ineffective assistance; where it concerned the appeal, at ineffective appellate counsel.

Why work with Elizabeth Franklin-Best, P.C.

Elizabeth Franklin-Best founded the firm in 2019 and practices federal criminal appeals and post-conviction relief nationwide. She is admitted to the U.S. Supreme Court and twelve of the thirteen U.S. Courts of Appeals — every regional circuit (1st–11th) plus the D.C. Circuit — and has handled more than 330 federal proceedings, over 100 of them appeals, with sixteen published decisions as counsel of record. She is recognized by Best Lawyers in America 2026 for Appellate Practice and by Chambers USA 2026 for Litigation: White-Collar Crime & Government Investigations, was named South Carolina Public Defender of the Year in 2010, and is the author of Reversing Your Criminal Conviction: Develop Your Winning Strategy (2023).

Christopher Zoukis, JD, MBA, is the firm’s Managing Director and a nationally recognized federal prison expert, author of the Federal Prison Handbook and four other books on the federal system, and Chairman of the Board of the Human Rights Defense Center. He earned his J.D. at UC Davis School of Law after serving roughly a decade in the federal Bureau of Prisons.

The firm runs a national federal practice on an intentionally small caseload, from offices in Columbia and Mount Pleasant, South Carolina, and is admitted pro hac vice in federal courts from Montana to Florida. A failure-to-investigate claim is an evidence-gathering project before it is a legal argument, and that is how the firm builds one.

Talk to a federal post-conviction lawyer

If something in your case was never looked into and you want to know whether it can be turned into a pleadable §2255 claim, schedule a case review or call (843) 620-1100. Bring the judgment, the docket sheet, the presentence report, the sentencing transcript, and the names of anyone who was never contacted. If your one-year deadline is close, say so when you call. Start from the hub at federal 2255 motion lawyer if you are still deciding which motion you need.

Frequently Asked Questions About Failure to Investigate

Reviewed for legal accuracy by Elizabeth Franklin-Best, Esq., Principal Attorney·September 2026