Why Quarterly Client Reports Are Still a Data Entry Problem

Every RIA sends a quarterly report. It's not optional — the custody rule requires clients get an itemized statement at least that often. What's optional, in practice, is how much of that process is still a person copying numbers from one screen to another. At most mid-market firms, the honest answer is: most of it.

We wanted to know why, specifically, before writing anything about fixing it. The popular explanation — "the software is old and bad" — turned out to be half right and half wrong, and the actual answer is more useful than either version.

The part of the popular explanation that's true

The core platforms most firms report out of really are old. Orion was founded in 1999. Black Diamond and Tamarac both trace back to around 2004, before Black Diamond was acquired by SS&C in 2014. These aren't fly-by-night tools — they're mature, widely used platforms that have had two decades to build out real reporting depth, and by most customer accounts, they've done that well. Orion carries a 4.4 out of 5 rating across 400 reviews on G2. Black Diamond's client-facing reporting is generally considered some of the best in the industry.

So "the reporting engine is broken" isn't really the story, whatever the age of the code underneath it. The story is what has to happen before any of that reporting engine ever sees clean data to work with.

The part that's actually true: these tools were built for one custodian, and firms don't run one custodian anymore

Nearly 30% of RIAs now run two or more custodians at once, according to a 2025 analysis of Form ADV data across more than 23,000 SEC-registered firms — up from the prior year, and part of a real, ongoing shift as firms hedge against outages, service problems, and custodian mergers. A platform built around a single clean custodial feed handles that reasonably well. A platform being asked to reconcile positions, cost basis, and performance across two or three custodians with different data formats, different update schedules, and different quirks in how they represent the same information — that's a much harder problem, and it's the one actually eating weeks of a reporting cycle.

It shows up in unglamorous ways. One G2 reviewer described just getting their firm's data organized and uploaded into a reporting platform as a six-month process that "stressed many people out" — not because the software was defective, but because pulling clean, consistent data out of multiple source systems and getting it into a shape the platform could actually use is genuinely hard, one-time or every quarter.

What a quarter actually looks like for the person building the report

Pull positions and transaction history from each custodian a household touches. Reconcile it against what the portfolio management system thinks it holds, because those two numbers don't always agree, especially right after a transfer or a corporate action. Check cost basis, which is its own minefield — brokers only have to report it accurately for securities bought after 2011, so anything older often needs a manual check against the firm's own records. Pull fee data and verify it against what was actually billed. Then, and only then, does the report itself get assembled — and for firms without a fully integrated stack, "assembled" often still means someone building slides or a document by hand, quarter after quarter, household by household.

Take a household with an IRA at one custodian and a taxable account at another — common enough that it's closer to normal than an edge case at this point. The two accounts update on different schedules. A dividend posts to one a day before it shows up in the consolidated view the ops person is working from. A transfer mid-quarter means the position history has a gap that has to be manually bridged so the performance number doesn't look wrong. Multiply that by however many households at the firm have a similar setup, and "build the quarterly reports" quietly becomes a multi-day reconciliation project before anyone has actually started writing anything a client will read.

None of that is judgment work. It's the same retrieval-and-reconciliation problem that eats an annual compliance review, just running four times a year instead of once, and touching every single client relationship instead of a sample of them.

Why this is worse than it looks on a task list

A quarterly report isn't just an operational deliverable — for most clients, it's the single most regular, tangible proof that their advisor is actually paying attention. A report that's late, or visibly wrong, or clearly a boilerplate template with the numbers swapped in, costs more trust than the hours it took to build it would suggest. And the households most likely to expose a data-reconciliation problem — the ones with money spread across several custodians and account types — are usually the same households generating the most revenue for the firm. The reporting process is hardest exactly where getting it right matters most.

What's actually solvable here, and what isn't automation's job

The retrieval and reconciliation layer is genuinely automatable, and increasingly is: a system that pulls position and transaction data from every custodian a household uses via API where one exists, or by reading a statement directly where it doesn't, reconciles it against the portfolio management system, flags anything that doesn't tie out, and assembles the underlying numbers into a consistent, verified data set before anyone has to look at it. That's the same document intelligence problem we've written about elsewhere on custodian paperwork — reading unstructured statements reliably and turning them into structured, checkable data — applied to a recurring quarterly cycle instead of a one-time transfer.

What that system shouldn't do is decide what the report says about the client's plan or write the narrative explaining a hard quarter — that's a separate, harder problem, and one where an advisor's judgment about their specific relationship with that specific client still matters more than anything a model can generate on its own. The data layer and the narrative layer are different problems with different right answers. Solve the data layer first; it's the one actually costing weeks.

This is already happening — just mostly at the top of the market

A handful of data-infrastructure vendors have already built real products around exactly this problem: pulling positions from custodians and other sources via API or document extraction, harmonizing the formats into one consistent schema, and handing a clean data set to whatever reporting layer sits on top. That's a genuine, working category now, not a hypothetical. What it mostly isn't, yet, is built for or priced for a firm managing a few hundred million dollars with a lean ops team. Most of what we found targets large RIAs, family offices, and enterprise platforms with the budget and technical staff to run a full data-infrastructure buildout.

That's the actual gap for a mid-market firm: the underlying problem is solved technology, proven at the top of the market, and not yet something you can reasonably buy off a shelf at a price and complexity that fits a 20-person firm. That's a build problem, not a research problem — the pattern is known, it just has to be assembled for a firm your size instead of a firm ten times it.

Frequently Asked Questions

Why do quarterly reports take so long if the software is supposed to handle it?

Because the software generates a report from data it's given — and getting clean, reconciled data out of two or three custodians with different formats and update schedules is the actual bottleneck, not the report-generation step itself. Nearly a third of RIAs now run multiple custodians, which most reporting platforms weren't originally built around.

Is quarterly reporting actually a regulatory requirement?

Yes. The SEC's custody rule requires that clients receive itemized account statements, including all disbursements and advisory fees, at least quarterly. Most firms send them quarterly specifically to satisfy this baseline.

Are Orion, Black Diamond, and Tamarac actually bad software?

Not by the evidence we found. All three are mature platforms with generally strong customer satisfaction — Orion holds a 4.4 out of 5 rating across 400 G2 reviews, and Black Diamond's reporting output is widely regarded as strong. The complaints that hold up are about data onboarding and multi-source reconciliation, not the reporting engines themselves.

Can this be automated without losing the personal touch clients expect?

Yes, if it's scoped correctly. Automating the data-gathering and reconciliation work doesn't touch the parts of a report that should stay personal — the narrative, the context, the advisor's read on what a quarter meant for that specific client. It just means the advisor isn't spending their limited time re-typing numbers instead of writing to their client.

See what this is actually costing your team

We built a free calculator that estimates what manual document and data handling costs a firm per year, using industry-standard per-document rates. It takes about two minutes and doesn't require your email to see the number. Try the Document Processing Cost Calculator, or if you'd rather just talk through what your own reporting cycle actually looks like, book 30 minutes with us.