Fuel costs disrupt a cruise, business operators share customer support plans
(Featured)
- AOL reported that fuel costs disrupted a long-term cruise and left Villa Vie Odyssey passengers stranded in Cambodia. Business operators share contingency-planning practices for supporting customers during sudden service suspensions.
Edited by Adrian James, Editor at Featured News · Roundup by Featured · Nov 3, 2026
AOL reported that Villa Vie Odyssey passengers were stranded in Cambodia after fuel costs disrupted the long-term cruise.
Business operators share contingency-planning practices they use to support customers when services are suddenly suspended.
RoundupsTaylor HornCo-founders, Rents4Baby · Sacramento, California
At Rents4Baby, we do whatever we can to fulfill an order, starting with another rental partner or an item substitution. If neither works, we often rent through a direct competitor. We would rather get the family what they need and keep their trust, especially when it involves sleep or safety equipment like a crib or car seat.
Sometimes we send a Target gift card or order directly from a local store for same-day delivery to the family’s destination. We have built a strong network of rental partners whose values align with ours and who don’t want to leave a family stranded.
In one instance, an expensive bassinet broke 24 hours before an order, and we couldn’t get another in time from a rental partner or the manufacturer. We offered an alternative bassinet, but the customer declined.
Todd HarmonFounder, AURA Modern Home · Denver, Colorado
Our contingency practice starts before anything goes wrong: we set customer expectations up front with delivery date ranges that have a little padding built in. When things shift a few days, we've already accounted for it from day one, so we don't say anything.
When a manufacturer delays an item, we wait a few days to see if the delay changes. Once we know it's firm, we update the customer with their new tracking or shipping information. Manufacturers usually offer full refunds when something is backordered much longer than expected, but we still try to make it right for the customer.
Recently a customer's bed was supposed to come back in stock. The manufacturer didn't get the shipment, and it was going to be two months late, so we offered the customer a refund. That's hard to do, because we pay about 3% on the credit card charge and another 3% on the refund. But it's the right thing to do. We couldn't meet the date we gave them, so that's our fault.
Mike BruceBrand Growth Architect and Consultant, Founder, Mike Bruce Design · Abuja, Nigeria
The practice I rely on is early, transparent communication, backed by compensation. When we realise a couple of days before a deadline that we can't meet it, we email the client straight away. We explain the reasons, give them the earliest new date we can commit to, and offer something free, on the house. Clients want transparency. When you aren't hiding anything, people are very understanding. That has been my experience.
I have also had to deal with a physical disruption. Our facility once lost power for two to three weeks. Our inverter system usually lasts about two days, and once that was used up, the power went out. We moved into another facility temporarily, booked for a couple of weeks, so that we could finish and submit our clients' projects.
Kiragu MainaDeveloper & Network Engineer, FyberPay · Nairobi, Kenya
Our hosting provider posted on their status page that they would reboot some or all of their servers in the coming days, but they didn't say exactly when. We couldn't plan around it, so we held on. The reboot came at a critical time, when most of our ISPs were online, and took our service down for about 20 minutes. The first sign was a call from one of our ISPs asking if everything was okay and why their customers couldn't get online. We immediately updated all our customers through our notification channels and explained it was a security reboot. With AI, flaws like these will keep turning up, so these reboots will keep happening.
So we fixed the root problem. Internet access is now a basic service, like water, and you cannot rely on one server. We now run three servers in different locations: two run the actual services and the third acts as a witness. Two people with equal rights can't settle a vote, but three can. If one fails, another picks it up.
If you provide a critical service, run more than two nodes, and have fallbacks.
Darryl StevensFounder & CEO, DIGITECH · Austin, Texas
Do not wait until after a disruption to improve your communications. Instead, test your communications plan twice yearly in tabletop exercises.
Be proactive by providing efficient rescheduling options to retain your customers and simultaneously provide your clients with a smooth experience. After all, your customers want to know someone is working on their problem. If the organization's decision maker is unavailable or incapacitated, is there clear guidance that indicates who is authorized to perform those duties, roles, and responsibilities? For complex issues that can't be resolved immediately, provide the customer with a clear action plan and a realistic timeline for the next update.
Create a status page to act as a central resource for updates. In addition, make sure that your lines of communication are well and truly open (and that your customers know that they're open). Once the systems are in place, the company should inform various stakeholders, in normal times, where to go for information during outages.
Carmelo ZaffaranaFounder & CEO, Lucernel LLC · Syracuse, Sicily
When it comes to customer support, when a service goes down, there are actually some very simple things you can do:
1. Add a status page, where it pings the service and shows the uptime. By doing this, the user can check the status themselves (Fun fact: did you know that 99.90% uptime means the service is offline for almost 9 hours a year? It seems like a little, but it's actually a lot).
2. Set up automatic routing for incoming emails: recent AI models based on the System One approach are becoming popular. They're decision-making models that don't generate any text, but calculate probabilities, yes/no decisions, scores and categories.
For an intelligent system, like I've done in my services at Lucernel, you can make it so that every email, when it arrives, passes through the System One model (Jev is the famous one at the moment), and by putting all the departments in the request, you can make it go to the most appropriate department, like Support.
Then, within Support, you integrate another request, where you analyze the type of inquiry. If, for example, it's a "the page won't load" issue, you can integrate the system that checks the status of the services and send an automatic email saying that there are service disruptions and to check the status page.
It sounds difficult, but it's actually a very easy thing to do!
3. (Not recommended anymore) Do it the old-fashioned way, where, in the automatic email, you add that there are currently service disruptions and to check the status page.
The typical automatic email: "Your request has been received and you will get a response as soon as possible", and below that, also the current status of the service.
This story was produced by Featured and republished with permission. Experts featured here are members of Connectively, Featured's expert network.
- As summer winds down, travelers looking to squeeze in one more getaway may find that post-Labor Day trips come with plenty of perks. Fewer crowds, lower rates and crisp late-summer weather can make September an ideal time to travel. But even as the busy season slows down, safety and peace of mind should remain a priority.
