MEMORANDUM AND ORDER
CASPER, District Judge.
I. Introduction
Lead class action plaintiffs (“Plaintiffs”) have filed this lawsuit against ModusLink Global Solutions, Inc. (“ModusLink” or the “Company”), its former Chief Executive Officer, Joseph Lawler and its former Chief Financial Officer, Steven Crane (“Defendants”), alleging securities fraud in violation of Sections 10(b) and 20(a) of the Exchange Act and SEC Rule 10b-5 promulgated thereunder. D. 29; see 15 U.S.C. § 788(b); 15 U.S.C. § 78t(a), 17 C.F.R. § 240.10b-5. Defendants have moved to dismiss the Amended Complaint. D. 32. For the reasons stated below, the Court DENIES the motion to dismiss.
II. Factual Background
The Court acknowledges that there is a related case before this Court. See In re ModusLink Global Solutions, Inc. Derivative. Litig. (Montini v. Lawler, Inc.), No. 12-cv-11296-DJC (the “Derivative Litigation”). Although the facts alleged in the Derivative Litigation are substantially similar to those in the instant case, the Court must evaluate Plaintiffs’ claims based upon the facts alleged in their complaint in this case and the applicable legal standards that apply in this direct action by the shareholders against the Company and its officers.
A. ModusLink and Its Business
Plaintiffs are shareholders of Modus-Link, which is a publicly traded “global supply chain service company” that procures raw materials for its clients. D. 29 ¶ 2. Plaintiffs bring this action “on behalf of all persons or entities that purchased ModusLink securities between September 26, 2007 and June 8, 2012.” Id. ¶ 14. The Company charges a percentage mark-up to its clients, which amounts to its profit margin. Id. ¶ 2. ModusLink aggregates client orders for a given product, purchasing a large quantity of the product from suppliers and then fulfilling client orders at a mark-up pursuant to contracts with individual clients. Id. ¶27. In some situations, ModusLink receives rebates from its suppliers when it purchases a large volume of products. Id. ModusLink’s clients do not receive notice when ModusLink has received a rebate from suppliers. Id. Ac[64]*64cording to Confidential Witness (“CW”) l,1 suppliers would provide volume discounts to ModusLink through lower upfront costs or rebates in the form of a check at designated intervals. Id. ¶ 28. According to CW 2,2 ModusLink typically sold products to clients at a price six to eight percent higher than the cost to the Company to procure the materials. Id. ¶ 29.
According to CW 1, most of Modus-Link’s larger customers had contracts that required ModusLink’s mark-up to be “calculated as a percentage of the Company’s final cost of raw materials as opposed to the initial contract price per unit.” Id. ¶ 30. As a result of the rebates that Mo-dusLink received from suppliers, Modus-Link’s “final cost of raw materials was often significantly lower than the price listed on the initial client contract.” Id. Accordingly, ModusLink’s contracts with many of its customers required the Company to remit the rebates to its customers. Id. ¶2. CW 43 and CW 54 confirm that most contracts “allowed for some volume discount to be passed on to the client,” either through “cost-plus model pricing” or “most-favored nation” pricing. Id. ¶¶ 33, 38.
B. The Alleged Fraud on Modus-Link’s Customers
CW 2 provided an illustration of the effect on customers that withholding rebates might have on ModusLink’s revenues:
CW 2 cited as an example the scenario where ModusLink would purchase a component at $100 per box and the client cost-plus model contract stipulated that the Company’s mark-up was 6%. Per the terms of the contract, Modus-Link would invoice the client at $106, which includes the $100 per-box cost of goods plus $6, which represented the 6% mark-up. However, if at the end of the quarter or the year, ModusLink received a volume discount of $20 per box, the Company’s actual cost of goods was $80, and the 6% mark-up applied to ModusLink’s actual $80 cost per box would yield a $4.80 mark-up instead of the $6 mark-up if the Company used the pre-discounted cost of goods, for a total price of $84.80 per box. In CW 2’s hypothetical scenario ... the client should have been invoiced $84.80 per [65]*65box (the true discounted cost of goods of $80 per box plus the 6% mark-up of $4.80), instead of the $106 per box (the $100 per box pre-discounted cost of goods plus the 6% mark-up of $6.00) ... [and] the Company would improperly receive an additional $21.20 in revenue to which it was not entitled under a cost-plus model contract.
Id. ¶ 34. This is the crux of Plaintiffs’ claims against the Defendants. According to CW 1, after signing a new client, the Company would “price out the cost of all of the components and give the true price to the client.” Id. ¶ 39. Over time, however, ModusLink would negotiate more favorable prices with suppliers, but continue to charge clients the initial price in contravention of ModusLink’s contractual obligations. Id.
As alleged by Plaintiffs, perpetrating a fraud of this nature required ModusLink to deceive its customers by revising purchase orders and invoices to reflect the original price of goods rather than the discounted prices that reflected rebates or discounts to ModusLink. Id. ¶¶ 40, 42. As CW 1 explained, “[i]t was a fundamental practice at ModusLink to hide our savings ... ModusLink was hiding the rebate costs and not passing that along to clients ... resulting] in inflated costs and markups.” Id. ¶ 43. Put another way, Modus-Link hid its rebates “on the backend. Instead of showing it on purchase orders, we just got a check that was lumped on a quarterly basis.” Id. ¶44. Although a portion of these quarterly remittances was supposed to go to clients, they were not reflected in ModusLink’s purchase orders to suppliers, which, in turn, facilitated the ease through which ModusLink could conceal receipt of additional revenue outside of the ordinary course. See id.
In addition, it was company policy at ModusLink to tell clients that they were not entitled to receive rebates. Id. ¶45. According to CW 2, one way in which ModusLink accomplished this alleged deception was to confirm to clients that “ ‘[ModusLink] got the rebate, but it wasn’t based on your order, it’s from a combination of orders so you don’t really deserve a discount.’.... [ModusLink] could have this type of conversation with multiple clients and then just kept the discount.” Id. ¶ 46. CW 3 confirmed that “[t]he complexity of the process also creates an effective shield against the customer figuring out an individual item cost.” Id. ¶ 49.
Eventually, certain of ModusLink’s clients caught on to the alleged scheme and launched audits of ModusLink’s practices that revealed the extent of the underlying fraud. By the end of 2010, two of the Company’s largest clients, SanDisk and Intuit, conducted audits of their accounts at ModusLink. Id. ¶¶ 57-58. Subsequently, the United States Securities and Exchange Commission (“SEC”) launched an investigation into Modus-Link’s accounting practices. Id. ¶ 89. This precipitated an internal investigation by the Company’s Audit Committee, which “uncovered improprieties in the Company’s accounting treatment of volume discounts and customer contracts.” Id. Moreover, the Company acknowledged that the Audit Committee “determined that certain client contracts have not been aligned consistently with ModusLink’s practice of retaining volume discounts ... [and] identified limited instances where vendor costs incurred where marked-up to clients in a manner not consistent with client contracts.” Id. ¶ 90. These investigations culminated in the resignations of Lawler and McLennan on June 11, 2012. Id. ¶ 89. On that same date, the Company announced a wide ranging restatement of its financials for the fiscal years of 2007 to [66]*662011 and the first two quarters of 2012. Id. ¶ 90. This announcement resulted in the Company’s stock price dropping 34.74 percent.5 Id. ¶ 91.
The restatement of earnings demonstrated, inter alia, that ModusLink had a net income of $500,000 during from 2007 to 2011 (the “Class Period”) instead of the $18.4 million that it had initially reported. Id. ¶ 76. The restatement also demonstrated that the Company’s actual gross margin was 11.6 percent as opposed to the 12.1 percent that the Company had initially reported. Id. ¶ 81.
C. ModusLink’s Alleged Material Misstatements
During the Class Period, Plaintiffs allege that Defendants repeatedly certified the accuracy of the Company’s financial results and certified the Company had presented these results in compliance with Generally Acceptable Accounting Principles (“GAAP”). Id. ¶ 99. As a result of the Company’s consistent overstatement of its revenues, Plaintiffs allege that its annual (10-K) and quarterly (10 — Q) reports to the SEC during the Class Period materially misreported the Company’s financial results. Id. ¶¶ 100-04 & n. 20-27. Similarly, Plaintiffs allege that Lawler and Crane (the “Individual Defendants”) falsely certified the accuracy of the Company’s reported earnings on multiple occasions. Id. ¶¶ 105-06 & n. 28.
Plaintiffs also allege that ModusLink made false statements through its securities filings regarding the Company’s profit margin, falsely attributing increases to decreases in cost of revenue. Id. ¶¶ 108-111 & n. 30-32. Lawler and Crane echoed these allegedly false conclusions during quarterly earnings calls. Id. ¶¶ 113-17. As part of ModusLink’s securities filings, Plaintiffs further contend that Lawler and Crane also executed false certifications regarding the Company’s initial controls. Id. ¶¶ 121-25 & n. 33-34.
Plaintiffs allege that ModusLink’s securities filings were materially false when they stated that problems in net revenue were related to the procurement and resale of materials on behalf of clients. Id. ¶ 127. Plaintiffs allege that this statement was false because it omitted that the Company was required to reimburse clients for overcharges inconsistent with contractual obligations. Id. ¶ 128. Moreover, Plaintiffs allege that the Company made materially false statements about the rebates themselves, asserting that “we believe the benefit of these rebates is effectively shared with our clients and that our accounting for such practices is correct.” Id. ¶ 131.
Finally, Plaintiffs allege that Modus-Link’s accounting practices violated GAAP and SEC regulations in their reporting of income. Id. ¶¶ 133-163.
In sum, Plaintiffs allege that by claiming the rebates as revenue, Defendants violated accounting principles and made public statements and filed documents with the SEC that purported to demonstrate that the Company’s financial performance was far more robust than it was. Id. ¶ 9. Plaintiffs allege that by facilitating an overvaluation of the Company, Lawler and Crane benefited personally. Id. ¶ 169.
[67]*67D. Defendants’ Awareness of the Company’s Scheme to Defraud Customers
As a general matter, CW 1 states that Lawler, Crane and McLennan were personally and solely responsible for decisions regarding if and when ModusLink passed rebates onto clients. Id. ¶ 56. It was only upon the approval of Lawler, McLennan, or Cathy Venable, Senior Vice President of Finance Transformation, that CW 1 would alter purchase orders or invoices. Id.
CW 4 reports that by October 2010, she was aware of various reports prepared by Lawler and Crane as well as McLennan and Chief Information Officer Matt Datillo that included inconsistent revenue figures on different versions of contracts and budgets. Id. ¶ 51. The Company had reason to manipulate these numbers; certain contracts specified that once ModusLink’s profits on a given contract reached a certain level, its customers would receive a discount. Id. By altering these numbers, the Company’s executives ensured that this “trigger number” would not be reached and the client would not be entitled to a discount. Id. When CW 4 approached Lawler and Crane about these discrepancies, she was told to “mind her own business,” and the Company then terminated her employment. Id. ¶ 69.
CW 66 corroborated CW 4’s account. According to CW 6, CW 4 told CW 6 in the fall of 2010 that she was “unhappy with some of the things that Joe Lawler was instructing her to do with rebates and finances” and was “being instructed to use them in a particular manner that she didn’t necessarily agree with.” Id. ¶ 74. Although CW 6 reported this conversation to the Director of Internal Audit, no one ever followed up with her about the matter. Id.
CW 6 also agreed that only six or seven individuals had full access to the Company’s entire breadth of financial information and asserted that the “direction of what to do with rebates had to come from Joe Lawler.” Id. ¶ 73. CW 2 noted that “Steve Crane set the policy on rebates” and that as a result “the problems had to occur at the CFO level.” Id. ¶ 63.
CW 3 also corroborated Lawler and Crane’s control over ModusLink’s operations, noting that the rebate scheme was “mishandled at the very top level ... the pressure to make targets would have been passed down thoughtlessly through McLennan’s organization and [in order to meet targets] the monkey business would start ... Any executive at ModusLink who tells you they didn’t know what was going on with rebates is lying to you.” Id. ¶¶ 66-67. Moreover, very few people at the Company had the authority to approve price changes; this group included Lawler and Crane. Id. ¶ 68.
There were also very few people at the Company, including Lawler and Crane, with authority to misreport a cost savings as revenue. Although rebate retention should have appeared as a cost savings, the Company reported these gains as revenue. Id. ¶ 72.
According to CW 1, Lawler instructed her to ensure that ModusLink was not passing rebates along to clients, telling her: “we are not making enough margin, you need to get more rebates to decrease the cost of goods. Don’t tell [the] sales [department] the real cost of our materials [68]*68because they will try to give it back to the client.” Id. ¶ 59.
During the SanDisk and Intuit audits, in response to Lawler’s inquiries about whether the Company was “covered” and whether they were “making sure that the purchase orders are matching the invoices” so if there were “any gaps” the auditors would not find them because “we don’t want them to know about the rebates,” id. ¶ 58, CW 1 told Lawler that the rebates would not be discovered because the purchase orders and invoices both reflected the Company’s prerebate prices. Id.
Plaintiffs also allege that Lawler and Crane had incentives to inflate the Company’s revenue numbers. First, by doing so, they made themselves appear as effective managers to the Company’s Board of Directors, who defended Lawler during a proxy contest. Id. ¶ 168. Second, this appearance of effectiveness made it possible for Lawler and Crane to misuse Company funds for personal purposes. Id. ¶¶ 171-73. Third, Lawler and Crane received performance-based compensation. Id. ¶¶ 175-80.
III. Procedural History
Plaintiffs commenced this action on June 12, 2012. D. 1. The Court consolidated this action with two other pending cases, No. 12-cv-11078 and No. 12-cv-11279, on November 18, 2012, appointing Plaintiffs the Lead co-Plaintiffs in this action. D. 11. Plaintiffs amended their complaint on February 11, 2013. D. 29. Defendants subsequently moved to dismiss the amended complaint. D. 32. The Court heard the parties on November 9, 2013 and took this matter under advisement. D. 50.
IV. Discussion
This is a direct action in which the Company’s shareholders have sued the Company and its officers through claims arising under the federal securities laws. D. 29. This is unlike the Derivative Litigation in which the plaintiffs in that action have sued the Company’s officers and directors on behalf of the Company through claims arising under the Delaware common law. No. 12-cv-11296-DJC, D. 36.
A. Standard of Review
At this stage, Defendants have only challenged the sufficiency of Plaintiffs’ scienter allegations. D. 33.7 Because the Private Securities Litigation Reform Act (“PSLRA”) applies to the instant case, Plaintiffs must “state with particularity facts giving rise to a strong inference that the defendant acted with the requisite state of mind,” here, scienter. Id.; see also Ernst & Ernst v. Hochfelder, 425 U.S. 185, 193 n. 12, 96 S.Ct. 1375, 47 L.Ed.2d 668 (1976) (defining scienter as “a mental state embracing intent to deceive, manipulate, or defraud”); Greebel v. FTP Software, Inc., 194 F.3d 185, 198-201 (1st Cir.1999) (noting that recklessness is sufficient to establish scienter). The Supreme Court has construed the PSLRA’s requirement to plead facts giving rise to a “strong inference” of scienter to mean that “a complaint will survive ... only if a reasonable person would deem the inference of scien-ter cogent and at least as compelling as any opposing inference one could draw from the facts alleged.” Tellabs, Inc. v. Makor Issues & Rights, Ltd., 551 U.S. 308, 324, 127 S.Ct. 2499, 168 L.Ed.2d 179 (2007). Thus, in evaluating a securities class action complaint, courts “must take [69]*69into account plausible opposing inferences.” Id. at 323, 127 S.Ct. 2499. This is necessarily a fact-specific inquiry. Aldridge v. A.T. Cross Corp., 284 F.3d 72, 82 (1st Cir.2002). Courts conducting such an inquiry must determine “whether all of the facts alleged, taken collectively, give rise to a strong inference of scienter, not whether any individual allegation, scrutinized in isolation, meets that standard.” Tellabs, 551 U.S. at 323, 127 S.Ct. 2499 (emphasis in original). “Under the PSLRA, the complaint must state with particularity facts that give rise to a strong inference of [the required state of mind], rather than merely a reasonable inference. ... The inference ... must be reasonable and strong, but need not be irrefutable.” In re Stone & Webster, Inc., Sec. Litig., 414 F.3d 187, 195 (1st Cir.2005) (internal quotation marks omitted) (quoting In re Cabletron Systems, Inc., 311 F.3d 11, 38 (1st Cir.2002)); see also Tellabs, 551 U.S. at 324, 127 S.Ct. 2499 (noting that inference of scienter need not be of the “smoking gun” variety). Nor does the PSLRA require the plaintiff to “plead evidence.” Id. at 33. “[W]here there are equally strong inferences for and against scienter, Tellabs now awards the draw to the Plaintiff.” ACA Fin. Guar. Corp. v. Advest, Inc., 512 F.3d 46, 59 (1st Cir.2008).8
B. The PSLRA Is a High Hurdle for Plaintiffs to Clear
Adopted in 1995, Congress enacted the PSLRA to elevate the procedural hurdle to stating securities fraud claims “in a bipartisan effort to curb abuse in private securities litigation, especially the filing of so-called strike [that is, meritless, “nuisance” actions brought to obtain a favorable settlement] suits.” In re Galileo Corp. Shareholders Litig., 127 F.Supp.2d 251, 260 (D.Mass.2001). There is a discussion of Congress’s intent in passing the PSLRA in Galileo:
In particular, Congress sought to reform private securities litigation to discourage unmeritorious class actions, including actions brought because of a decline in stock prices. The aims of the PSLRA are three-fold: (1) to encourage the voluntary disclosure of information by corporate issuers; (2) to empower investors so that they — not their lawyers — exercise primary control over private securities litigation; and (3) to encourage plaintiffs’ lawyers to pursue valid claims and defendants to fight abusive claims. The PSLRA seeks to curtail the filing of abusive lawsuits at the pleading stage of litigation by establishing uniform and stringent pleading requirements.
Id. (internal citations and quotation marks omitted). Even before the Supreme Court decided Tellabs, commentators acknowledged the PSLRA’s success, finding that dismissals as a percentage of dispositions had doubled in the years following the PSLRA’s passage. Mark H. Gitenstein and Charles A. Rothfield, Securities Litigation Reform — Early Signs of Success, 16 No. 9 Andrews Corp. Off. & Directors Liab. Litig. Rep. 16 (2001).
As part of the Tellabs inquiry, courts engage in a particularized scrutiny of private securities complaints. See, e.g., Zucco Partners, LLC v. Digimarc Corp., 552 F.3d 981, 996 (9th Cir.2009) (discrediting one confidential source’s conclusory allegations and another’s that were based upon triple hearsay). Nevertheless, the [70]*70PSLRA’s pleading standard is not an insurmountable bar. Courts may consider circumstantial evidence, see Simon v. Am. Power Conversion Corp., 945 F.Supp. 416, 434 (D.R.I.1996), and allegations that do not amount to admissible evidence at all. Cabletron, 311 F.3d at 33. Plaintiffs have made a wide variety of allegations in support of their argument that Defendants intentionally defrauded ModusLink shareholders. As required by Tellabs, the Court considers the allegations collectively to determine whether the allegations give rise to a strong inference of scienter. Zucco, 552 F.3d at 992 (citing Tellabs, 551 U.S. at 323, 127 S.Ct. 2499).
C. The Allegations Pled in the Amended Complaint Collectively Support a Strong Inference of Scienter
The lynchpin of Plaintiffs’ theory is that throughout the Class Period, Lawler and Crane were aware of the manner in which the Company was deceiving its customers. Consequently, Lawler and Crane, both personally and as agents for ModusLink, were aware that them and the Company’s statements to investors were materially false. See D. 39 at 20 & n. 12. Defendants argue that the facts pled in the Amended Complaint support more strongly an alternate inference. Defendants argue that ModusLink and its executives honestly believed they were entitled to retain rebates from suppliers and “like many in business, jealously guarded what they believed at the time to be proprietary information about a properly retained economic benefit.” D. 33 at 10. According to Defendants, to accept that Plaintiffs’ allegations support a showing of scienter, the Court would be concluding that “Defendants committed fraud in the light of day,” a notion that they argue is implausible. D. 42 at 8. Against this backdrop, the Court analyzes the facts pled in the Amended Complaint.
1. The Statements of the Confidential Witnesses Proffered by Plaintiffs are Strongly Indicative of Scienter
a. The CWs Have a Sufficient Basis of Knowledge
By far, much of Plaintiffs' factual basis in support of their theory of the case is the statements of the six CWs, some of whom had high-ranking positions in the Company. Securities fraud plaintiffs may rely on confidential witnesses without specifically naming them "provided they are described in the complaint with sufficient particularity to support the probability that a person in the position occupied by the source would possess the information alleged." Cabletron, 311 F.3d at 29; see also New Jersey Carpenters Pension & Annuity Funds v. Biogen IDEC Inc., 537 F.3d 35, 51 (1st Cir.2008) (noting that confidential sources must be "described in the complaint with sufficient particularity to support the probability that a person in the position occupied by the source would possess the information alleged"). Some courts have looked askance upon confidential sources that "were simply not positioned to know the information alleged, ... report only [multi-layer] hearsay[ ] and ... allege conclusory assertions of scienter.” Zucco, 552 F.3d at 996.
Here, however, as alleged, some CWs clearly have a basis of knowledge to report on the inner workings of ModusLink’s financial operations. CW 1 worked in purchasing and sourcing, reporting directly to McLennan, the Company’s President of Global Operations. D. 29 ¶ 28 n. 9. As such, CW 1 was directly involved with purchasing raw materials from suppliers and was therefore aware of the manner in which ModusLink received rebates. See, e.g., id. ¶¶ 30-31, 39. Similarly, as the Senior Vice President of Sales for the [71]*71Americas, CW 2 reviewed and approved contracts with and pricing levels for clients. Id. ¶ 29 n. 10. This gave CW 2 a basis of knowledge similar to that of CW 1. CW 2 also had a direct awareness of Mo-dusLink’s relationships with its customers. See id. ¶ 46. CW 3 was ModusLink’s Business Unit President of the Americas and was a member of the Company’s “Executive Leadership Team.” Id. ¶ 32 n. 11. Accordingly, CW 3 reported directly to Lawler and worked alongside Crane. Id. ¶ 32 & n. 9, 11. Because CW 3 had operational and functional responsibility for finance, id., he had direct knowledge of the company’s financial reporting. See id. ¶ 35. CW 4 was ModusLink’s Manager for Travel and Commodity. Id. ¶ 33 n. 13. She reported to William McLennan. Id. The Court agrees with Defendants, D. 33 at 28 n. 10, that Plaintiffs have not shown that CW 4 was involved with the rebate program. CW 4 was, however, a member of the Executive Leadership Team and therefore had access to top-level presentations. Id. ¶¶ 25, 51. CW 5 was ModusLink’s Senior Vice President of Supply Base. Id. ¶ 33 n. 14. As he had negotiating authority for a large amount of purchasing volume, he had intimate knowledge of the manner in which ModusLink purchased raw materials. Id. CW 6 was the Company’s IT Director. Id. ¶ 73. Similar to CW 4, CW 6 does not have a basis of knowledge about the rebate program itself, but an IT employee does have a basis of knowledge for opining as to who had access to certain electronic files.
CW’s 1, 2, 3, 4, 5 interacted with Lawler and Crane. D. 29 ¶ 28 & n. 9, 10, 11, 13, 14. This further bolsters their reliability. Compare Selbst v. McDonald’s Corp., No. 04-2422, 2005 WL 2319936, at *7 (N.D.Ill. Sept. 21, 2005) (crediting confidential source that “regularly interacted” with Defendants) with Konkol v. Diebold, Inc., 590 F.3d 390, 401 (6th Cir.2009) (rejecting reliability of confidential witnesses where they were “not identified as having any contact or interaction with any of the Defendants”) (emphasis in original), abrogated on other grounds, Frank v. Dana Corp., 646 F.3d 954 (6th Cir.2011). The Court therefore credits the allegations of the CW’s to the extent described above.
b. Certain CW Statements Are Sufficiently Particular
Plaintiffs identify seventeen specific allegations that the CWs have provided that they allege support a strong inference of scienter. D. 29 ¶ 166 (summarizing all confidential witness statements previously relayed in the complaint). For the Court to consider and credit these allegations, these allegations cannot merely be “conclusory allegations of fraud, but specific descriptions of the precise means through which it occurred, provided by persons said to have personal knowledge of them.” Cabletron, 311 F.3d at 30.
The Court agrees with the Company that some the CW’s allegations are conclu-sory and non-specific, particularly where allegations assume that Lawler “had to be” responsible. D. 29 ¶ 61; see In re A123 Sys., Inc. Sec. Litig., 930 F.Supp.2d, 278, 285 (D.Mass.2013) (discrediting allegation that defendant “must have known” of fraud). On the other hand, the Court finds that certain allegations are sufficiently particular. In this regard, for example, the Court points to CW l’s statement that “(i) she received directions concerning rebates, financials and margins directly from Defendant Lawler and McLennan; (ii) the Individual Defendants and McLennan were solely responsible for decisions regarding if and when rebates were passed along to ModusLink clients” as this allegation identifies who instructed CW 1 and what individuals were responsible for the [72]*72decision to remit rebates to clients. Id. ¶ 166(b). Similarly, the Court points to other allegations that demonstrate that the decisions to authorize rebates were made at the top levels of management and those that specifically identify Lawler and Crane as the individuals with authority to authorize rebates, prices changes and reorganize the ways in which ModusLink accounted for revenues and costs. See id. ¶ 166(h), G), (k), (n), (o).
In addition, the Court notes CW 4’s narrative regarding the manner in which she identified accounting inconsistencies and was consequently dismissed from her position. Id. ¶ 166®, (m), (q). This allegation is particular as to the timing of the allegation, demonstrates an attempt to silence dissent and is corroborated by CW 6. Id.
Also in the mix are statements Lawler allegedly made to CW 1 about attempting to conceal discrepancies from clients. See id. ¶¶ 166(c)-(d) (recounting Lawler’s statements that “are we covered and are we making sure that the purchase orders are matching the invoices? Do we have any gaps here that they are going to find because we don’t want them to know about the rebates” and “we are not making enough margin, you need to get more rebates to decrease the cost of goods. Don’t tell sales the real cost of our materials because they will try to give it back to the client”). These allegations identify specifically what was said, identify the speaker as one of the Defendants and tie Lawler’s statements to a customer audit. Id. The Court, therefore, disagrees that all of the “confidential witness statements are ... so vague and of such unreliable origin as to be unpersuasive.” Zucco, 552 F.3d at 1000.
The Court concludes that these allegations from the confidential witnesses, collectively, support a strong inference of scienter. CW l’s allegation that Lawler did not “want [SanDisk] to know about the rebates” and that wanted to ensure that the Company was “covered” so the customer did not discover their conduct leads the Court to infer that Lawler reasonably knew rebates were being wrongfully withheld from SanDisk. Defendants argue generally that the Court could infer from such allegations that ModusLink and its executives concealed rebate information from consumers to “jealously guard[ ] what they believed at the time to be proprietary information about a properly retained economic benefit.” D. 33 at 10. Although Defendants’ reading of this allegation is plausible, the Court cannot say that their reading is more likely than an inference of fraudulent intent. Where an inference of fraudulent intent is “at least as likely” as an inference of non-fraudulent intent, the Court errs in favor of the Plaintiffs. Tellabs, 551 U.S. at 329, 127 S.Ct. 2499; ACA, 512 F.3d at 59.
Additionally, the Court finds that CW 4’s narrative regarding her dismissal from ModusLink supports an inference of scien-ter. Based on the close temporal proximity between CW 4 raising the issue of revenue discrepancies (October 2010) and her dismissal from the Company (December 31, 2010), it is reasonable to infer from this narrative that the Company retaliated against CW 4 for raising the issue to Lawler and Crane and terminated her employment to prevent CW 4 from raising more questions that could expose the fraud. Defendants offer no competing inference as to CW 4’s termination from the Company. The Court, therefore, concludes that this narrative supports an inference of scienter.
Moreover, the contemporaneous nature of the confidential witness statements discussed above contradicts Defendants’ assertion that the allegations merely demon[73]*73strate that Defendants knew that they had improperly retained rebates only after the fact. Although it is true that the First Circuit has “consistently rejected” the notion of “fraud by hindsight,” see In re The First Marblehead Corp. Sec. Litig., 639 F.Supp.2d 145, 160 (D.Mass.2009) (quoting Rodriguez-Ortiz v. Margo Caribe, Inc., 490 F.3d 92, 97 (1st Cir.2007)), Plaintiffs’ allegations support a strong inference of a contemporaneous scheme to conceal rebates from customers and the Company’s shareholders during the Class Period itself. Defendants’ argument is therefore unavailing. Taken together, these statements “reasonably could be read as admissions that [the Defendants] knew” that what it was doing was wrong. Simon v. Am. Power Conversion Corp., 945 F.Supp. 416, 434 (D.R.I.1996).
The Court expresses doubt as to the Defendants’ argument that the confidential witness statements in this regard lack particularity. The Court cannot construe the PSLRA’s pleading requirement to mean that confidential witnesses, who are former employees of the Company, must recall all possible details from their former positions. See In re St. Jude Medical, Inc. Sec. Litig., 836 F.Supp.2d 878, 899 (D.Minn.2011) (noting that there is nothing requiring that confidential witnesses provide direct evidence of individual defendants’ scienter); In re Dura Pharm., Inc., Sec. Litig., 548 F.Supp.2d 1126, 1132 (S.D.Cal.2008) (finding confidential witness statements sufficient and noting that level of detail is weighed corroborative nature of other facts, reliability of sources, number of sources and plausibility of allegations).
2. Plaintiffs’ Motive and Opportunity Allegations Give Rise to an Inference of Scienter
Plaintiffs’ allegations as to motive and opportunity also have some heft. The Second Circuit has determined that where a plaintiff shows that he has “benefited in a concrete and personal way from the purported fraud” that this is sufficient to allege scienter. Novak v. Kasaks, 216 F.3d 300, 311 (2d Cir.2000). The First Circuit takes a slightly less formalistic view, noting that “[a]s part of the mix of facts, the plaintiff may allege that the defendants had the motive ... and opportunity ... to commit the fraud.... However, ... while mere allegations of motive and opportunity alone may be insufficient, together with additional factual support, evidence of motive and opportunity may establish a strong inference of scienter.” Aldridge v. A.T. Cross Corp., 284 F.3d 72, 82 (1st Cir.2002) (citing Novak, 216 F.3d at 307; Greebel, 194 F.3d at 197).
As to motive, Plaintiffs have alleged that Lawler and Crane personally benefited from the fraud. D. 29 ¶¶ 168 et seq. For example, Plaintiffs allege that by inflating the Company’s revenues, Lawler and Crane made themselves appear as effective managers to the Company’s Board of Directors, who defended Lawler during a proxy contest. D. 29 ¶ 168. In addition, Plaintiffs allege that this appearance of effectiveness made it possible for Lawler and Crane to misuse Company funds for personal purposes. Id. ¶¶ 171-73. Third, Lawler and Crane received performance-based compensation. Id. ¶¶ 175-80.
As to opportunity, Plaintiffs have demonstrated that Lawler and Crane number among the few people who had the authority to disburse or withhold rebates from clients. D. 29 ¶ 166(b) (alleging that “[Lawler, Crane and] McLennan were solely responsible for decisions regarding if and when rebates were passed along to ModusLink clients”). Moreover, Plaintiffs have alleged that Lawler and Crane were among the few who could misreport retained rebates as revenues. Id. ¶ 72 (noting that rebate retention “should not ap[74]*74pear in the revenue line unless someone purposely took that cost savings and booked it as revenue to inflate the sales line”); id. ¶ 166(n), (o). That Lawler prepared presentations with inconsistent revenue figures also bolsters the notion that he had the opportunity to manipulate the manner in which ModusLink reported its revenue. Id. ¶ 166©. These allegations support Plaintiffs’ assertion that Defendants had motive and opportunity to engage in the conduct alleged in the Amended Complaint.
Defendants argue that Lawler and Crane’s stock purchases during the Class Period undercut the assertion that they believed the Company was overvalued due to inflated revenue figures. D. 33 at 34. It is true that if ModusLink was inflating their revenues by improperly retaining rebates, Lawler and Crane could have taken advantage of this by selling their stock in the Company. The Court acknowledges that the stock market could find stock sales by insiders as being highly suspicious and could draw attention from regulators or shareholders at a time when they were attempting to conceal fraud. See In re Apple Comp. Sec. Litig., 886 F.2d 1109, 1117 (9th Cir.1989) (concluding that “insider [sales] in suspicious amounts or at suspicious times is probative of bad faith and scienter”) (citing Goldman v. Belden, 754 F.2d 1059, 1070 (2d Cir.1985)).9 It is, therefore, entirely reasonable for Lawler and Crane to refrain from selling stock during the Class Period to avoid the appearance of wrongdoing. Moreover, as Plaintiffs note, Lawler and Crane purchased their stock “immediately after a marked drop in the price of ModusLink’s common stock, or at prices below Modus-Link’s stock price at the end of the Class Period.” D. 39 at 36. This behavior is consistent with “dollar cost averaging,” when stockholders purchase shares of a company just after its price has fallen to make up losses. See McPhail v. First Command Fin. Planning, Inc., 247 F.R.D. 598, 603 n. 6 (S.D.Cal.2007) (adopting National Association of Securities Dealer’s definition of “dollar cost averaging” and explaining that by investing the same amount each period, the investor purchases more shares when the security price is low, and fewer shares when the price is high). Thus, this behavior does not necessarily bear upon Plaintiffs’ scien-ter allegations. Finally, the SEC reports cited by Defendants that allegedly demonstrate that Lawler and Crane purchased ModusLink shares during the Class Period do not provide any insight into their so-called “short positions” (i.e., sales of borrowed stock) in the Company that would effectively negate their long position (shares actually owned). See Robertson v. Clayton Brokerage Co. of St. Louis, Inc., 587 F.Supp. 678, 684 (N.D.Ga.1984) (discussing how holding an equal number of long positions and short position or a “spread” position results in the sharehold[75]*75er from effectively holding no interest in the company).
At oral argument, Defendants presented the Court with a chart demonstrating that Lawler and Crane’s equity investments in the Company decreased in value during the Class Period by roughly $100,000 for Lawler and roughly $70,000 for Crane. By holding their investments in the Company, Defendants noted, Lawler and Crane lost money. D. 51 at 17-18. Accordingly, Defendants argued, Lawler and Crane could not have possibly expected the Company’s value to decrease. However, Plaintiffs have alleged that Lawler and Crane also earned over $20 million in total compensation, including performance-based bonuses, during the Class Period. D. 29 ¶ 175. Whatever percentage of this compensation was equity, the fact remains that this loss represented a mere 0.85 percent of Lawler and Crane’s total salary. The benefit to Lawler and Crane by remaining in their positions at the Company thus far outweighed any financial damage they incurred by holding their equity positions in the Company.10 Accordingly, even if not dispositive, Defendants’ motive and opportunity are part of the “mix of facts” that may give rise to a strong inference of scienter. Aldridge, 284 F.3d at 82.
3. Plaintiffs' Remaining Allegations Snpport an Inference of Scienter
Plaintiffs also allege that Mo~ dusLink's issuance of a restatement is indicative of scienter. “In general, the mere publication of a restatement is not enough to create a strong inference of scienter.” Zucco, 552 F.3d at 1000; In re Segue Software, Inc. Sec. Litig., 106 F.Supp.2d 161, 169 (D.Mass.2000) (same). However, the “magnitude and frequency of the restatements may support an inference of scienter.” In re Focus Enhancements, Inc. Sec. Litig., 309 F.Supp.2d 134, 159 (D.Mass.2001). Moreover, the Ninth Circuit has found a restatement of earnings highly probative where general allegations about “management’s role in a corporate structure and the importance of the corporate information about which management made false or misleading statements” are buttressed with “detailed and specific allegations about management’s exposure to factual information within the company.” Zucco, 552 F.3d at 1000 (citation omitted). Here, Plaintiffs have indeed made general allegations about the importance of the information that was the subject of the restatement, noting that its result was that ModusLink’s profits during the Class Period were nearly wiped out. D. 29 ¶ 76. Plaintiffs have also plausibly alleged that Lawler and Crane were exposed to the information that was the subject of a restatement. Id. ¶ 166(b), (Z). Accordingly, the restatement is properly before the Court in the mix of facts that the Court may consider in evaluating Plaintiffs’ allegations of scienter.
[76]*76Defendants argue that the resignations of Lawler and McLennan, the alleged lack of internal controls and the SEC’s inquiry into ModusLink are insufficient to meet the Tellabs standard. D. 33 at 28-34. Even if this is true, this does not undercut the strength of Plaintiffs’ remaining allegations discussed above. The Court must consider the totality of all of the allegations pled by Plaintiffs. Considering all of these allegations together, Plaintiffs paint a picture that strongly suggests Defendant’s fraudulent intent.
The Defendants seek to “divide and conquer,” D. 39 at 33, Plaintiffs’ allegations of scienter by picking apart each allegation one-by-one. The cases cited by Defendants are valid for the proposition that certain factors such as resignations of high-ranking company officers or government investigations are, in isolation, insufficient to support a strong inference of scienter. See, e.g. Zucco, 552 F.3d at 1000; Garfield v. NDC Health Corp., 466 F.3d 1255, 1265 (11th Cir.2006); Abrams v. Baker Hughes Inc., 292 F.3d 424, 434 (5th Cir.2002); Isham v. Perini Corp., 665 F.Supp.2d 28, 35 (D.Mass.2009); Grillo v. Tempur-Pedic Int’l, Inc., 553 F.Supp.2d 809, 818-19 (E.D.Ky.2008). Plaintiffs have, however, not only provided the Court with allegations by confidential witnesses that support a contemporaneous knowledge of the alleged wrongdoing as well as motive and opportunity allegations, but have also provided the Court with a litany of other allegations bearing on the issue of scienter. Individually, the facts pled may not be sufficient to defeat a motion to dismiss, but collectively they are. See Kafenbaum v. GTECH Holdings Corp., 217 F.Supp.2d 238, 247 (D.R.I.2002) (finding factual allegations that “taken together, create a strong inference of scien-ter”). This is especially true where no “smoking gun” is required to allege a strong inference of scienter, especially at this stage. Tellabs, 551 U.S. at 324, 127 S.Ct. 2499.
Finally, the Court addresses Defendants’ contention that it is implausible that Defendants could have “committed fraud iri the light of day.” D. 42 at 1. Defendants suggest that astute businesspersons would never dare to defraud their customers and rely on the hope that their buyers were simply too unsophisticated to understand the nature of the fraud. This argument both assumes its conclusion and is inconsistent with recent history regarding the financial downturn. See, e.g., IBEW Local 90 Pension Fund v. Deutsche Bank AG, No. 11-4209, 2013 WL 1223844, at *13 (S.D.N.Y. Mar. 27, 2013) (denying motion to dismiss securities class action where plaintiffs alleged that investment banks pooled mortgages of substandard quality into residential mortgage backed securities and collateralized debt obligations, sold these investment vehicles to investors while failing to disclose their poor quality, yet bet against them, generating billions of dollars in revenue for the banks).
D. As the Court Finds Plausible Evi- . dence of a Primary Violation, the Court Denies Defendants’ Motion to Dismiss the Section 20(a) Claim
Plaintiffs assert a claim under § 20(a) of the 1934 Act, 15 U.S.C. § 78t, against Lawler and Crane. Under § 20(a), “a person who controls a party that commits a violation of the securities laws may be held jointly and severally liable with the primary violator.” Maher v. Durango Metals, Inc., 144 F.3d 1302, 1305 (10th Cir.1998). A § 20(a) claim must be predicated on liability under § 10(b) and Rule 10b-5.
In their motion to dismiss, Defendants argue that the § 20(a) claim must be dismissed because the plaintiffs have not ade[77]*77quately pled a primary violation of § 10(b) and Rule 10b-5. However, as the Court has now held that the plaintiffs have stated a claim under § 10(b) and Rule 10b-5 against ModusLink, the Court concludes that Plaintiffs have adequately pled their § 20(a) claim.
V. Conclusion
For the above reasons, the Court DENIES Defendants’ motion to dismiss.
So Ordered.