Preparing Returns for Friends Was Never Supposed to Get Complicated

The situation

The plan had always been small and manageable. Vartan worked full time as a sales director for a manufacturing distributor, and on evenings and weekends he ran a modest corporation on the side preparing tax returns for a handful of people he knew well: family, coworkers, and a few friends who trusted him because he was careful and because he charged less than an accounting firm would. Lusine, a police sergeant, had been a client for six years and never had an issue. Adnan, who ran a small corporation doing residential renovation work, had been a client for three, and his returns were always a bit more work, because Adnan kept his own books in a mix of spreadsheets, paper receipts, and text messages to himself, and handed the whole pile over each spring for Vartan to make sense of.

That arrangement had worked well enough for three tax years. Vartan would take Adnan's records, ask a few clarifying questions, and file a return that matched what Adnan told him about the business. When the Canada Revenue Agency selected Adnan's corporation for an audit, the auditor found that a substantial amount of revenue had gone unreported over two of those years, along with a set of expense claims that did not match anything in Adnan's own paperwork. Adnan was reassessed. Then, months later, so was Vartan personally, under a provision that allows the Agency to penalize a person who prepares or helps prepare a return containing a false statement, if that person knew or should reasonably have known about the problem.

The number attached to Vartan's penalty was close to $280,000, calculated as a percentage of the tax Adnan's corporation had understated across the two years in question. For a man whose side business existed mainly to help people he liked and to earn a bit of extra income on evenings he would otherwise have spent doing something else, it was a figure that did not match his sense of what he had actually done. He had not built the numbers. He had typed up what Adnan gave him, asked the questions that seemed relevant at the time, and filed what looked like a complete and consistent return.

The problem was proving that distinction to an auditor who, understandably, could not tell from the file alone where Adnan's record-keeping ended and Vartan's preparation began. From the outside, a preparer penalty and an ordinary reassessment can look like the same finding wearing two different names, and untangling them meant going back further than either the audit or the original penalty notice had bothered to go.

What the law actually said

The penalty provision the Agency relied on is not aimed at every preparer whose client turns out to have misreported something. It targets a narrower group: preparers who knew about a false statement, or who were wilfully blind to circumstances that should have made the problem obvious, or who behaved with a degree of carelessness so far below what a reasonable preparer would do that it amounted to the same thing. A preparer who reasonably relies on records and representations a client provides, without obvious red flags suggesting those records are unreliable, is not the target of the provision even if the client's numbers later turn out to be wrong.

That distinction mattered enormously here, because the whole case against Vartan rested on an implicit assumption: that the gap between what Adnan's business actually earned and what got reported was something a competent preparer working from Adnan's records should have caught. Whether that assumption held depended entirely on what those records actually showed at the time Vartan prepared the returns, not on what a full audit years later revealed once every bank statement and invoice had been pulled together and cross-referenced by professionals with tools and time Vartan never had.

The law also drew a line we thought the Agency's initial position blurred: reasonable reliance on a client's own account of their business is different from reasonable reliance on numbers a preparer had reason to doubt. If Adnan's spreadsheets and receipts, as handed to Vartan each spring, were internally consistent and did not flag obvious gaps, then Vartan's job was to prepare from what he had, not to conduct his own investigation into whether Adnan was telling him everything. Preparers are not expected to audit their clients. They are expected to use reasonable judgment about the material actually in front of them, and the standard the Agency has to meet before penalizing a preparer personally is meaningfully higher than the standard for reassessing the underlying taxpayer.

Where the case would turn, we understood early, was not on abstract argument about the standard but on a concrete reconstruction of exactly what Adnan had given Vartan, year by year, and whether anything in that material should have raised a flag a careful preparer would have noticed.

What we did

1. Requested the complete preparer file, not just the returns. We obtained everything Vartan still had from each of the three tax years, including emails, the spreadsheets Adnan had sent, and notes from their conversations, because the penalty case would ultimately depend on what Vartan actually saw at the time, not on what the audit later uncovered using records Vartan never had access to.
2. Rebuilt the accounting from Adnan's own source material. Working from the spreadsheets and receipts Adnan had provided, we reconstructed what a preparer working only from that material would reasonably have concluded about the business's income and expenses in each year, which let us compare that reconstruction directly against what the audit had eventually found using additional records Vartan never saw.
3. Identified where the discrepancy actually originated. The reconstruction showed that the unreported revenue came from cash payments Adnan had never recorded anywhere in the material he gave Vartan, not from a miscalculation of numbers that were already present in the file. That distinction was the central fact of the whole case, and establishing it credibly took real accounting work tracing each disputed dollar back to its source, not argument or assertion about what a reasonable preparer generally does.
4. Documented the absence of red flags in the records themselves. We prepared a year-by-year summary showing that the spreadsheets Adnan provided were internally consistent, arithmetically sound, and did not contain the kind of obvious gaps, round numbers, or contradictions between documents that would have put a careful preparer on notice that something was missing. That summary supported the reasonable-reliance argument directly, because it answered the question the Agency actually needed answered.
5. Prepared a written submission distinguishing Vartan's role from Adnan's conduct. We set out plainly, with the reconstructed timeline attached, that Adnan's under-reporting was a failure to record income in the first place, occurring entirely before any material reached Vartan, while Vartan's role was limited to preparing a return from what he was actually given. That distinction framed the penalty question correctly for the reviewing officer instead of leaving the two conducts blurred together.
6. Negotiated the applicable penalty percentage down. Once reasonable reliance was established for the bulk of the disputed amount, we argued that a smaller residual portion, tied to one expense category where Adnan's material had been genuinely ambiguous rather than simply incomplete, was the only part of the file where any preparer carelessness was even arguable, and negotiated the penalty down to reflect that narrower slice rather than the full amount originally assessed against Vartan.
7. Advised Vartan on his side business going forward. We recommended he formalize an engagement letter and a records checklist for any future preparation work, not because anything he had done this time was improper, but because a documented intake process makes reasonable reliance far easier to demonstrate quickly if a client's records are ever challenged again, rather than needing a costly reconstruction after the fact.

The outcome

The Agency accepted the reconstructed accounting and agreed that Vartan had reasonably relied on Adnan's own records for the large majority of the understated amount. The penalty was reduced from roughly $280,000 to just under $40,000, tied to the one expense category where the material Adnan provided had been genuinely unclear rather than simply incomplete, and where a more careful preparer might reasonably have asked another question before filing. The reduction was not a technicality; it reflected an actual line drawn between what Vartan saw and what Adnan never showed him.

Vartan paid the reduced amount and kept his side business running, though on a smaller and more formal footing than before, with fewer clients and a clearer paper trail for each one. Adnan's own reassessment was unaffected by any of this; his liability for the unreported income stood on its own and was never seriously in question, since the audit had traced the cash payments directly to his business activity regardless of who prepared the return. What changed was narrowly and specifically Vartan's personal exposure for someone else's under-reporting, once the file showed clearly where the responsibility for each part of the discrepancy actually sat.

Lusine, whose returns had never been in question at any point, stayed on as a client, as did most of the others who had used Vartan's services over the years without incident. For Vartan, the episode reshaped how seriously he treated a business he had always thought of as informal and low-risk. Preparing returns for people he trusted turned out to carry real personal exposure if a client's own records were incomplete, and the only protection against that exposure, it turned out, was a clear, contemporaneous, documented account of exactly what he had actually been given to work with at the time each return was filed.

What you can learn from this