Ascente Business Consulting, LLC, doing business as LIBERTYID v. DR myCommerce, doing business as eSellerate; Digital River, Inc.

United States Court of Appeals
For the Eighth Circuit
___________________________
No. 20-2474
___________________________
Ascente Business Consulting, LLC, doing business as LIBERTYID
Plaintiff - Appellant
v.
DR myCommerce, doing business as eSellerate; Digital River, Inc.
Defendants - Appellees
____________
Appeal from United States District Court
for the District of Minnesota
____________
Submitted: May 12, 2021
Filed: August 18, 2021
____________
Before SMITH, Chief Judge, SHEPHERD and GRASZ, Circuit Judges.
____________
GRASZ, Circuit Judge.
Ascente Business Consulting, LLC hired DR myCommerce to build a web
portal so that Ascente could sell its identity-theft-protection services to consumers.
When the portal fell below Ascente’s expectations, it sued DR myCommerce and its
parent company, Digital River, Inc. (collectively, “Digital River”), for Minnesota

-- 1 of 14 --

-2-
contract, fraud, and fraud-adjacent claims. Ascente challenges the district court’s1
futility-based denial of its motion to amend some claims and the adverse grant of
summary judgment on others. We affirm.
I. Background
Ascente hired Digital River to build a customer-facing web portal. In May
2014, Ascente agreed to pay Digital River $73 per hour ($44,822 total) for its
services with a launch set for early October 2014, as set out in the First Statement of
Work. But setbacks twice delayed the launch. In a Second Statement of Work,
Digital River split the remaining tasks into two phases, prioritizing pre-launch tasks
over post-launch ones.
Eventually, the portal launched on October 29, 2014. The same day, Travis
Mills (Ascente’s President) “agree[d] and acknowledge[d]” that: (1) the portal
“satisfie[d] all requirements set forth in the Implementation Specifications”;
(2) Digital River “met its obligations with regard to the implementation of” the
portal; and (3) Ascente “approved” the portal and its launch.
Yet by December 2014, Ascente found problems with the portal. For
example, if a customer tried to create an account to buy the identity-theft-protection
services but exited before entering payment information, the customer could not ever
buy the services on return visits unless Digital River stepped in to help. And even
if a customer checked out, the portal “put [the customer’s] personal data at risk.”
Ascente also learned that multiple post-launch tasks remained unfinished.
The next month, the parties planned to meet in Denver to discuss Digital
River’s cost overruns. Before the meeting, Chad Johnson (Digital River) sent
Ascente an agenda outlining “the current status of our partnership,” stating:
1 The Honorable Joan N. Ericksen, United States District Judge for the District
of Minnesota, affirming the order of the Honorable Katherine Menendez, United
States Magistrate Judge for the District of Minnesota.

-- 2 of 14 --

-3-
a. Contract Date: 5/22/14 (2 year)
b. Site Live: 10/29/14
c. Digital River build hours YTD: 3180.25 @ $73 hr = $232,158.25
d. Ascente initial build fees invoiced: $44,822 (invoiced and scheduled
for payment)
e. Total orders YTD: 28
f. Total Gross Revenue YTD: $2950.90
Ex. 27, ECF No. 98-4 at 35.
At the Denver meeting, representatives from Ascente (Mills and two
consultants, Bret Busse and Elizabeth Sipple) and Digital River (Johnson and
Rodney Salazar) discussed the portal’s status. Tracking its agenda, Digital River
explained that the portal’s build time (and thus, its fees) ran over the budget by
$187,336.25. Ascente agreed to pay those fees. In exchange, Digital River agreed
to: (1) deliver the portal’s source code (the portal’s software in its computer-
programming language); and (2) complete the web-portal-specific services from the
existing Statements of Work. The parties planned to ink those terms in a Software
Development Agreement.
In February 2015, when Digital River sent its invoice, it excluded the hours
worked and the hourly rate. Digital River’s purposeful exclusion stemmed from a
debate between Johnson and his colleague Jennifer Manwarren over a two-dollar
difference in the hourly rate. Manwarren stated that she used “wild math” to reverse
engineer the invoice.
In July 2015, the parties signed the Software Development Agreement. While
the district court sealed that agreement, we describe relevant terms in a general way
to give some context for the contract that Digital River allegedly breached.
Under that agreement, Digital River needed to “develop Software and provide
services . . . (collectively, the ‘Services’)[.]” “Software” meant a “[c]ustomer-facing

-- 3 of 14 --

-4-
Web portal supporting an identity monitoring subscription service[.]” The
agreement listed Software as one of the “Deliverables” that Digital River needed to
“deliver or otherwise make available to [Ascente][.]” In very similar terms, the
parties twice equated the Software’s acceptance with Ascente’s receipt of the
Software.
Still within the Software Development Agreement, the Scope-of-Services
provision explained that through the Software, after “[c]ustomers enter[ed] the data
they wish to have monitored[,]” the data would be “supplied to a third party identity
monitoring service[.]” Another company would send emails. “In addition to” those
duties, “the Scope of Services expressly include[d] the scope” from the earlier
Statements of Work but “relating to the Web Portal only.” But Digital River would
have “no obligation to modify, fix, correct, update, enhance, or otherwise maintain
the Software.”
Beyond that limitation, the parties agreed that Digital River would provide
“ALL SOFTWARE, SERVICES AND WORK PRODUCT . . . ‘AS IS[.]’” And,
after generally “DISCLAIM[ING] ALL WARRANTIES,” the agreement expressly
stated that Digital River:
MAKES NO WARRANTY OF ANY KIND THAT THE SOFTWARE
OR WORK PRODUCT, OR ANY PRODUCTS OR RESULTS OF
THE USE THEREOF, WILL MEET [ASCENTE’S]
. . . REQUIREMENTS, OPERATE WITHOUT INTERRUPTION,
ACHIEVE ANY INTENDED RESULT, BE COMPATIBLE OR
WORK WITH ANY SOFTWARE, SYSTEM OR OTHER
SERVICES, OR BE SECURE, ACCURATE, COMPLETE, FREE OF
HARMFUL CODE OR ERROR FREE.
Ex. B, ECF No. 24 at 6.
In April 2016, Ascente finished paying Digital River. After Ascente spent
months refusing to accept the Software’s source code, Digital River delivered it via
flash drive in November 2016. Fourteen months later, Ascente sued.

-- 4 of 14 --

-5-
At the pleading stage, the district court dismissed Ascente’s fraud and fraud-
adjacent claims but not the contract claim. Discovery revealed Digital River’s
internal communications about Ascente, the parties’ working relationship, and the
portal.2 Based on those revelations, Ascente asked to amend its fraud, fraudulent-
inducement, and reckless-misrepresentation claims.
The magistrate judge considered some internal Digital River emails as
embraced by the pleadings. The magistrate judge concluded, and the district court
agreed, that futility barred Ascente from amending everything except one
fraudulent-inducement claim. But that claim did not survive summary judgment.
Nor did Ascente’s contract claim.
II. Discussion
Now, Ascente asks us to reverse the leave-to-amend denial and the adverse
grant of summary judgment.
A. Leave to Amend
Ordinarily we review leave-to-amend denials for an abuse of discretion, but
we review legal conclusions de novo. U.S. ex rel. Raynor v. Nat’l Rural Utils. Coop.
Fin., Corp., 690 F.3d 951, 957 (8th Cir. 2012). A futility-based leave denial “means
the district court has reached the legal conclusion that [Ascente’s] amended
complaint could not withstand a motion to dismiss under Rule 12(b)(6) of the
Federal Rules of Civil Procedure.” Munro v. Lucy Activewear, Inc., 899 F.3d 585,
589 (8th Cir. 2018) (quoting Cornelia I. Crowell GST Tr. v. Possis Med., Inc., 519
F.3d 778, 782 (8th Cir. 2008)). Taking Ascente’s allegations as true and construing
2 The emails showed that pre-launch, some Digital River employees saw the
then-existing software flaws as fatal. Employees also joked about and flouted post-
launch tasks. And during the Denver meeting, Digital River’s employees (who did
not attend the meeting) openly rooted for their company to cut ties with Ascente.

-- 5 of 14 --

-6-
all reasonable inferences in its favor, we will reverse if it can show that its proposed
amendments could have saved its otherwise meritless claims. See id.
Ascente asks us to reverse the district court’s futility-based leave denials of
its fraud and reckless-misrepresentation claims. Because Minnesota law treats even
the fraud-adjacent claims like fraud, we apply Rule 9(b)’s heightened pleading
requirements to each claim. See Trooien v. Mansour, 608 F.3d 1020, 1028 (8th Cir.
2010) (applying Minnesota law).
“In alleging fraud . . . , a party must state with particularity the circumstances
constituting fraud[.] Malice, intent, knowledge, and other conditions of a person’s
mind may be alleged generally.” Fed. R. Civ. P. 9(b). For each claim, Ascente
needed to “plead the ‘who, what, where, when, and how[.]’” Ambassador Press,
Inc. v. Dust Image Tech. U.S., LLC, 949 F.3d 417, 421 (8th Cir. 2020) (quoting U.S.
ex rel. Joshi v. St. Luke’s Hosp., Inc., 441 F.3d 552, 556 (8th Cir. 2006)); see also
Trooien, 608 F.3d at 1028, 1030 (noting a fraud claim must include “such matters as
the time, place[,] and contents of false representations” (quoting Bennett v. Berg,
685 F.2d 1053, 1062 (8th Cir. 1982))). It did not.3
More specifically, Ascente’s fraud claim needed to allege: (1) “a false
representation of a past or existing material fact susceptible of knowledge”;
(2) “made with knowledge of the falsity of the representation or made without
knowing whether it was true or false”; (3) “with the intention to induce action in
reliance thereon”; (4) “that the representation caused action in reliance thereon”; and
(5) “pecuniary damages as a result of the reliance.” U.S. Bank N.A. v. Cold Spring
3 We decline to review the underdeveloped reckless-misrepresentation claim.
See Dormani v. Target Corp., 970 F.3d 910, 916 (8th Cir. 2020) (“To be reviewable,
an issue must be presented in the brief with some specificity.” (quoting Meyers v.
Starke, 420 F.3d 738, 743 (8th Cir. 2005))). Additionally, while Ascente’s opening
brief asserts that the district court erred in denying leave to amend its fraudulent-
inducement claim, Ascente only discussed that claim in the summary-judgment
context.

-- 6 of 14 --

-7-
Granite Co., 802 N.W.2d 363, 373 (Minn. 2011). This appeal deals with the second
element—knowing misrepresentation.
In terms of relevant facts, the knowing-misrepresentation analysis comes to
an abrupt halt after the speaker makes the allegedly false statement. Put simply: we
only look at what the alleged fraudster knew when he or she made the
misrepresentation. See Martens v. Minn. Mining & Mfg. Co., 616 N.W.2d 732, 747–
48 (Minn. 2000) (“[T]he complaint is devoid of any specific claim that appellant’s
statements were representations the appellant knew were false or had no intention of
fulfilling at the time they were made.”).
Ascente says that its proposed amendments adequately alleged three separate
knowing misrepresentations. We disagree.
1. October 2014
First, Ascente says that its proposed amendments adequately alleged that on
October 28 and 29, 2014, Digital River committed fraud when Johnson and his
colleague Corey Husfeldt told Ascente (via Busse) that the web portal was ready for
commerce, fully functional, and met all of the requisite specifications. But those
amendments do not point to Johnson or Husfeldt. See Ambassador Press, 949 F.3d
at 421 (noting fraud claims must “plead the ‘who, what, where, when, and how’”
(emphasis added) (quoting Joshi, 441 F.3d at 556)).
Ascente asserts that it “need only allege that [Digital River] had the requisite
knowledge[,]” rather than Johnson or Husfeldt. Not so. “It is not sufficient to
attribute alleged false statements to ‘defendants’ generally.” Trooien, 608 F.3d at
1030.
Ascente also urges us to use general corporate-law principles to cloak Johnson
and Husfeldt in their colleagues’ statements. But Ascente waived that argument by

-- 7 of 14 --

-8-
raising it for the first time on appeal.4 See, e.g., St. Paul Fire & Marine Ins. Co. v.
Compaq Comput. Corp., 539 F.3d 809, 824 (8th Cir. 2008) (argument not raised in
the district court is waived).
Ascente views its amendments as adequately alleging that Johnson and
Husfeldt personally knew about the portal’s shortcomings. This argument rises and
falls with three email chains.
First, an October 3–6 Digital River email chain discussed the company’s
capability to meet the new deadline (October 29, 2014). Husfeldt (with Johnson
copied) asked colleague Jerel Thompson to “provide feedback on 10/29[.]”
Thompson confirmed that “[a]s long as the planned resources . . . do[n]’t change, we
can meet Oct. 29th from a Development perspective.” Manwarren replied to ask if
“that include[s] QA time or not?” Ascente did not tell us where to find that
question’s response.
Next, Husfeldt listed the pros and cons of two adjustment options and asked
if one adjustment “allow[s] for October 29th (including testing) to be the go live
date?” Tracing the chain chronologically leads to two affirmative responses, one
from Thompson and another from Matthew Kleinsasser.
Those email chains fit the timeframe (before the alleged misrepresentations).
See Martens, 616 N.W.2d at 747. And they involve Johnson and Husfeldt. But the
emails say nothing about what those two knew about the portal’s readiness for
commerce, its functionality, or whether it met the specifications.
For the October 2014 slice in time, Ascente’s only other knowing-
misrepresentation allegations come from a December 2014 email chain. Ascente
tries to use those emails to explain what Digital River (via Johnson or Husfeldt)
4 For the same reason, Ascente cannot use this argument to revive its January
2015 allegations.

-- 8 of 14 --

-9-
knew two months earlier.5 But because we can only look at what Johnson or
Husfeldt knew when they made the allegedly false statements, the December 2014
emails do not help Ascente. See Martens, 616 N.W.2d at 747.
For these reasons, Ascente failed to adequately allege that Johnson and
Husfeldt made knowing misrepresentations in October 2014.
2. January 2015
Next, Ascente says that its proposed amendments adequately alleged that
Digital River committed fraud at the Denver meeting in January 2015 by purportedly
misrepresenting its then-present intentions. Specifically, Ascente points to when
Digital River’s Johnson and Salazar agreed—without intending—to finish the
company’s work and deliver a functional portal if Ascente paid Digital River another
$187,336.25.
In support, Ascente relies on Digital River’s post-meeting actions, including
the undisputed fact that it stopped development work after the Denver meeting. No
record cite directs us to any pre-meeting actions that would hint at Johnson or
Salazar’s intent at the meeting.
Ascente also tries to marry Digital River’s purportedly “clear financial
motive” and the January 2015 misrepresentations. Ascente’s only case that involves
a “competing interest” does not equate a conflict of interest with an adequately
5 The December 2, 2014, email chain began when Sipple (an Ascente
consultant) asked Husfeldt (Digital River) about test accounts for the portal.
Husfeldt forwarded that email to his colleague Christine Roe for help responding to
Sipple. In turn, Roe forwarded it to Thompson and both weighed in. Dropping
Thompson from the chain, Roe said that “the cleanup job [for reconciling accounts]
wasn’t implemented.” But Roe “wasn’t aware of that” and said that it “[m]ust have
been a corner that was cut at the end. As far as [she] knew[,] [Thompson] had written
it.”

-- 9 of 14 --

-10-
alleged knowing misrepresentation. See Damon v. Groteboer, 937 F. Supp. 2d 1048,
1071–73, 1090 (D. Minn. 2013) (denying summary judgment on fraud claim against
realtors when facts could have supported an inference that realtor lacked intent to
market a property because doing so cut against his financial interest, he did not try
to lease units, and he claimed to have shown the building many times). At best,
Damon (a nonbinding district court case) suggests that a competing interest can
present a factor in the knowing-misrepresentation analysis—not the only factor.
For these reasons, the January 2015 statements cannot save Ascente’s fraud
claim.
3. July 2015
Finally, Ascente contends that Digital River committed fraud in July 2015 by
signing the Software Development Agreement without intending to perform it. In
support, Ascente relies on the “mountain of unfinished work” and post-execution
emails. This argument fails for several reasons.
First, the Software Development Agreement expressly contemplated Digital
River’s duty to finish its outstanding work. So, everyone (including Ascente) knew
that Digital River had more work to do. And Ascente concedes that the July 2015
misrepresentation turns on Digital River’s “intent not to perform,” not its capability
to do so. Yet Ascente does not explain how the “mountain of unfinished work”
translates into an intent not to perform an agreement to finish that exact work.
And second, Ascente cannot use emails that post-date the Software
Development Agreement to adequately allege that Digital River signed that
agreement without intending to perform it. See Martens, 616 N.W.2d at 747. So,
the July 2015 misrepresentation does not help Ascente’s fraud claim.
Even when we drew all reasonable inferences in Ascente’s favor, its proposed
amendments could not save its fraud claim. As a result, we agree with the district
court’s futility holding.

-- 10 of 14 --

-11-
B. Summary Judgment
Next, we review de novo the grant of summary judgment to Digital River on
Ascente’s breach of contract and fraudulent-inducement claims. See Paskert v.
Kemna-ASA Auto Plaza, Inc., 950 F.3d 535, 538 (8th Cir. 2020); Fed. R. Civ. P. 56.
1. Contract
Everyone agrees that Minnesota law governs the Software Development
Agreement and its incorporated Statements of Work.6 On the contract claim, we
only need to decide if Ascente raised a triable fact on the breach element.
The Minnesota Supreme Court’s “primary goal of contract interpretation is to
determine and enforce the intent of the parties.” Motorsports Racing Plus, Inc. v.
Arctic Cat Sales, Inc., 666 N.W.2d 320, 323 (Minn. 2003); accord Jensen v. Minn.
Dep’t of Hum. Servs., 897 F.3d 908, 913 (8th Cir. 2018) (applying Minnesota law).
When parties use unambiguous words to express their intent, Minnesota courts give
those words their plain meaning. Jensen, 897 F.3d at 913. However, a contract
becomes ambiguous if its plain terms are “reasonably susceptible to more than one
construction[.]” Id.
But “[i]nterpretation of unambiguous contracts is a question of law for the
court, as is the determination that a contract is ambiguous.” Staffing Specifix, Inc. v.
TempWorks Mgmt. Servs., Inc., 913 N.W.2d 687, 692 (Minn. 2018) (holding trial
court erred in allowing jury to decide “whether the contracts at issue were
ambiguous, rather than instructing the jury that the contracts were ambiguous”). “A
6 The parties’ contract arguments use common law and Uniform Commercial
Code concepts, interchangeably. Although the Software Development Agreement
involves both goods and services, we do not decide if common law or the UCC
applies. See Vesta State Bank v. Indep. State Bank of Minn., 518 N.W.2d 850, 854
(Minn. 1994) (using the predominant-purpose test for hybrid goods-and-services
contracts).

-- 11 of 14 --

-12-
contract’s terms are not ambiguous simply because the parties’ interpretations
differ.” Id. If a court views a contract as ambiguous, “it may admit . . . extrinsic
evidence of the parties’ intent” (anything beyond the contract’s four corners). Id.
“When extrinsic evidence has been admitted, the interpretation of ambiguous terms
becomes a question of fact for the jury.” Id.
At summary judgment, the district court concluded that only three (of the six)
purported breaches even implicated Digital River’s contractual duties. But to the
district court, no evidence raised a triable fact on whether Digital River breached its
purported duties.
Now, Ascente’s two contract-based challenges hinge on whether the district
court erred by reading the Software Development Agreement differently than
Ascente’s expert, Jonathan Hochman. As Ascente sees it, we can only understand
that agreement’s “highly technical” terms by looking at extrinsic evidence—namely,
Hochman’s opinion.
Ascente’s expert-based challenge skips several steps. See Staffing Specifix,
913 N.W.2d at 692–93. Hochman’s testimony cannot answer the legal question of
ambiguity. See S. Pine Helicopters, Inc. v. Phx. Aviation Managers, Inc., 320 F.3d
838, 841 (8th Cir. 2003) (expert testimony on legal matters is inadmissible). And
despite Ascente’s assertions otherwise, the district court’s Daubert decision on
Hochman—which limited his testimony to ecommerce software’s development—
did not empower Hochman to declare the Software Development Agreement
ambiguous. See Staffing Specifix, 913 N.W.2d. at 692.
Without any other substantive arguments, Ascente cannot convince us that the
district court erred in dismissing the contract claim.

-- 12 of 14 --

-13-
2. Fraudulent Inducement
“[A]bsent the essential element of reliance, an action for fraud must fail.”
Breezy Point Airport, Inc. v. First Fed. Sav. & Loan Ass’n of Brainerd, 179 N.W.2d
612, 615 (Minn. 1970). So, to defeat summary judgment on its fraudulent-
inducement claim, Ascente needed to raise a triable fact on its actual reliance. See
Hoyt Props., Inc. v. Prod. Res. Grp., 736 N.W.2d 313, 321 (Minn. 2007); accord
Popp Telecom, Inc. v. Am. Sharecom, Inc., 361 F.3d 482, 491–92 (8th Cir. 2004)
(affirming adverse grant of summary judgment on Minnesota fraud claim when
shareholders could not establish actual reliance).
Here, Ascente needed to show that a triable fact existed on whether it signed
the Software Development Agreement because it relied on Digital River’s
representations about the cost-overrun figure. See Hoyt, 736 N.W.2d at 321.
Ascente primarily looks to its president’s deposition excerpts for support. When
asked why he agreed to pay Digital River $187,336.25, Mills testified:
Because regardless of what our contract said, the agreement was made
between people. People that agreed to form a partnership whom I
believed were working ethically and on top of the table to continue that
partnership in good faith. I saw it as two professional organizations
that were in a partnership together, and that there would be a bigger
downside of me – for me not to pay my partner and keep them happy
and going than there would be to try to get away with something.
Digital River was my partner, and my values don’t allow me to even if
contractually get away with something, do it.
Ex. 1, ECF No. 122-1 at 7–8. Mills also testified that he “was willing to pay what
they told me that – that would make our relationship whole and ongoing. I was
willing to consider that.”
For the district court, Mills’s testimony resolved actual reliance against
Ascente. In its view, “[n]o reasonable jury could find that Ascente signed the

-- 13 of 14 --

-14-
[Software Development Agreement] in reliance on [Digital River’s] representation
of the amount of their cost overruns.” We agree.
The undisputed evidence speaks for itself. To preserve its business
relationship with Digital River, Ascente “was willing to pay what [Digital River]
told” Mills. Compare Popp, 361 F.3d at 491–92 (upholding adverse grant of
summary judgment when shareholders admitted that corporation’s pre-merger
activities did not induce them to sell their shares), with Hoyt, 736 N.W.2d at 321
(remanding for trial when lessor testified that he agreed to settle “because he relied”
on opposing counsel’s representations).
Ascente suggests that its “reliance is also confirmed by Johnson’s own
emails,” because the Digital River employee “noted”—to other Digital River
employees—“that if the invoice w[as] submitted with the wrong hourly rate, Ascente
‘will likely question it.’” We have no idea how Digital River’s internal prediction
about what Ascente might do tells us what Ascente actually did. And Ascente cites
no authority to convince us that Minnesota courts decide the actual-reliance
element—which looks at the alleged victim—by examining what the alleged
fraudster expected its victim to do.
Without raising a triable fact on its actual reliance, Ascente’s fraudulent-
inducement claim fails. See Breezy Point, 179 N.W.2d at 615. Consequently, we
conclude that the district court properly granted summary judgment on that claim.
III. Conclusion
For these reasons, we affirm the district court’s futility-based leave denial and
its grant of summary judgment to Digital River.
______________________________

-- 14 of 14 --