Kettler & Scott, Inc. v. Earth Technology Companies

JUSTICE COMPTON delivered the opinion of the Court. *452 In this action by an engineering firm against a land developer for recovery of sums allegedly due for services rendered, we limited the appeal to the question whether the trial court erroneously admitted in evidence certain computer records and daily work reports under the business records exception to the hearsay rule. Appellant Kettler and Scott, Inc., the defendant below, develops property by buying raw land, planning its development, obtaining rezoning, completing roads and underground utilities, and reselling sections of the development to others for residential and business uses. Appellee.Earth Technology Companies, Inc., the plaintiff below, provides geotechnical engineering services to developers, including testing and analysis of soil as well as inspections and certifications necessary for site development to meet design and governmental requirements. In May 1992, the plaintiff filed this action against the defendant, having nonsuited during trial a previous action against the defendant based on the same claim. In the present case, the plaintiff sought recovery of $218,265.85 allegedly due as the result of defendant’s failure to pay for engineering services rendered under contracts related to development of a large mixed-use community in Prince William County known as “Wellington.” There was no dispute that plaintiff had performed work under the contracts nor was the quality of plaintiffs work at issue. Rather, defendant declined to pay the disputed amount claiming “it was impossible for [defendant] to determine whether the hours claimed actually had been worked” and whether certain lab tests actually had been performed. Additionally, defendant asserted that if the tests actually had been performed, “it was impossible for [defendant] to determine whether they had already been paid or even whether these were tests which were actually required by the contracts.” Following a three-day bench trial, the court found for the plaintiff, fixing the damages at $148,711.00 with interest. We awarded the defendant this appeal from the May 1993 judgment order. According to settled appellate principles, we shall summarize the relevant facts in the light most favorable to the plaintiff, the prevailing party below. During the 1980s and prior to development of Wellington, the parties had worked together on various projects. The business relationship was informal and there were virtually no written contracts involved. The plaintiff charged “unit prices” for performing laboratory tests and hourly charges for *453 engineers’ time spent in the field. The plaintiffs billing for work done was irregular at defendant’s request so that defendant could time its discharge of the obligation to coincide with the sale of a particular section of developed land. In 1988, as the development of Wellington commenced, the parties employed written contracts because of a new accounting system defendant had established to better allocate “costs associated with different services.” The contracts took the form of brief, one or two-page documents with virtually no terms, giving an estimate of the work to be performed by plaintiff based on a fixed amount per hour worked or per test performed. There was a separate contract for each residential and commercial section, each drainage structure (such as a detention pond), and each major road in the project. There were 25-30 contracts between the parties, of which 16 are the subject of this controversy. Even with written contracts in place, the parties continued their informal course of dealing with respect to billing procedures. Plaintiff performed necessary testing, engineering, and inspections to obtain governmental approval, with billing being submitted irregularly for field work and some lab work. Defendant did not always insist that supporting data for the hourly charges or testing accompany the billing invoice. Defendant’s main interest was that all necessary work be done to obtain governmental approval and that the required certifications be provided to defendant’s design engineers or appropriate authorities. In June 1990, plaintiff learned of the potential sale of one of the sections and met with defendant’s personnel to discuss “unpaid invoices.” The plaintiffs work under the contracts was virtually complete with no question having been raised about the quality of plaintiffs performance. During the meeting, Roy K. Anderson, plaintiffs financial officer in charge of contracts, records, billings, and accounts receivable, informed defendant that the bills outstanding did not represent “all the monies that are owed. You have a lot of unbilled time here, unbilled items.” Defendant’s representatives told plaintiff to submit “final bills.” During the two years of work on Wellington, plaintiffs billing system was a combination of modern computer technology and manual posting. As we have said, the services billed had two components: laboratory soil testing and employees’ hours worked. The lab tests were partially computerized. Test results were posted by hand on forms in the lab. Those report forms were for *454 warded to the business office, which was supervised by Anderson. Anderson’s staff put the test results into a computer as a regular part of their duties. The computer was programmed to sort the test results and to print a report form. The report form generated would be reviewed by the engineer, who issued the necessary certification based on the report. To save space, no hard copies of the reports were kept by plaintiff after the reports were submitted to the concerned authorities. But when an invoice was prepared to be submitted to defendant, an additional copy of the test report would be printed, attached to the file copy of the invoice, and placed in plaintiff’s contract file. The computer program prepared only reports, not billing invoices. In order to bill for lab work, Anderson or his staff counted each type of lab test reported on the computer-generated report and manually multiplied that number by the applicable unit price for such test. The hours worked were not computerized. They were maintained primarily on a form called a “daily report.” In addition to the daily report forms, plaintiff’s engineers and lab technicians recorded their time in personal journals or on available pieces of paper. Whatever the form of the individual record, Anderson placed it in a file for a specific contract. When Anderson prepared a bill, he manually sorted the hourly records according to category, and assigned the unit price for the particular level of work to those hours. The invoice, which included the lab report and the hourly charges, was prepared manually by Anderson’s staff, with the hourly records returned to the file. Ordinarily, a bill was prepared by gathering all the hourly records and printing all the lab test reports for a particular contract and comparing this information against the file records of what already had been billed on that contract to avoid “double billing.” Following the June 1990 meeting, Anderson and Patty Endicott, the data processor and billing specialist employed by plaintiff, began the normal document review to identify all unbilled lab tests and hourly charges. The final hourly billings were prepared based on the records described above. The final billing for the lab tests was more complicated. Some time before the June meeting, plaintiff had retained a computer programmer to improve plaintiff’s computer software. Instead of improving the system, the programmer destroyed the *455 program that sorted the lab data and generated the lab reports. According to custom, no hard copies of unbilled lab time had been retained. In attempting to retrieve this information, Endicott discovered that the lab data originally put into the computer memory. was still available in the memory on a “back up” disc. Even though the program to print reports was nonexistent, Endicott was able to print the raw data in what, at a glance, appears to be a mass of words, numbers, and symbols. These printouts, covering approximately 91 pages, were received in evidence. Endicott studied these printouts and discovered the code number assigned by the computer program to sort the information by contract. In addition, she was able to identify plaintiffs abbreviations and numbering for the various test results, including defendant’s contract numbers. Armed with the sort code, Endicott manually sorted and reported, like the destroyed program would have done mechanically, all the test results in the memory. In order to determine which test results appearing on the 91-page printout had been billed, and which had not, Anderson and Endicott engaged in a time-consuming process. As we have noted, the file for each contract had the previously billed test results attached to a file copy of the invoice. Each test result on the printout was compared against all previously billed invoices for that particular contract and identified as billed or unbilled. Endicott said she could determine “with a high degree of certainty” what had not been billed. She then sorted the test results into a list of billed and unbilled items. From the unbilled list, she printed a separate list of all unbilled tests for each contract. These unbilled test results were counted manually by Anderson and Endicott, the unit prices were applied, and the final invoices prepared. If either person had a question whether a test had been billed previously or had been paid, charges for the test were omitted from the final invoices. Accordingly, the final billing for each contract consisted mainly of three items: all unbilled lab tests, all previously unbilled hourly charges, and all previously billed but unpaid invoices. The final bills were submitted to defendant in September 1990. Defendant refused to make any payments without being furnished documentation, that is, all test results and all hourly records, taking the position that such documentation was a condition precedent to payment under the contracts. On conflicting evidence, the trial court found as a fact that submission of documen *456 tation with the invoices was not a condition precedent to payment under the contracts; and we refused an assignment of error attacking that ruling. After suit was filed and during pretrial discovery, the plaintiff provided defendant with the printouts of the lab results, test summaries prepared by Endicott, and a summary of the entire claim, showing the testing and hourly claims for each contract. These documents were among the mass of exhibits received in evidence at trial. But not all the hourly records and paid invoices were available pretrial. Many of these records had been lost when plaintiff was evicted from its offices after the submission of the final billing. Additionally, a portion of the final billing included charges for a “shear test.” The lost files included records for those tests. As we have said, the final claim submitted by plaintiff at trial was in the amount of $218,265.85. Daily reports for hourly charges were missing to support $17,755.00 of the claim. The total sum included for the shear tests was $51,800.00. The trial court allowed recovery for all the documented lab work, hourly charges, and unpaid invoices; the court denied recovery for the undocumented hourly charges and shear tests, thus making the award $148,711.00. On appeal, defendant contends that the 91-page printout of “raw computer data” is “the only foundation supporting the invoices for lab tests valued at $105,461.00 which are included in the judgment.” Defendant argues that “Endicott used this confusing mass of computer data, supposedly reflecting portions of destroyed paper test results which allegedly had been created in 1988, 1989 and 1990, to guess which tests had been run, and she then compared her guesswork with documentation attached to old invoices previously paid by [defendant] to guess again regarding whether the tests already had been billed.” “Unfortunately,” the defendant continues, “this raw computer data, an incomplete copy of the original computer records, did not contain enough information for Endicott or any other witness to guess whether the work which allegedly was done actually was within the contracts.” Summarizing, defendant argues that these exhibits “are not the sort of verifiable regular business entries which are trustworthy and therefore can be admitted into evidence. They should have been excluded as impermissible hearsay.” We do not agree. *457 In determining the admissibility of computer records, when the argument has been advanced that they are inadmissible hearsay, we have employed the traditional business records exception to the hearsay rule. Frye v. Commonwealth, 231 Va. 370, 387, 345 S.E.2d 267, 279 (1986). See generally Randy Snyder, Note, Assuring the Competency of Computer-Generated Evidence, 9 Computer/L.J. 103 (1989); James A. Sprowl, Evaluating the Credibility of Computer-Generated Evidence, 52 Chi.-Kent L. Rev. 547 (1976); Donald M. Zupanec, Annotation, Admissibility of Computerized Private Business Records, 7 A.L.R.4th 8 (1981). See also Jerome J. Roberts, A Practitioner’s Primer on Computer-Generated Evidence, 41 U. Chi. L. Rev. 254 (1974). “Under the modern Shopbook Rule, adopted in Virginia, verified regular entries may be admitted into evidence without requiring proof from the regular observers or record keepers,” generally limiting admission of such evidence to “facts or events within the personal knowledge of the recorder.” "Automatic” Sprinkler Corp. v. Coley & Peterson, Inc., 219 Va. 781, 792, 250 S.E.2d 765, 773 (1979). However, this principle does not necessarily exclude all entries made by persons without personal knowledge of the facts recorded; in many cases, practical necessity requires the admission of written factual evidence that has a circumstantial guarantee of trustworthiness. Id. “The trustworthiness or reliability of the records is guaranteed by the regularity of their preparation and the fact that the records are relied upon in the transaction of business by the person or entities for which they are kept.” Id. at 793, 250 S.E.2d at 773. “Admission of such evidence is conditioned, therefore, on proof that the document comes from the proper custodian and that it is a record kept in the ordinary course of business made contemporaneously with the event by persons having the duty to keep a true record.” Id. Accord Marefield Meadows, Inc. v. Lorenz, 245 Va. 255, 264, 427 S.E.2d 363, 368 (1993); Ashley v. Commonwealth, 220 Va. 705, 707-08, 261 S.E.2d 323, 324-25 (1980). This approach “necessarily requires that a determination as to admissibility be made on the facts of each case.” "Automatic” Sprinkler Corp., 219 Va. at 793, 250 S.E.2d at 773. Dealing first with the lab test charges, we are satisfied from a study of this record that the disputed, written, factual information admitted in evidence has the required circumstantial guarantee of trustworthiness. In the present case, there is credible evi *458 dence to support the trial court’s explicit and implicit findings that the records were prepared, and regularly relied upon, in the ordinary course of business under the supervision of Anderson, the custodian, and that the entries were promptly made at or near the time of the event by plaintiffs employees having the duty to keep accurate records. The focus of the defendant’s attack is upon the lab test data placed into the computer memory, the 91-page printout produced from the back-up disc, and the retrieval and interpretation of the data. Thus, it becomes important to examine the flow of information into, through, and out of the computer. Here, of course, we are not concerned with one step in the normal production of computerized evidence, that is, employment of the software or the computer program that dictates the actions and calculations made by the computer. That link in the usual process had been destroyed. When examining the flow of information, we shall not repeat in detail what we already have recited in summarizing the facts. It is sufficient to point out that plaintiffs employees as a regular part of their duties at each step in the process: posted the lab test results by hand on forms in the lab; forwarded those forms to the business office, which was supervised by Anderson; and, using the forms, put the test results, which were reliable and trustworthy, into the computer memory. Because the program to retrieve the test results from the computer was destroyed, the raw data was printed in a form that enabled Endicott and Anderson, utilizing records of previously billed invoices, to determine “with a high degree of certainty” what had not been billed. Endicott actually performed manually the function of the destroyed program when she assembled information employing the raw data printout. The only difference between the previous billing procedures and the final billing procedures, as the plaintiff points out, was that the lab test reports could not be reprinted but had to be sorted manually from the printed computer memory instead of automatically by the software program. There is no hint in the record that the plaintiffs computer hardware malfunctioned, or that the data Endicott says she retrieved from the printouts had been altered in any way from the time the data had been collected and put into the computer to the time when the information translated by Endicott was retrieved *459 from the back-up disc. The defendant’s attack on the reliability of the information assembled by Anderson and Endicott to create the final billings goes merely to the weight of their testimony about the accuracy of the methodology used, and does not affect the admissibility of the disputed test-result evidence as a whole. Turning to admission of the daily reports supporting the hourly charges, which defendant says represents $43,250.00 of the judgment, we reject defendant’s contention that they are not within the business records exception because they were not shown to be contemporaneously prepared, “and it would not have been inconvenient for the preparers of the reports to authenticate them personally and lay the proper foundation.” Under the circumstances of this case, practical necessity dictated presentation of the hourly records in the form offered by the plaintiff, that is, through Anderson, the custodian, instead of parading to the witness stand every engineer and lab technician who worked on the Wellington project. And, as our recitation of the evidence on this issue has already demonstrated, this evidence also has the required guarantee of trustworthiness. The daily reports, supplemented by the journals and paper memoranda, were created at the time of the work, placed by Anderson in separate files for each contract, and manually sorted later when the bills were prepared, either during the normal course of the project or at the time of the final billing. Any complaint by the defendant about these records merely affects the weight the trier of fact would accord them, and not whether they are admissible. Consequently, we hold that the trial court did not err in admitting any of the disputed records, and the judgment below will be Affirmed.