With clients, I emphasise that they should test any new campaigns and processes.  People ask “Why?”  “Isn’t it a waste of their time?”  Shouldn’t I do it, instead of them?  And why haven’t I tested my own work?  What kind of shoddy job was that?

Good questions.  Let’s explore them.  Ask yourself

  1. Is it OK for your system to flawlessly create the wrong result?
  2. Is it OK if the process does exactly what you thought it should at the time, but when you use it, you wish it did something else?
  3. Is it OK if the process has the perfect outcome, but your staff can’t use it?

Finally:

How do you know this new process doesn’t have one of those problems?

 

I’ve had a few discussions about the how & why of testing.  So let’s explore …

In short, testing is insurance that you pay for with time, rather than money.  It takes time to test – but it can take much more time to clean-up after not-testing.  You’re automating your business, largely to ensure that every customer receives your “best customer experience”.  Testing helps ensure this happens.

Let me explain …

Testing checks three things:

  1. The system works as designed.
  2. The design works for your business.
  3. You can use the system.

Often people stop at the first point.  Bad move.  It’s great to make sure that something does what we said we wanted.  But that’s only the beginning.  A system is a combination of people, process, and software. Having software that works is a great start – but only a start.

The goal of testing is to check that when you use the process you designed, the outcome works. That means you need to understand how to use the system, and that the result of the system is actually what your business needs.  Remember those questions:

  • Is it OK for your system to flawlessly create the wrong result?
  • Is it OK if the process does exactly what you thought it should at the time, but when you use it, you wish it did something else?
  • Is it OK if the process has the perfect outcome, but your staff can’t use it?  

The answer to each of these is “no”.  While testing is never perfect, we try to minimise these outcomes.

So how do you test?

At its simplest, testing is just walking through the process manually.   Let’s take a simple example: testing a signup form for a newsletter.

The actions here are:

  • First, the person signs up on my form.
  • Then they see an acknowledgement or thank you page.
  • Next, they receive and opt-in confirmation email.

At this point there’s a choice: Either they click the link or they don’t.  If they click the link I will see them having opted-in within the contact record in my system. If they don’t click the link, they will still be “unconfirmed”.

In this case there are a few things we check:

  • Did the process flow happen as we expect?
  • Did each email and web form look correct? (Was the branding correct, for example?)
  • Was the person’s status “unconfirmed” until they clicked the link?
  • After they click the link, did their status change?

This is a simple example.  We can complete this with a single test contact.

I create a Crash Test Dummy contact in my system.  I use this dummy contact and make them step through every part of the process. Each step of the new process will do things.  Maybe it triggers an update on their contact record.  Perhaps it triggers communications or integrates with another system (such as a payment gateway or finance package).  Using this dummy person we can check every piece works as planned.

Sometimes there are more branches – more choices that people can make. In these cases you may need multiple contacts to complete your tests.  When testing complex processes, it is best to plan out these decision points in advance. Against each of these parts, write down which Crash Test Dummy email address, you will use to test this scenario. In effect you’re making fake people and assigning a series of actions to that person.

Tip: If your process needs more than one Crash Test Dummy, you need several email accounts.  These may be personal accounts, business accounts, or possibly accounts to set up with free email providers.

 

Phew!  Testing sounds like a lot of work when it’s written out. It does take time.  You may have better things to do with your time (don’t we all?)

On the other hand, it takes a lot more time to clean up situations with your clients when something goes awry.  Imagine if any of these scenarios happened to you:

  • People sign-up to buy your product and you never reply (because you were looking in the wrong place for sign-ups).
  • You need to contact people to explain why they paid money and didn’t receive or can’t access your product.
  • People try to buy your product, but can’t because your process missed a stage – like the payment entry page is missing.

Testing is completely optional, of course…

 

If you want to talk about how to plan, build or test your campaigns, I’ll be happy to help – just let me know.