I have a bit of a love-hate relationship with video mystery shopping.

On one hand, it is probably the most detailed mystery shopping methodology available. You can see exactly what happened, hear the language used, watch how the interaction developed and pick up behaviours that would be very difficult for a shopper to remember accurately afterwards.

That makes it incredibly powerful. It also makes it incredibly easy to overdo.

A large video programme can generate hundreds/thousands of hours of footage. Hundreds of hours that someone then has to watch, analyse, validate, interpret and somehow turn into something useful. The fieldwork might be excellent, and the evidence might be incredibly rich, but if the client team (and the actual sites) cannot realistically consume it, you have created an insight problem rather than solved one.

And this is where I think video mystery shopping is sometimes misused.

Because it is detailed, it often gets sold as though more must automatically be better. More locations. More interactions. More footage. More analysis. Bigger programme.

But more evidence does not automatically mean more insight.

If the business problem is around sales conversion, service recovery, needs discovery or confidence in a particular part of the customer journey, the first question should not be “how many video visits can we do?”

It should be “what are we trying to understand?”

That sounds obvious, but it is the bit that gets missed (or even ignored in my experience), far too often.

Mystery shopping is market research. It should be designed like market research.

Start with the objective. Decide what behaviour or part of the journey might be influencing it. Use the right methodology & sample to investigate it. Then let what you learn shape what happens next.

Ten well-designed video visits might uncover the same behavioural pattern repeatedly. That does not mean you can confidently say the same issue exists in every location, but it gives you something extremely valuable: a hypothesis.

At that point, the smart thing might be to stop.

You may then use a much lighter mystery shopping programme across the wider estate to understand how widespread the behaviour is. You might change the questionnaire. You might target a particular region. You might work with L&D on a coaching intervention. Then you re-measure.

That, to me, is what a good programme should do. Evolve.

The alternative is running another 100 video visits just because that was the original plan and ending up with another 100 hours proving largely the same thing.

That is not better research. It is just more expensive research.

There is another part to this too, and it is something we are thinking about a lot at Silent Customer.

Even if you collect brilliant evidence, it still has to reach the people who can use it. A senior leader does not need hours of footage. They need the pattern, why it matters and what decision needs to be made. A manager needs to understand the behaviour they should be coaching. A frontline team needs fair, practical feedback they can actually use.

The answer is not to dump everything into a portal and call it insight.

It is to take the evidence, understand what matters and serve the right bit to the right audience.

That is where video mystery shopping can be brilliant. Used selectively, it can tell you far more about why something is happening than a conventional questionnaire ever could. Used badly, it can swallow budget, overwhelm teams and ultimately make a client question whether mystery shopping was worth doing in the first place.

And that is the bit our industry should be thinking harder about.

Sometimes the right recommendation is a large video programme.

Sometimes it is a small diagnostic piece followed by something much lighter.

Sometimes the best thing a supplier can tell you is that you do not need to spend the whole budget. Because mystery shopping is market research.

Treat it like research.

And, being frank, if your supplier is not helping you work out what you actually need to measure, how much evidence is enough and how the programme should evolve once you learn something, I would be asking a few questions… and speak to us!