Last month a fellow practitioner watched me onboard a new client over a screen share and said something I get a lot. "You're still doing this in Excel?" The file was a Xero export. My tool was a sheet with the trial balance pasted in and four columns of working notes alongside it. The assumption among cloud-fluent accountants is that you should be in Syft, Fathom, or Spotlight from minute one. I disagree.
TL;DR: Syft, Fathom and Spotlight Reporting are excellent once you know what story a client's numbers are telling. Excel is for figuring out what that story is. On a new engagement, the first 90 minutes belong in a flexible grid, not a structured reporting layer.
Why Does Excel Still Win the First 90 Minutes?
Because it's the only tool in the stack that doesn't make you commit to a question before you ask it.
Syft, Fathom and Spotlight all need you to have picked a question. A dashboard, a forecast horizon, a KPI band. Since 14 January 2026, Xero ships Analytics powered by Syft embedded in its Comprehensive plan and above (Xero Central, 2026), and Spotlight Reporting has been the dominant UK management report tool for over a decade (Acuity Magazine, 2024). Each is excellent at what it does. None is built to diagnose a file you've never seen.
The first 90 minutes of a new engagement are pre-structure. I'm pulling a trial balance into a flexible grid and watching how lines move. The debtors line that's drifted up four months running. The miscellaneous expenses code with three round-number entries against it. The credit card reconciliation that's nine months stale. That's diagnostic work, the same instinct I picked up in audit (see the bank-feed hygiene check I still run on every new file).
ICAEW's CEO Alan Vallance, opening the institute's Twenty Principles for Good Spreadsheet Practice, puts it more politely than I would: despite rapid technological leaps in accountancy, spreadsheets "appear as important as ever."
Doesn't This Just Mean I Haven't Learned Syft Properly?
No. I'm certified, I use it weekly for close packs and consolidated views, and I like the product. Syft scores 9.0 for reporting hierarchy and 9.3 for report templates on G2; Fathom rates higher on ease of setup; Spotlight has the most refined pre-built UK report layouts. Each is a strong second-stage instrument.
The category mistake is treating any of them as a first-stage instrument. Once you know which story you're telling, structured reporting is faster and more client-presentable than anything you'll hand-build. But you can't decide the numbers look right inside the tool reporting them. That work happens upstream, in a grid that lets you ask three questions before you commit to one.
What Does the First 90 Minutes Actually Look Like?
A trial balance pulled into Power Query, shaped, and dropped into a tab with comparative columns for the last three or six month-ends. A variance check. A column for "do I expect this to move?" which is directional expectation written down. A column for "what would a competent reviewer ask about this?" which is audit documentation discipline applied to my own work.
The profession agrees this is real work. ACCA runs a dedicated Power Query course for finance professionals; AccountingWEB has been publishing long-form pieces on the Power Query trial balance workflow for years. The cloud-only practitioners I see getting stuck in the first month of a new engagement are the ones who reached for the reporting tool before they had a hypothesis worth reporting on.
When Should I Move into Syft?
When the diagnostic work is done. When you know what story the file is telling, what's clean and what's still being argued with, and what your opening conversation with the client needs to cover. At that point the tool's job is presentation: make the output legible to the client, the board, or the lender. The thinking happens upstream.
Roughly 57% of entry-level accountants lack advanced Excel or Power BI modelling skills (SoftwareOasis, 2025). For that cohort, "Excel first" reads as gatekeeping. It isn't. The diagnostic skill belongs upstream of the reporting tool, and three months with Power Query is a better investment for a junior than three months with any dashboard product.
For my files the rule stays simple. First 90 minutes in Excel; everything after that in whichever AI-augmented cloud tool actually ships the feature the client needs. The grid and the dashboard do different jobs, in sequence.
Frequently Asked Questions
When should I use Excel versus Syft, Fathom or Spotlight for client analysis?
Use Excel during the first diagnostic phase of a new engagement, when you're pulling a trial balance, comparing month-on-month movement, and figuring out which lines don't look right. Move to Syft, Fathom or Spotlight Reporting once you know the story and need a structured output for the client, the board, or a lender.
Is it old-fashioned for a cloud accountant to still use Excel?
No. ICAEW maintains a long-running Excel Community precisely because spreadsheets do diagnostic work no reporting layer replicates. The discipline lies in knowing when a flexible grid is the right tool and when a structured reporting product is.
What kind of analysis is Excel actually faster at than Syft or Power BI?
Anything pre-structure. Month-on-month variance scans, ad-hoc grouping, applying a directional-expectation column, sense-checking a trial balance you've never seen before. Structured tools require you to commit to a question first; Excel lets you ask several before you commit.
If you want a second pair of eyes on a new file before you commit it to a reporting layer, get in touch. Happy to walk through what a first-90-minutes diagnostic looks like for the kind of clients you take on.