HARLINGTON WOOD, Jr., Circuit Judge.

Defendants Kenneth Boula and Earl Dean Gordon pleaded guilty to three counts of mail fraud. 18 U.S.C. § 1341. The district court sentenced defendants to 108 months’ incarceration. The defendants appeal this sentencing determination, arguing that the district court’s application of sections 2F1.1(b)(2) and 3B1.1 of the United States Sentencing Guidelines (“Guidelines”) constituted impermissible “double counting” — inappropriate sentence enhancement through the application of two Guidelines provisions to the same conduct — and that the district court’s upward departure from the applicable Guidelines range was unjustified and improper.

I. BACKGROUND

Defendants conducted a massive mail fraud scheme that began in the late 1970s and early 1980s. Through their advertisements, free seminars, personalized investment counseling, and real estate tours, defendants induced investors to purchase interests in a series of limited real estate partnerships. Defendants informed investors that their money would be used to purchase property and to build improvements on the property. When the property sold, the investors would receive the principal investment plus interest. Unfortunately, these real estate plans did not function as promised. To raise money to pay existing investors, defendants created a pyramid or Ponzi scheme in which they attracted more investors, then diverted funds from the stated purposes and used the new investor money to pay interest to the old and new investors. 1 From these criminal investment activities, defendants took between ten and twenty per cent of the unit prices in commissions and fees. Defendants’ efforts in their Ponzi scheme resulted in $5.2 million in losses to investors.

The defendants fraudulently acquired an additional $1.8 million through three income partnerships. They did not use the money for the purposes explained to investors, but instead used the investors’ money to pay off demand notes to other investors and to pay operational costs.

*653 In March 1988, the Illinois Secretary of State issued an order of prohibition forbidding defendants from continuing to establish limited real estate partnerships. A class action commenced in the Northern District of Illinois in April 1988, and resulted in the appointment of a receiver to control the properties, partnerships and companies owned by defendants. In May 1989, the defendants were indicted by the Illinois Attorney General for sale of unregistered securities. The plea agreement with the state provides that the state sentences are to be concurrent with the federal sentences.

Defendants were charged with three counts of mail fraud, in violation of 18 U.S.C. § 1341, in an information filed February 15, 1990. At the arraignment the next day, defendants pleaded guilty to the charges and acknowledged their responsibility for the schemes alleged.

In the presentence report, the government recommended a sentence of forty-six months based on sentencing calculations under the Guidelines. Section 2Fl.l(a) covers offenses involving fraud and deceit and establishes the base offense level of six. The government increased the sentence eleven levels because the loss amount exceeded $5 million. U.S.S.G. § 2Fl.l(b)(l)(L). Because the scheme involved “more than minimal planning” and “more than one victim,” the government added two more levels. U.S.S.G. § 2F1.1(b)(2). The government suggested an additional increase of four levels for the defendants’ roles as organizers or leaders of criminal activity that was “otherwise extensive.” U.S.S.G. § 3Bl.l(a). The government then subtracted two levels for the defendants’ acceptance of responsibility, as provided in section 3E1.1 of the Guidelines, and arrived at the sentencing level of twenty-one. The government also asked for an unspecified departure from the sentencing level based on application notes one and ten of section 2F1.1. 2

The defendants agreed with most of the government’s recommendations, but objected to the enhancement for “more than minimal planning” and organization of “otherwise extensive” activity. Defendants argued, as they do on appeal, that application of the section 2Fl.l(b)(2) and 3B1.1 provisions constituted double counting. They also objected to any upward departure.

Following the sentencing hearing on June 15, 1990, the district court sentenced defendants to 108 months’ incarceration. The district court found, though the defendants disputed the finding, that the initial real estate, the Ponzi, and the income partnership schemes were part of the overall criminal fraudulent activity. The court therefore determined the total loss from the fraud to be $7 million. The district court also found that defendants targeted vulnerable victims, and that they used their gains to support an “extravagant” lifestyle. The district court accepted all of the government’s recommendations and did not agree with defendants’ double counting argument. The court also agreed that an upward departure was warranted. Reasoning that the Sentencing Commission had not accounted for schemes involving as many victims as in this case, the district court created its own “vector” system to figure an appropriate departure for the number of victims. 3 For the three thou *654 sand victims in this case, the district court determined that an upward departure of ten levels would be appropriate. The district court found the offense level to be 31, and sentenced defendants to the lowest sentence at that level — 108 months’ imprisonment. Defendants appeal this sentencing determination.

II. ANALYSIS

A. Double Counting Argument

The district court enhanced defendants’ sentence by two points under section 2F1.1(b)(2) and by four points under section 3Bl.l(a) of the Guidelines. For a fraud offense, section 2F1.1 permits a two-point enhancement if the offense involved “more than minimal planning” or “a scheme to defraud more than one victim.” U.S.S.G. § 2Fl.l(b)(2)(A) & (B). Section 3B1.1 pertains to the defendant’s role in the offense and provides for a four-point enhancement “[i]f the- defendant was an organizer or leader of a criminal activity that involved five or more participants or was otherwise extensive....” Defendants concede that the court properly enhanced their sentence under section 3B1.1 because they were organizers of an “otherwise extensive” scheme. They add, however, that because the section 3B1.1 enhancement is proper due to the “otherwise extensive” provision, enhancing their sentence for “more than minimal planning” under section 2F1.1 is enhancing the sentence twice for the same conduct. They argue that this constitutes improper double counting. Defendants’ argument fails for several reasons.

Defendants misinterpret the section 3B1.1 meaning of “otherwise extensive.” The commentary to that section explains that “[i]n assessing whether an organization is ‘otherwise extensive,’ all persons involved during the course of the entire offense • are to be considered. Thus, a fraud that involved only three participants but used the unknowing services of many outsiders could be considered extensive.” U.S.S.G. § 3B1.1, comment, (n. 2). See United States v. DeCicco, 899 F.2d 1531, 1536 (7th Cir.1990). Section 3B1.1 states that the enhancement applies if there were five or more participants or the activity was otherwise extensive; “otherwise extensive” activity therefore applies to the number of people involved in the operation, not the extent of the criminal activity. Boula and Gordon were the active participants in and organizers of the fraud, and during the course of the fraud, they employed between eight and eleven salesmen and between eight and eleven staff persons who contributed to their fraudulent activities. Their use of these additional participants made their fraud scheme otherwise extensive for section 3B1.1 enhancement purposes.

Defendants incorrectly argue that according to the section 3B1.1 commentary, the degree of planning of the offense is the primary factor in this enhancement and that such “otherwise extensive” planning is the same as “more than minimal planning” in section 2F1.1. The commentary actually states, however, that “the degree of participation in planning or organizing the offense” is the key factor, not the planning itself. U.S.S.G. § 3B1.1, comment, (n. 3) (emphasis added). The purpose of section 3B1.1 is to increase the sentence due to a defendant’s leadership role in an offense, not because of the nature of the offense itself. The Guidelines indicate that a defendant could have a key role in a criminal scheme that required minimal planning, and thus receive the section 3B1.1 enhancement because of the number of people involved and the defendant’s leadership role.

“More than minimal planning,” as defined in the commentary to the Guidelines application instructions, includes the following: more planning than would be typical for the particular offense; steps taken by defendants to conceal the criminal activity; or repeated acts over time. U.S.S.G. § 1B1.1, comment, (n. 1) Defendants’ *655 scheme lasted for eight years. This time element, the elaborate advertising, investment seminars designed to lure investors, and the staff involved are just some of the many examples of “more than minimal planning” by defendants in conducting their criminal scheme.

Defendants’ double counting argument is further hindered by the fact that section 2F1.1(b)(2) allows for a two-point enhancement when the scheme to defraud involves more than one victim. The defendants’ sentence could have been increased by two points regardless of their degree of planning, simply because there were three thousand victims in their scheme. As it is, the defendants participated in elaborate planning, justifying the section 2F1.1(b)(2) enhancement, and leadership of an extensive criminal enterprise, justifying the section 3Bl.l(a) enhancement. These Guidelines sections are not mutually exclusive and therefore no double counting occurred from these sentencing determinations.

B. Departure from the Guidelines

Boula and Gordon pleaded guilty to three counts of mail fraud. The base offense level for the fraud offense is six. The addition to this base level of the section 2F1.1 and 3B1.1 enhancements and the decrease for acceptance of responsibility resulted in a sentencing level of twenty-one. The sentence range for this level is 37-46 months. The district court added a ten-point departure to reach a sentence level of thirty-one. The resulting sentence of 108 months’ incarceration, while the lowest sentence for an offense level of thirty-one, is more than double the Guidelines sentence range for the predeparture offense level. We must determine whether departure was warranted, and whether it was warranted to this extent.

In reviewing a district court’s departure from the Guidelines sentence range, we determine “whether it was reasonable in light of the district court’s explanations for its departure at the time of sentencing.” United States v. Gaddy, 909 F.2d 196, 198— 199 (7th Cir.1990). See 18 U.S.C. §§ 3553(c), 3742(e)(3). A departure may be appropriate if “the court finds that there exists an aggravating or mitigating circumstance of a kind, or to a degree, not adequately taken into consideration by the Sentencing Commission [(“Commission”)] in formulating the guidelines.... ” 18 U.S.C. § 3553(b). 4 The court may also depart where unusual circumstances exist that make the Guidelines level attached to the factor inadequate. U.S.S.G. § 5K2.0. Where a factor is accounted for in the Guidelines, departure is warranted “only if the factor is present to a degree substantially in excess of that which ordinarily is involved in the offense of conviction.” Id.

We first examine the district court’s grounds for departure to determine whether they are adequate. Gaddy, 909 F.2d at 199. This is a question of law that we review de novo. We next determine whether the facts underlying the grounds for departure exist. Id. We accept the district court’s findings of facts unless they are clearly erroneous. United States v. Jordan, 890 F.2d 968, 972 (7th Cir.1989). The final element of the departure analysis is whether the degree of departure is appropriate. The structure of the Guidelines is used for guidance in making that determination. Gaddy, 909 F.2d at 199; United States v. Ferra, 900 F.2d 1057, 1062 (7th Cir.1990). We defer to the district court’s findings on the appropriate degree of departure “so long as [the departure] adequately reflects the structure of the Guidelines.” Gaddy, 909 F.2d at 199.

The district court gave the following as its grounds for the ten-point departure: (1) the commentary to section 2F1.1 explains that when the offense involves more than minimal planning and more than one victim, upward departure may be warranted; (2) the Commission did not account for fraud schemes involving as many victims as in this case; and (3) the $7 million loss exceeded the $5 million floor in the highest *656 loss category for fraud offenses. Defendants argue that these were improper grounds for departure because these factors are adequately addressed by the Guidelines.

The first application note to section 2F1.-1, the first of the district court’s stated grounds for departure, explains that departure may be warranted if two or more of the section 2F1.1(b)(2) provisions are satisfied. U.S.S.G. § 2F1.1, comment, (n. I). 5 As noted above, defendants participated in “more than minimal planning” over eight years. Their fraudulent activities affected not just “more than one victim” but three thousand victims. Because the Commission explicitly provided for departure in such a situation and because the defendants’ scheme so exceeded the Guidelines base provisions in section 2Fl.l(b)(2), we find the district court’s departure based on this ground to be proper. The other two stated grounds, however, are more difficult to accept.

Defendants’ eight-year scheme, involving three thousand victims and $7 million in loss, no doubt was inore extensive than the typical fraud case; yet not of a scale unknown to the Commission when setting up the Guidelines provisions. Ponzi’s scheme, in 1919, involved “thousands” of investors and millions of dollars. Cunningham, 265 U.S. at 7-8, 44 S.Ct. at 425. Numerous mail fraud cases have arisen since then and it is highly likely that when formulating the fraud provisions in the Guidelines the Commission anticipated that such major fraud schemes could occur again.

As stated in a policy statement on departure,

The Commission intends the sentencing courts to treat each guideline as carving out a “heartland,” a set of typical cases embodying the conduct that each guideline describes. When a court finds an atypical case, one to which a particular guideline linguistically applies but where conduct significantly differs from the norm, the court may consider whether a departure is warranted.

U.S.S.G. Ch. I, pt. A(4)(b), pp. 1.5-1.6 (policy statement). The district court’s second stated reason for departure, number of victims, actually was included by the Commission in the fraud sentencing provisions, but the only number required for enhancement is “more than one.” Despite this minimal numerical requirement, number of victims is expressly included in the fraud sentencing calculations and is not outside the heartland of the fraud provisions. The omission of any further consideration of number of victims from section 2F1.1 sentencing surely is more than mere oversight by the Commission.

Three thousand victims is an extremely large number, but the Commission accounted for massive fraud through the dollar amounts of the loss table. It is difficult to imagine that in setting up a loss table for $5 million, the Commission did not anticipate a large number of victims. The Commission may have anticipated that it would become too complicated to have different sentence level accounting based on both amount of loss and number of victims. In this case, for example, there were approximately three thousand victims and a total loss of $7 million. The average loss per person was $2333. If there were 10 victims and the loss remained $7 million dollars, making an average loss of $700,000 per person, would that be a less heinous crime that would merit a lower sentence than defendants received? This is not to make light of the seriousness of this crime. We simply wish to point out that the Commission elected to factor the sentencing level according to the total loss, not the number of victims.

*657 Section 2F1.1 has been amended since this case commenced, and while the new loss table does not apply to this case, it further indicates that number of victims is accounted for through monetary losses. The floor amount of the highest loss level is now $80 million, so the Commission definitely contemplated massive fraud schemes. Even with the applicable Guidelines loss level of $5 million, it is likely that the Commission would not have set the particular offense level based on monetary loss at so high a level had it wished number of victims to be accounted for as well.

We also recognize that note 9 of the section 2F1.1 commentary gives examples of when departure is proper. The number of victims is not among the factors, and while the list is intended to provide examples only and not exclusive factors for departure, it is significant that a factor as prominent in fraud as number of victims is not given as a valid reason for departure.

The district court’s final ground for departure was that the $7 million loss exceeded the $5 million loss that constituted the highest amount in the loss table. Based on that “over $5,000,000” loss amount, defendants received an eleven-point enhancement. The commentary to section 2F1.1 establishes that departure may be warranted if the loss “substantially exceeds” the $5 million amount. The previous loss level in the table covers amounts from $2 million to $5 million — a $3 million difference that results in the same sentence level increase. Given that degree of difference within one category of the loss table, and the $3 million difference between the last two loss levels in the table, we cannot find that $7 million, only $2 million greater loss than the $5 million ceiling of loss amounts, is in “substantial excess” and worthy of departure. 6

Having determined that departure was reasonable, because of the application note to section 2F1.1, we must examine the facts underlying that ground to make sure they exist. Defendants pleaded guilty and accepted responsibility for their crime. Defendants do not dispute the two- factors underlying the section 2F1.1 enhancement — that the scheme involved “more than minimal planning” or that there was “more than one victim.” The only dispute defendants have with the section 2F1.-1(b)(2) enhancement for “more than minimal planning” is the double counting argument that we rejected in the preceding section. Defendants do argue that the district court erred in finding that the total loss caused by defendants was $7 million, not the $5.2 million claimed by defendants, and that the defendants lived a life of luxury. These facts do not affect the grounds for departure provided for in the section 2F1.1 commentary. There was no clear error by the district court in making the factual determinations relevant to the section 2F1.1 departure. U.S.S.G. § 2F1.1, comment, (n. 1).

The last departure element to be determined is whether the degree of departure imposed by the district court — a ten-level enhancement based on the court’s own vector system — was reasonable. While we give deference to the district court in reviewing the reasonableness of the degree of departure, United States v. Schmude, 901 F.2d 555, 560 (7th Cir.1990), we must recognize that “[ujnless there is discipline in determining the amount of departure, however, sentencing disparity will reappear.” Ferra, 900 F.2d at 1062.

In United States v. Miller, 874 F.2d 466, 471 (7th Cir.1989), we stated that “ ‘[Reasonableness’ implies that a sentencing judge must provide articulable reasons, of *658 a type contemplated by the Act and the Guidelines, and based on a sufficiently sound factual foundation, to justify a departure from the guidelines.” If a departure is not reasonable, “we must vacate a sentence and remand the case for further proceedings.” Id. We commend the district court for its innovative efforts, in calculating the sentence departure, to construct a vector system for number of victims that was comparable to other sentencing tables. However, we conclude that number of victims was an element anticipated by the Commission and provided for through the “more than one victim” enhancement, as well as the large loss level enhancements. Defendants received an eleven-point enhancement based on the amount of loss, and a two-point enhancement for involving more than one victim. The district court’s ten-level departure that it based on number of victims under these circumstances was not reasonable.

Because of the seriousness of this crime and the application note to section 2F1.1, which does allow for departure when “more than minimal planning” and “more than one victim” were involved in the fraudulent activity, some departure from the offense level, as properly calculated under the Guidelines, is appropriate. We find, however, that the ten-point increase in this case was an improper degree of departure. Because this degree of departure from the Guidelines was not reasonable, we must vacate defendants’ sentence and remand the case for resentencing not inconsistent with this opinion. 18 U.S.C. § 3742(f)(2). Circuit Rule 36 shall not apply-

VACATED AND REMANDED.

. The name for this pyramid scheme is derived from Charles Ponzi, a notorious swindler. Starting in 1919, Ponzi received $9,582,000 within eight months by inducing investors to give him $100 for the promised repayment of $150. The case Cunningham v. Brown, 265 U.S. 1, 44 S.Ct. 424, 68 L.Ed. 873 (1924), discusses Ponzi's escapades and the resulting actions against him.

. Application note 1 states that a departure may be warranted if several section 2F1.1(b)(2) factors apply. U.S.S.G. § 2F1.1, comment, (n. 1). Application note 10 allows for departure if the loss "substantially exceeds” $5 million. U.S. S.G. § 2F1.1, comment, (n. 10).

. It’s hard to draw a valid vector of what would be appropriate with increased numbers [of victims], but I would suggest to you you start at point zero and you could take a minimum and a maximum that would go upward from that consideration.

For example, one could say, well, it said if it’s more than one [victim] you increase, so if it’s two to five, you could add two, if it’s six to ten you could add three, and so forth.

It’s my suggestion that it would still flatten out by numbers as it continued to get larger, but the number of people here is something that hasn’t been considered at all.

Now it is my thought that between two thousand and five thousand would come up to an enhancement of eleven points. I want to acknowledge that my own calculations here could be a little excessive just because I want to be cautious in what I think we are doing here, which is establishing new ground, that I have nothing to compare it with on previous *654 sentencings, so I’m trying to give it as much real thought as I can.

So I’m suggesting to you that a number of victims in the range of three thousand deserves an incremental addition of 10, which would give me a score of 31.

Transcript of Sentencing Hearing at 98-99.

. Section 5K2.0 of the Guidelines also explains that grounds for departure include any factors in addition to those identified by the Commission "that have not been given adequate consideration by the Commission.” U.S.S.G. § 5K2.0.

. This pre-November 1989 Guidelines application note allowing upward departure for crimes involving "more than minimal planning” and "more than one victim" has been superseded. Comment 1 to section 2F1.1 in the current Guidelines allows for upward departure only if the offense involved both "a misrepresentation that the defendant was acting on behalf of a charitable” or other such organization, and "violation of any judicial or administrative order, injunction, decree or process.” U.S.S.G. § 2F1.1, comment, (n. 1). Therefore, this ground for departure given by the district court would no longer be appropriate.

. The district court correctly noted that the new Guidelines table, though not applicable, also would indicate that f 7 million is not substantially excessive. The court concluded, however, that the dollar amount did contribute to the overall need for a large departure in this case.

The new table provides for increases at losses of $2.5 million, $5 million, and $10 million. The highest level for the new table is $80 million. The new table, however, does set a fourteen-point increase for a loss of $5 million— three points more than the table applicable in this case. This difference may serve as guidance in determining an appropriate degree of departure, but does not influence our decision regarding the appropriateness of departure based on the monetary loss in this case.