Hacker News Reader

Comment by alexpotato | Hacker News Reader

One of my favorite stories about logistics and quarterly earnings deadlines (from when I worked at a pharmaceutical company:

"In our business, a truckload of various drugs can easily reach $10-$15 million. Now, if that truck arrives at the depot at 11:59pm March 31st then it's first quarter earnings. If it arrives at 12:01am April 1st then it's second quarter earnings.

$15 million is a BIG shortfall, even for us, so you better believe those truck drivers will roll the stop signs, blow red lights etc to make sure that truck arrives before 11:59pm"

View on HN

That doesn't make any sense because the revenue is already booked for the sale which has nothing to do with when the delivery truck actually arrives.

If it is cash based accounting, revenue and expenses are booked when the money changes hands.

If it is accrual-based acocunting, it takes place when the event legal triggering the change of ownership of goods in the transaction takes place, which depends on the shipping terms, which could be anywhere from when it is available for the buyer’s transport agent to pick up at the seller’s facility (EXW) to when it is delivered, unloaded, and at the buyers door (DDP) or any of a variety of places in between (FOB Origin, FOB Destination, and a bunch of other potential shipping terms with their own rules on when ownership—and responsibility—transfer from seller to buyer.)

Yes, to add on -

- incoterms (https://www.dhl.com/content/dam/dhl/global/dhl-global-forwar...)

- cash flow statement v. income statement (accurals)

Those diagrams of risk vs cost bared are helpful.

It's got to be one of these:

FOB Shipping Point (or Origin): Responsibility transfers to the buyer as soon as the goods leave the seller's premises. You book it when it leaves your loading dock.

FOB Destination: The seller retains risk and costs until the goods reach the buyer’s location.

The sale doesn't happen until the asset transfer occurs. Before that any cash you get from the sale is balanced by the liability to actually produce the good or refund the money. Or more likely you don't get any cash but can't record the bill as accounts receivable. It's not receivable until the transfer point is crossed.

You can account a transaction that's been placed but not fulfilled. I think when someone orders $15m of goods, you can immediately book $15m accounts receivable (asset) and $15m goods owed (liability) as soon as you have the expectation it will happen. If the transaction falls through, you delete them.

Under GAAP you cannot recognize revenue before the service is delivered or product is shipped. You can accrue revenue that is earned but not yet paid (if you are paid on Net 30, for example), but even if pre-paid you have to book that as deferred revenue, which is a liability (until you ship).

There's no deleting anything in accounting.

As someone with a partner who’s an accountant, I love seeing technologists be confidently wrong about accounting fundamentals vs. the type of technicalities that she has to deal with. Your comment highlights the absurdity of their confidence; kudos.

This is not correct. A business this big would definitely be using accrual accounting (not cash) which generally means you count the revenue when the actual ownerships transfers to the buyer. Since the truck was operated by the seller, the transfer of ownership is almost certainly counted as when the buyer receives the goods.

Rewrite the anecdote with the truck racing to the supplier to make the pickup on time.

Accounts receivable, revenue, and cash are related, but separate, accounting items.

Ya but op's anecdote is cute and funny.

That’s not why it doesn’t make sense. It doesn’t make sense because in the forward guidance you’d be able to say you expect the $15 million coming in.

Cash vs Accrual

How many F500 companies use cash accounting? How many public companies altogether?

I worked at a previous listed company where a single $6MM order of hardware being pushed out a week made quarterly p&l positive. Im absolutely sure the same situation occurred every other quarter as well in some part of the business I didnt see.

Sounds more realistic than a low level truck driver running stop signs.

this kind of deal timeline management happens at all companies. this is why contracts get structured in complicated pricing structures to make it easier for revenue recognition to occur in the quarter it’s supposed to. the timeline can move from 3 months to 6 it’s still going to be a huge focus area for a lot of people at every company

This is why Netflix broke up the final season of Stranger Things in such a weird way... they wanted new episodes at the end of quarters, to have good subscriber numbers for the quarter report

Over the long run, companies that spend resources on this micromanagement of earnings probably are not seeing the forest for the trees. I cringe thinking about how much time and top talent at a company is spent preparing for earnings rather than spending those hours improving the business itself.

Likewise, if you know you've already got the current quarter in the bag, but the next quarter is looking soft, you tell that truck driver to slow down!

If you can't be arsed to ship it before the last day of the report, it's your fault and asking the truck driver to drive like a maniac to compensate is amateurish, dangerous.

That's not something I'd be proud of.

It's an extreme example that shortens the time frame and shrinks the cast of characters to illustrate the point.

Substitute "telling the truck driver to run stop signs" with "order the factory to increase production", it's the same thing.

Or, sometimes, you order the factory to _reduce_ output to 50% of what it can do for the last week of Q1 so you don't have excess unsold inventory on the books.

Then in Q2, you panic because you don't have enough inventory, so you order the factory to produce at 150% to catch up. Both 50% and 150% are inefficient factory states; if you weren't thinking about snapshot reporting you'd have just let it run at 100% and your Q1+Q2 results would be overall better.

I have personally seen this happen at a household-name Fortune 50 company. It's insane and causes real damage to the business in many ways.

in two word: being short-sighted.

Early in my career I worked at a place where the sales people would half-joke about signing deals on December 40th -- to claim it in the previous quarter/year.

December 40th is last day of Q5.

Okay, what's that to say it won't be the same but even worse on a 6mo reporting schedule?

as the time period gets longer, the the more likely it is that the numbers represent the true performance of the business rather than randomness. That has to be balanced against the fact that investors get less frequent updates i.e. the information is now potentially 6 months out of date rather than 3 months at worst. But then its just a judgment call of the relative benefit of each - you could argue that with modern accounting systems, modern companies could deliver weekly or even daily earnings , which would give investors much more timely information, and the high frequency would probably mean it wouldn’t be worth making the effort for management to fudge the numbers to bring forward or delay revenue one day or one week. There would be a lot more variance in the numbers if they were daily, but thats a good thing - it would just reflect the underlying randomness, and then the investors could decide when the accumulated trend over a period of time is meaningful or not, instead of management wasting time massaging numbers into a fairy tale of steady growth.

In every sales-led company quarter end is a shitshow. It'll be even worse if there's only one chance to bring the numbers back in instead of 2 or 3. It's used to put pressure on sales teams, but the net result over the year is never good because it sours relationships and reduces overall deal value.

The best thing would be continuous daily or weekly reporting with no defined year end. Unfortunately the entire global system of tax and accounting is set up around annual reporting, so change is impossible.

How is 2 data points a year "representing the true performance of the business" but 4 data points a year is randomness?

You also get less frequent CPU usage % datapoints when you want to be sure about usage? That makes no sense at all.

"We risk everyones lives in order to have a barely just-in-time warehouse on shipments in the low 8 figures."

Cool.

What company is this?

It's very clearly an analogy.

To be fair, this is a side effect of financial deadlines in general. A business reporting annually will still try to beat that deadline right up to the last second.

In this particular example a single truckload would be less significant annually than quarterly though.

General Electric has a history of using that exact trick... just with jet engines and power generators and medical devices that can represent much larger amounts of revenue.

Not just - Welch pioneered using GE Capital to “smooth” earnings - lots of judgment calls in those finance companies in the early 90s.

GE's latest trick is to roll long term maintenance contracts into the price of the product and then sell off the unit holding the bag on the maintenance contract. Very shady but very clever.

I don’t get it, why not just say “oh we were 15 million short that quarter and 15 million ahead this one so it’s all good”

Like why get hung up on these arbitrary cutoffs

This is what confuses me about tech sales too. Why do I always get a discount for buying right at the end of the quarter? Seems like you could just get ahead on the next one.

Because of how performance targets are defined. If sales figures “fell” or didn’t grow at the expected rate it can be worse than having a few more points over next quarter.

Why do they define performance targets that way then? Are they making it hard on purpose for some reason?

Sure but in most instances the system won’t change. Someone will usually define some reporting time period and some metrics to look at and agree when a lower threshold is reached that things start to look bad or alternatively, really good and they get an extra bonus or something.

What would you suggest? It sounds like you have a better system in mind already?

You have to set a deadline at some point. Now, I think any rational manager would agree if the sale shows up on April 1 instead of March 31st, that's totally fine. But HR/Finance systems aren't always rational.

Because many people are lazy and need deadlines or else they won't work.

It, unfortunately, always comes down to people being dumb. I wish there was an another answer, but that's always the essence of the issue.

I've seen someone trying to explain to an investor that the numeric change between quarters wasn't meaningful. Investor seemed to understand explanation, but was clearly deeply unhappy about the chart not being tidy. People respond to that by trying to make the chart tidy, even if it means losing the company money at times.

Reply on news.ycombinator.com