What to send an account manager when you claim lost trading hours

Restaurant chains write these messages as grievances and get generic replies, because the recipient is not an adjudicator. A delivery platform account manager is closer to a holder of switches, and the published documentation names several of them explicitly. A claim built around what that person can change gets acted on. A claim built around what was unfair gets acknowledged. The package below is short on purpose: the platform’s own identifiers, one table of intervals, one documented mechanism, and a single ask that corresponds to something the account manager is documented as able to do.

What can an account manager actually change?

More than most operators realise, and the platforms say so in their own words rather than ours.

On Talabat and HungerStation, the Check-in feature that decides whether a branch opens on schedule without a human is documented as switched off unless somebody asks, and both write “By default, this feature is disabled and does not require a check-in status to be sent. To activate the check-in flow, please contact your account manager.” Sources: developer.talabat.com and developer.hungerstation.com. HungerStation adds portal access itself to the list, stating “you will be given access by your account manager or during registration with us”, and routes identifiers the same way: “Please reach out to your Account Manager to get the Vendor ID.” Source: developer.hungerstation.com.

On Jahez, according to Foodics, menu synchronisation runs on a fixed nightly schedule and anything else is a request: “Jahez has an enabled auto sync everyday at 3 AM, If a manual sync is required please contact your Jahez Account Manager.” Source: help.foodics.com. On Careem, part of the promotional toolset is not self service at all: “You can add offers/discounts/cpc activation by emailing partnerssupport.uae@careem.com or by emailing your Account Manager.” Source: careem.com. Snoonu gates a delivery service the same way, telling operators to “reach out to your Snoonu account manager”. Deliveroo routes its automatic opening feature through a request in Partner Hub rather than through a toggle.

Write the ask against that list, and the message stops being a complaint.

What identifiers does the message need?

The platform’s, not yours. Every one of these systems indexes a location under its own value, and a message that identifies a branch by its trading name and street forces the recipient to do a lookup before they can do anything else.

Talabat and HungerStation both address a location as a vendor inside a chain, with a chain identifier and a vendor identifier. Careem works in outlets. Deliveroo works in sites. Jahez uses a restaurant identifier alongside an integration branch identifier. Supply both the platform’s value and your own internal branch code in the same row, so that the reply comes back in a form your operations team can act on without a second translation.

If you do not know the platform side identifier for every listing you own, that is the first thing to ask for, and it is a request the documentation above says the account manager fulfils.

What does the incident table need to contain?

One row per interruption and no prose. Branch, in both vocabularies. Platform. The trading hours that listing published for that day. The first minute it stopped being orderable from a customer address it serves. The first minute it was orderable again. The duration, capped at the end of the published trading window. And the state or label the storefront showed, if it showed one.

Two more columns turn a list into an argument. The hour of day and the day of week, because a branch failing at the same hour on the same weekday describes a mechanism rather than an accident. Keep the table to the incidents that support the ask. Twelve rows that all point the same way carry further than sixty that include everything.

Give the total a benchmark, because a number with nothing beside it invites a shrug. A typical monitored listing loses about a full trading day over a month, and roughly a third of listings lose nothing at all, so a branch several times above the market rate for its own market is describing an outlier rather than a bad week. Our published market figures are the UAE, Saudi Arabia and Kuwait reports, and quoting the market figure alongside your own is what turns a claim about one branch into a question about a configuration.

What should the ask be?

One thing, expressed as a change rather than as a judgement. Enable or disable Check-in on these vendors. Restore the visibility flag on this branch. Correct the trading hours held on the platform side for these three locations. Confirm which of our outlets are catalogue locked because they are point of sale integrated. Tell us what triggered the closure recorded at this timestamp.

Set the expectation on money separately and honestly. We read the published partner materials of the eight platforms we monitor most closely and none of them publishes a compensation rule for downtime the restaurant did not cause. There is no clause to invoke, which is why the ask should be a configuration change or an explanation. The longer argument about evidence and payouts is set out in claiming for delivery platform downtime.

What should you leave out?

The revenue estimate as a headline, because it invites a debate about your arithmetic instead of about their switch. The adjectives. Any second topic, since a message with two asks gets the easier one answered. And the incidents you caused.

That last one is counterintuitive and it is the most effective item on the list. A package that quietly drops the evenings when your own branch paused itself, or when nobody acknowledged a scheduled opening, is a package with an easy rebuttal inside it. A package that names those separately and excludes them from the total is one the other side has to engage with on the merits.

When should the message land?

Earlier in the month than feels natural, because at least one platform runs a calendar deadline that changes what is at stake. Deliveroo assesses marketing eligibility “three days before the first day of each month”, and a business that misses a threshold loses access “for a full month”. Source: help.deliveroo.com. A claim that hours lost to a platform side closure distorted a rejection rate is worth far more before that assessment than after it.

The same logic applies to features rather than metrics. Deliveroo’s automatic opening requires “At least 95% availability to request Auto-open”, so a chain trying to qualify has a reason to get platform caused hours acknowledged rather than absorbed. Source: help.deliveroo.com.

What does a package that works look like?

One page. A three sentence summary naming the branch, the platform and the total hours. The incident table. One quoted mechanism from the platform’s own documentation where the state was one you could not lift. One ask. And a note that the same measurement runs continuously across your whole estate, which is what makes the twelve rows a sample rather than a selection.

That last line is what changes the tone of the reply, because it says the record was not assembled for this conversation. Kitchain (kitchain.co) produces it as a by product of checking every listing on the same cycle, so a claim is an export rather than a project.

Related

Start Monitoring



    No credit card. No integrations.
    We'll configure your first location and confirm within 24h.
    Request a Demo

    Book a personalized walkthrough of Kitchain Products.



      We'll get back to you within 24 hours.