Book an intro call
Case studies
BeeBee

How BeeBee automated its daily close

Every evening BeeBee reconciles its POS, online shop and delivery portals. With Pharen it takes 5 minutes instead of 60 and saves about €1,400 a month.

Book a no-obligation intro call
Client
BeeBee
Industry
Hospitality
Location
Ilmenau, Thuringia
Services
Process design, Workflow automation, Data reconciliation

About 35 hours

less reconciliation a month, almost 40 before

About €1,400

a month, from saved working time and resolved portal settlements

0 to 3

exceptions for the team to check on most days, instead of about 120 orders

In short

The restaurant BeeBee in Ilmenau receives about 3,700 orders a month, at the POS, in its online shop and through delivery portals. At daily close, a workflow in Pharen Hub brings together the day's data from all 3 sources. It maps the values to the same fields, adds up orders, revenue and cancellations, and checks portal settlements and cancellation reasons against fixed rules.

Before, the reconciliation took about 60 minutes every evening, almost 40 hours a month with the month-end close. Today it takes about 5 minutes in the evening, and on most days 0 to 3 exceptions remain. That saves about 35 hours a month, worth about €950 at €27.40, the labour cost per hour in German hospitality. On top of that come about €430 a month from portal settlements whose gaps now show up the same evening. Together, that is about €1,400 a month. Resolving open items stays with BeeBee.

With Pharen, BeeBee has:

  • brought about 3,700 orders a month from the POS, online shop and delivery portals into one daily overview
  • cut the check at daily close from about 60 to about 5 minutes, and the month-end close from about 8 hours to 1
  • saved about 35 hours of working time a month, worth about €950
  • made gaps in portal settlements visible the same evening, about €430 a month
  • resolved missing cancellation reasons the same evening, while the shift still remembers
  • limited open items to the exceptions, 0 to 3 on most days

Starting point

At 8 pm, delivery ends at BeeBee in Ilmenau, 7 days a week. After that, there used to be about 60 more minutes that no guest sees: the daily close.

About 3,700 orders come in each month, about 120 on a normal day, at the POS, in the online shop and through delivery portals. Every order was recorded, but in 3 systems that knew nothing about each other. Only when someone put orders, revenue, cancellations and payouts side by side was the day complete.

3 systems, 3 formats

The POS provides the sales on site and the day’s cancellations. The online shop lists the direct orders with the payment method. Each delivery portal keeps its own order lists and pays out later, minus commission and often for several days at once. The values did not arrive in the same format or under the same labels.

Before: about 60 minutes after closing

Someone from the team opened all 3 views, copied the daily totals into a spreadsheet and checked whether order count, revenue and payout matched. Cancellations needed a reason. If one was missing, the search went back to the shift, often only the next day, when nobody remembered exactly.

If a portal settlement was lower than its orders, someone had to work out whether a fee, a delayed payment or an error was behind it. After a full day, there was often no time for that. At the end of the month, someone laid the portals’ payouts against the orders once more, about 8 hours of work, and small differences still slipped through.

About 60 minutes on about 30 evenings make 30 hours a month. With the month-end close, it came to almost 40, as much as 1 full working week. The Federal Statistical Office puts labour costs in German hospitality for 2025 at €27.40 per hour worked. So the reconciliation cost BeeBee just over €1,000 a month, for copying and recalculating.

Implementation

What we built in Pharen Hub

We built the daily close as one workflow. It takes the day’s data from the POS, online shop and delivery portals, maps it to the same fields and writes orders, revenue and cancellations per source into one shared overview.

Then the workflow applies fixed rules. Does a portal’s settlement match its orders? Fees that the portal lists do not count as a discrepancy. Does every cancellation have a reason? Anything left unresolved appears as an open item with its source and amount. A day without a discrepancy is closed without the team recalculating every total by hand.

BeeBee's daily close as a workflow in Pharen Hub. After closing time it pulls the POS, online shop and delivery portals, maps their fields to one shared structure, separates the portal fees, adds the values per source and stores the totals. Checks record mismatched settlements and cancellations without a reason as open items, otherwise the day is closed.
The workflow, rebuilt with sample data in Pharen Hub. An open item appears in the daily overview only when a check fails. Open the full-size view.

How the daily close runs today

  1. After closing, the daily totals from the POS, online shop and delivery portals go into the workflow.
  2. The workflow maps the different fields to one shared structure.
  3. Orders, revenue and cancellations are added up per source and stored as the daily total.
  4. Portal settlements and cancellation reasons pass through the agreed rules.
  5. With no discrepancy, the day gets the status “Complete”. With a discrepancy, only the affected item stays open.
  6. The team checks the receipt or the shift note, adds the reason and closes the item. For an uneventful day, a look at the overview is enough, and with 2 open items it takes about 5 minutes.

The workflow does not change any payout and does not post anything to the accounts. It prepares the reconciliation and shows where a decision is needed. Resolving it stays with BeeBee.

Result

One day, 2 open items

The daily overview below shows 7 sample days with anonymised figures. On 26 September there were 75 POS orders worth €2,520.00, 30 online shop orders worth €945.00 and 19 portal orders worth €655.00. Together, that is 124 orders and €4,120.00.

There were 3 cancellations. The team was left with 2 items: a portal settlement €43.10 below the order total, and 2 POS cancellations without a reason. For the €43.10, the team checks the portal’s settlement, and for the cancellations it asks the shift that was on the till that evening. Everything else had already been reconciled.

BeeBee daily overview in Pharen Hub with 7 sample days from 20 to 26 September, each with orders, revenue, cancellations, open items, status and a note
The daily overview with sample data. On 26 September, 124 orders, €4,120 in revenue, 3 cancellations and 2 open items are ready for review, and the days before are closed or resolved. Open the full-size view.

About 35 hours less reconciliation a month

Today the evening check takes about 5 minutes instead of 60. The month-end close needs about 1 hour instead of 8, because the days are already checked. Almost 40 hours a month have become about 3.5, so about 35 hours less. At €27.40 per hour, the labour cost in hospitality, that is about €950 a month.

Gaps in portal settlements show up the same evening

A gap like the €43.10 on the sample day now shows up the same evening, while it can still be documented and claimed back from the portal. With about 10 such gaps a month, that is about €430 that used to get lost in the monthly settlement.

Together with the saved working time, that is about €1,400 a month.

What changed

The daily close no longer starts with 3 views and an empty spreadsheet. The team opens the daily overview and sees right away whether anything needs resolving. It resolves missing cancellation reasons the same evening, while the shift still remembers. For the month-end close, the days are already checked.

Of the 60 minutes after 8 pm, 5 are left.

Published September 2, 2026, updated October 1, 2026

Get the PDF

What are your plans for AI? Let's talk about it.

Whether it is strategy, training or delivery, you talk to our founder Johannes for 30 minutes. There is nothing to prepare.

We have already worked with more than 50 clients.

How the call works

1

Your topic

You tell us what it is about and where you stand with AI.
2

Our assessment

We tell you where AI helps you and where it does not.
3

Your next step

You get a proposal and decide for yourself.