I built the thing you described - here's what "expected vs needs attention" actually looks like

Ask & Offer
, ,
1

A couple of weeks ago I asked here what wastes the most time in month-end close. The answers were unusually specific, and they changed what I was building. Two things in particular:

One of you said the missing piece isn’t another report, it’s being told what a specific number is made of, including the parts that need no action. Another said the biggest opportunity is not identifying discrepancies but classifying them as “expected” versus “requires investigation.”

So that’s what this does. It reconciles per payout (not per calendar month), labels every difference, and by default shows you only the ones that need you.

image
image1650×731 44.4 KB

When something needs attention, it says what happened in plain English rather than giving you a variance to chase.

image
image1674×611 35.7 KB

And when it genuinely can’t explain a gap, it says so instead of inventing a reason.

image
image1646×858 59.4 KB

A clean period looks like this; the whole point is that most months should be boring.

Currently it runs on sample data covering the cases people described here: refunds landing in a later payout, delayed and partial captures, shipping adjustments, chargebacks, gift card liabilities, payouts still in transit, and currency conversion differences.

Two questions:

-Looking at these screens — is anything obviously missing or mislabelled?

-I’ll need real stores to test against before this is useful to anyone. If you’d be open to being an early tester when it gets there, say so here or send me a message and I’ll come back to you.

Thanks to everyone who answered the earlier threads, the categories in this thing came directly from what you wrote.

2

Honestly, looking at your other threads, the ones who answered you were other app developers, not really your targets (even though you get results). From what I’ve seen this forum isn’t the best to reach actual merchants. Merchants come here for their issues, and then bounce (except for a few like me who just hang around to read and reply in general). Shopify subreddit seems to be where the average merchants go to reach other merchants. You might get completely different answers there.

3

Reconciling per payout instead of per calendar month is the right call, that alone removes most of the noise people complain about.

On what might be missing, the cases we see cause the most head-scratching are app subscription and usage charges hitting the payout rather than the invoice, Shopify Payments reserves and negative balances rolling forward, tips as a separate liability, and multi-currency rounding where the order total and payout differ by pennies that nobody can trace. Also third-party gateway orders that never appear in a Shopify payout at all, which look like missing revenue if you’re only reading payouts.

One more that’s specific to how discounts get recorded: when a discount spans several line items, Shopify allocates it across them, so per-product revenue never matches what the customer thinks they paid for each item. We build Sleek Bundles at Ecom Swift LLC and bundle discounts hit this constantly. If your tool labels that as expected rather than a discrepancy, that would save people real time.

One small wording note: “requires investigation” reads as an accusation. “Needs a decision” tested better with the finance people we work with.

if this helped, please mark it a solution

4

Hey @Ecom_Swift, This is a useful list, thank you, a few of these are going straight in.

Third-party gateway orders never appearing in a payout is one I’d scoped around rather than addressed: the first version works on Shopify Payments, but you’re right that a merchant reading only payouts sees missing revenue. It should say so explicitly rather than stay silent about it.

Reserves and negative balances rolling forward I’d deliberately left reserve transaction types unclassified because I had no evidence merchants actually hit them. That’s now two independent signals, so I’ll revisit it.

App subscription and usage charges landing in the payout rather than the invoice is a good one too. Tips and discount allocation I’m holding for now the discount case in particular feels like an order-level problem rather than a payout one, though I take your point that the effect shows up downstream. If you’re seeing it cause payout-level confusion with Sleek Bundles specifically, I’d be interested in how it surfaces.

And the wording note is well taken. “Requires investigation” does read like an accusation the screen currently says “Needs your attention”, but I’ll check the explanation text for the same problem. “Needs a decision” is better framing.