AccountingWEB

The screen-scraping segment carries real operational value - explaining why firms use banned tools (HMRC's own API gaps), what the detection mechanism looks like, and what questions to ask vendors. But this is sandwiched by a lengthy news recap of the P&L timeline and a generic soft-skills segment, plus several minutes of football banter and podcast housekeeping that contribute nothing.

“So if you're not using technologies like this then to get a full picture of your client base up, uh, to date picture should I say? You either have to manually piece this information together or you fly blind and hope for the best.”

“accounting firms can't use software to pull down client details like uh, penalties, uh, interest or specific balances on account. So it can't access the information via software and it can't aggregate them.”

The episode's most original framing is the argument that HMRC's own failure to build agent-facing APIs is the structural cause of the screen-scraping problem - that's a coherent, non-obvious inversion of the usual blame narrative. Everything else, including the P&L timeline recap and the soft-skills discussion, is conventional news commentary and generic career advice.

“if the front door's not wide enough, people will keep looking for the side entrance. Right. The challenge is to shut down that access without making ordinary compliance work even harder for accounting firms.”

“I think HMRC would find a lot more tax being paid on time if they provided that.”

There are no external guests at all - this is a roundtable of the show's own editors (managing editor and technology editor), who are journalists rather than practitioners who have run accounting firms at scale. Named practitioners (Alton Utah, Steve Bryce, Dan Healing) appear only as brief quoted references from separate off-show conversations, not as interviewees.

“Hello and welcome to the latest episode of no Accounting for Taste with me, Matthew Ort, and the familiar voices of Accounting Web's managing editor, Richard Hattersley.”

“I spoke to Alton Utah recently, fab speaker and finance and operations director at, uh, Geocoustics.”

The episode provides a usable regulatory timeline with concrete dates (March 2022 announcement, July 2025 Labour shift, January 2026 pause, 2028 implementation), named companies (Yodlee, Tink, Xero, QuickBooks), and named individuals. However there are no quantitative metrics - no firms affected, no cost estimates, no breach statistics - and the soft-skills segment is entirely anecdote-driven.

“And looking back over the timeline since the moment that this was announced, way back in March 2022”

“if HMRC sees that every day at 3 o' clock in the Morning, that firm is logging in to HMRC using the agent account, goes into every single client record, checks, balances, penalties, interest, and then copies the figures. Well, I mean, that's fairly easy to spot.”

The screen-scraping Q&A shows solid structural craft - the host sequences logical follow-ups (what is it, why do firms use it, how does HMRC detect it, what happens if caught, what should firms do) and presses on whether a roadmap is coming. However, the P&L section is largely an uninterrupted monologue with almost no follow-up or challenge, and the soft-skills closing segment is entirely unchallenged assertion.

“So sticking with hmrc, then how do they know that firms have been using screen scrapers?”

“I asked if there had been like a specific breach. I mean not, not asking for names obviously, but just to confirm whether there had been a breach involving one of these technologies. And they, they didn't get back to me on that one.”