National Library of New Zealand
Harvested by the National Library of New Zealand on: Apr 16 2016 at 9:22:22 GMT
Search boxes and external links may not function. Having trouble viewing this page? Click here
Close Minimize Help
Wayback Machine
The Department of Internal Affairs

The Department of Internal Affairs

Te Tari Taiwhenua

Building a safe, prosperous and respected nation


Services › Casino and Non-Casino Gaming › Mystery Shopper campaign

The Department of Internal Affairs is working with gambling operators to improve their identification and management of problem gamblers following the results of its new research.

In the first of its kind for the gambling regulator, Internal Affairs initiated a mystery shopper campaign that sent undercover gamblers to casinos, pubs, and hotels to test whether patrons showing signs of potential gambling addiction to pokie machines were adequately dealt with by venue staff.

For more information about the mystery shopper campaign, view the information below.

Mystery Shopper Project Summary

Mystery Shopper Information Summary

Casino Data Summaries

PowerPoint slides

Media release

Questions and answers


Q: What was the point of the exercise?

The exercise was designed to test how staff at gambling venues put their problem gambling training into use, and the extent to which potential problem gamblers may still be at risk of gambling harm.

It was not undertaken for enforcement purposes but to provide a baseline of information and to identify areas for improvement.

The Gambling Act 2003 requires casino and class 4 operators to develop and implement policies for identifying problem gamblers.

Q: Does this exercise show that there is lots of problem gambling happening?

No. The prevalence of problem gambling has not increased.

This exercise reveals new information and that is about how well staff are identifying problem gamblers.

Q: Will the proposed changes to problem gambling services make the situation worse?

No. The Ministry of Health is committed to ensuring that free to use services for those experiencing harm from their own or some else’s gambling are continued. Current services are being maintained until 30 June 2015.

Q: Don’t the results show that the sector is completely failing with its harm minimisation obligations/practices?

Studies indicate that most people gamble without negative effects. However, we are concerned about the harm that gambling can have and particularly concerned about problem gambling. The results of our research are disappointing.

The exercise focussed on specific areas of gambling staff’s harm minimisation obligations, so do not reflect all harm minimisation practices. However, the results in this area reflect badly on the gambling operators who have an obligation to minimise problem gambling harm.

Q: How was the project undertaken?

An independent company experienced in mystery shopper exercises was contracted to undertake the visits.

The exercise was undertaken from May to July 2014.

Gambling operators were not informed the exercise was underway.

Q: How much did the shoppers gamble? Who paid for this?

For the Class 4 exercise shoppers had a float of $40.

They started gambling with $20 and then on two occasions withdrew a further $20 from venue staff. The last $20 withdrawn was accompanied by a scripted problem gambling statement and was not gambled.

For the casino exercise the floats varied depending upon the scenario and length of time the shopper was in the casino. The floats ranged from $240 - $480.

These costs were included in the overall cost of the project.

Q: How much did the project cost?

The project cost around $57,000 ($34,000 Class 4; $23,000 Casinos)

Q: Did any of the shoppers win money from their gambling? If so what happened to the winnings?

Winnings from Class 4 ($900) are to be distributed back to societies.

Winnings from casinos were re-gambled as floats – overall there were no winnings.

Q: How did you ensure shoppers didn’t become addicted to gambling as part of the exercise?

The Department undertook a number of steps to ensure this risk was minimised. This included:
  • Screening undertaken of all mystery shoppers to determine no prior gambling dependency problems.
  • A clause in the mystery shopper contract declaring no such past problem.
  • Comprehensive training provided, including focus on how machines work, gambling odds, and signs of problem gambling.
  • Game play restricted to set limits and rates.
  • Shoppers introduced to problem gambling service providers; contact details and brochures provided.
  • Second screening of casino shoppers by problem gambling service provider.
  • Further assistance offered at request.

Q: Were any of your shoppers sprung / identified?

Not that we are aware of.

Q: What are the details for each sector and the results?

Class 4 venues (pubs and hotels)

102 venues were visited. This equates to 10 percent of New Zealand venues. They were a mix of urban and rural, and a mix of size.

Shoppers gambled for two hours (generally during the day) and displayed general problem gambling indicators such as sighing, head resting on hands, talking to machine, expressing frustration.

They observed the sweeping of pokie rooms by staff, and other patron behaviours.

They role played and re-created one of the following scenarios while withdrawing $20 cash from a staff member by saying:
  • “I need to go but I need to win some money back “ (used in 77% of scenarios)
  • “I can’t really afford it but I think I’m getting close to a win” (used 16%)
  • “I’m meant to get home to the kids but another few minutes won’t hurt” (used 7%)
Interventions that we expected to see staff use:
  • Questioning whether it was a good idea to withdrawal the money.
  • Asking if the person is OK.
  • Suggesting they take a break.
  • Providing problem gambling information to the person.
  • Suggesting they leave the venue.
  • Asking them about their gambling.


99% of scripted scenarios delivered by mystery shoppers did not result in an intervention from staff.

However, comments recorded indicate many scripted scenarios caught the attention of staff (e.g. the staff member seemed embarrassed or said something else in response).

The Department is in discussion with gambling operators to see whether these scenarios were at least recorded in any incident logs held by the venues.

Recognised best practice for monitoring gambling areas is to conduct ‘sweeps’ of the area every 15 minutes – that equals between six to eight sweeps for our scenarios.

Fourteen venues achieved the desired result of six or more sweeps. Ten had no sweeps at all.

A “sweep” of a gambling area refers to a staff member entering the area and checking on gambling patrons to develop an awareness of patrons and their behaviour, and to ensure they are not displaying signs of gambling harm.

Floor sweepsInstances% Sweep Count
1 sweep2120.6%
2 sweeps2221.6%
3 sweeps2019.6%
4 sweeps109.8%
5 sweeps54.9%
6 sweeps22.0%
7 sweeps22.0%
8+ sweeps109.8%
No sweeps109.8%
Grand Total102100.0%
Comments indicate that many sweeps involved staff focussing on other activities (e.g. re-filling a hopper).

Q: What does this indicate?

The results indicate that the need to improve harm minimisation practices is a nationwide issue - not a problem faced by only one or two societies.

Staff in pubs and hotels may not be adequately trained to carry out their harm minimisation obligations.

Staff in pubs and hotels may be aware of what they should do, but choose not to put it into practice due to a number of barriers, conflicting priorities, or set views over the thresholds or behaviours that indicate problem gambling.


A total of 16 scenarios were carried out across all casinos:
  • Skycity Auckland = five visits
  • Christchurch casino = three visits
  • Skycity Hamilton, Dunedin casino, Skycity Queenstown, Skycity Wharf (Queenstown)= two visits
Five different scenarios were carried out:
  • Length of play: 10 hours of play – no problem gambling indicators displayed.
  • Length of play: 10-12 hours of play plus problem gambling indicators displayed while playing gaming machines.
  • Frequent cash withdrawals from the cashiers – with problem gambling indicators displayed while playing gaming machines.
  • Frequent cash withdrawals from an ATM– with problem gambling indicators displayed while playing gaming machines.
  • Setting gambling spend limits / pre-commitment – with problem gambling indicators displayed while playing gaming machines.
The exercise focussed on behavioural indicators of potential problem gambling - rather than casino use of gambling data generated by carded players. (Carded players are those with casino loyalty-type cards).

What were the casino scenarios based on?

All New Zealand casinos have a Host Responsibility Programme and Problem Gambler Identification Policy (the Policy) approved by the Gambling Commission.

The Policy sets out a selection of “Strong Indicators” and “General Indicators” of potential problem gambling. Casino staff are trained to recognise these signs and there are requirements to record all observations made.

The Department set out to test the casinos response to the “General Indicators”. The General Indicators tested (which are set out under various headings in the Policy) include:

Intensity and Frequency of Play
  • Very few breaks from gambling – almost continuous play
Visible Emotional Disturbance
  • Visible emotional disturbance such as agitation, holding head in hands, personalising machines, rudeness and complaints to staff about gambling outcomes
Excessive Access to Money
  • Repeat ATM visits and/or multiple declined transactions
Dysfunction in Social Behaviour
  • Claims of malfunction of gaming machines or gaming errors
One of the scenarios was an opportunity to stretch casino systems and host responsibility. In this scenario, prior to displaying behavioural indicators, the mystery shopper seeks to set limits of spend and length of play. This indirect disclosure of limits of time and money, which were then breached, provided an opportunity to casino staff.


14 out of 16 scenarios did not receive any known direct intervention from casino staff to indicate they had noticed the behaviour.

One direct intervention was where a shopper asked about pre-set expenditure limits on arrival and it resulted in information being given about problem gambling.

In one other scenario a staff member asked if a shopper was OK, but did not take any follow up action or record the interaction.

Interventions that we expected to see staff use (and record):
  • Asking if the person is OK.
  • Suggesting they take a break.
  • Providing problem gambling information to the person.
  • Suggesting they leave the venue.
  • Asking them about their gambling.
Staff were observed visibly monitoring the gambling floor for the majority of casinos and in some scenarios staff engaged in friendly conversations with the undercover gambler. But no specific intervention was made for fourteen of the scenarios

It is possible these interactions involved casino staff making an assessment of the shopper, but checks of the daily logs held by casinos (a record of all security and harm minimisation incidents and observations) do not indicate this.

What does this indicate?

  • Improvement is needed by the casinos to monitor behavioural indicators of potential problem gambling.
  • Casino staff may be using casual conversations to establish a rapport which can build an environment in which customers are more willing to ask for help if they’re experiencing gambling problems.

Design of the project.

Q: How robust is the mystery shopper exercise? Can we trust the results?

Mystery shopper exercises aren’t a perfect science; however, given the scale of the exercise in pubs and hotels we’re confident the results depict an accurate picture of harm minimisation practices.

Although there were fewer casino visits, the results have given us enough cause for concern that potential problem gamblers could slip through the cracks.

The mystery shoppers were carefully selected through a screening process and were given comprehensive training before undertaking the exercises.

Department’s current / future actions:

Q: Why isn’t the Department taking enforcement action?

At this stage we believe education and support is the best way to improve the situation. If we believe progress is not being made this approach is likely to change.

Q: What are you going to do about the results?

We will work with gambling operators to ensure they are meeting their obligations.

In class 4 – pubs and hotels - we currently have a programme of work underway focussed on initiatives designed to reduce problem gambling, such as checking that the regulations for gaming rooms are complied with.

More specifically we are reviewing the training that venue staff receive from the gaming machine societies which own the pokie machines at venues and are responsible for this training. We are also supporting the Health Promotion Agency in its work to improve harm minimisation at venues.

For casinos we will work with them to improve processes to ensure they are meeting their host responsibility policy requirements.

Q: How will the Gambling reforms help?

Reforms in the Gambling Amendment Bill (No 2) which is before the House will help, by providing the ability to prescribe the content and standards of harm minimisation training in the class 4 sector.

The Bill also includes the extension of the people required to identify actual or potential problem gamblers.

There are also measures to improve the use of exclusion orders for problem gamblers.

Q: Will you undertake another mystery shopper exercise?

A follow up exercise is a definite option. Now that we have this baseline information, a follow up exercise would allow us to gauge whether gambling operators have successfully improved the situation.

This would be in addition to the activity we do continually with the sector which involves inspections, audits and specific projects to target areas of concern.

*You need Adobe Reader installed on your computer to view our files. If you are unable to open the files we recommend you get the latest version of Adobe Reader. You can download and install Acrobat Reader for free from the Adobe website.

** This document is in Microsoft PowerPoint (PPT) format. You need to have Microsoft PowerPoint viewer installed on your computer. You can download a free version from the Microsoft site.