Insights |

What should I do if my website goes down?

Written by

Tim Baker

Founder / Website and Mobile App Developer

Making mistakes online can be detrimental. The result can be seriously damaging to online businesses. Not only can money be lost, but also long-term user trust in the reliability and security of your website/service can be shaken too.

In this article I explore what you can do to get ahead of any potential problems, as well as providing steps you should consider when things go wrong.

Get ahead of the problem

Make sure you know about the problem before your customers do. To do this, you can use one of the online services like Pingdom or Status Cake. These companies will monitor your website and online services 24/7. These types of monitoring tools will alert you and send you a notification if a problem occurs. They also provide detailed reports containing information about when the problem occurred and information about the problem. This type of information is invaluable to a developer who will then use this to fix the problem. The fix can be so quick that your customers/users don't even notice there has been a problem.

Prepare for the worst

Larger organisations normally have a network infrastructure which handles system blips. A lot of companies utilise Docker and Kubernetes to automate and solve these blips by whirling up servers as required. This could be to handle web traffic generated by a large marketing campaign, or could be to replace a problematic server/container. This same automation can be used to power down any unnecessary servers to save money on hosting costs.

For a lot of small/medium-sized businesses this level of complexity isn't required. There is also no bullet-proof solution which will cover all scenarios, as each business is different. You may have an e-commerce website, or perhaps an administration/online portal, which your staff rely on to conduct daily business. It is important that any system nuances are covered in a redundancy plan, keeping your website(s) and critical systems operational.

I also recommend using Cloudflare with your website. Even the free tier of their service will offer your website some reliability and performance benefits. For example, their free service crawls your website once a week and caches/stores a copy of it on their servers. This means that if your website goes down users can still see the static portions of your site, as the last cached version stored on Cloudflare will be automatically served up to them. There are many more benefits to Cloudflare which are outside the scope of this article. When you get a moment take a look at the features listed on their website.

Speak to your tech team

Speak to the technical/development team who manage your website. Ask them if they know the reason why the website or a particular piece of functionality isn't currently working. Ask if it affects just a particular section or page. If possible also try to get an estimated timescale from your development team on when a fix will be in place.

If the team isn't aware of the problem then try to provide as much information about the problem as possible. For example - steps to reproduce the problem and supporting screenshots.

Pause your marketing campaigns

It's likely that any down time to your website or service is already impacting business and is costing you money. If the problem is going to take time to fix, then I would recommend contacting your marketing team and pausing any paid campaigns you have running.

Keep your users in the loop

If the issue impacts user experience then I would recommend notifying your customers. Most SAAS (Software as a service) products have dedicated status web pages which customers can visit for updates in the event of a problem. These are great for reducing telephone call volumes into your customers service team. In addition, you may want to consider making an announcement about the problem through your marketing channels (social, email etc). Normally, I would suggest this is only done if you have a large volume of traffic experiencing problems, or you expect the functionality and user experience to be affected for a long period of time.

Have a retrospective

Problems do happen from time to time. All tech companies, regardless of size, have the odd hiccup now and again. However - when they occur, it's important to learn from them and sometimes modify or introduce systems and processes to try and prevent them from happening again. Work closely with your tech team, find out what the root cause was - and what steps can be put in place to prevent it from happening again.

Stay open-minded about external help

A fresh pair of eyes can really help a development team. Sometimes just talking the problem through with another developer or having a couple of new ideas on the table can really help lead to a solution quickly.

Related links

Pingdom
https://www.pingdom.com
Status Cake
https://www.statuscake.com
Cloudflare
https://www.cloudflare.com

Become our friends