Worker voice - things to know

The Ask the workers platform connects workers anonymously, collectively and continuously to their workplace management team and to stakeholders.

This is to enable the workplace management team and its stakeholders to hear directly from workers about how they are treated.

The platform covers all the workers, all the time, and provides real-time results. It is continuous. It is not a survey and it is not based on a sample of the workers.

With continuous real-time transparency direct from the workers, the Ask the workers platform discourages deliberate poor practices and encourages workplaces to self-remediate matters which are being raised. Workplace C-suite management has real-time visibility on the effectiveness of their ethical policies and their implementation,

Ask the workers measures discrimination and harassment, abuses that are only reported by workers who are confident of their anonymity and safety from retailiation.

Ask the workers provides analysis by different worker demographics (gender, nationality, employment type, job role) although results are monitored and can be suppressed in order to ensure that anonymity of workers is not compromised.

The platform:

• discourages deliberate poor practices due to the real-time transparency provided by workers to workplace management and stakeholders.

• provides workplace C-suite management real-time visibility on how effectively their ethical policies are being delivered in the workplace.

• measures discrimination and harassment, abuses that are only reported when workers are confident of anonymity/ safety from retaliation and remedy.

• identifies when worker demographics have a bearing on their treatment, as long as reporting numbers are large enough to provide anonymity and responsible reporting. Many unethical practices only affect cohorts of workers and are hidden when reporting is based on average results or is infrequently carried out (eg: 6 monthly or annual surveys or audits).

• is cheaper to set up and run because it is continuous, when compared to the start-stop-start process of a survey.

There are four parts to the Ask the workers platform:

- the Ask the workers app which is used by workers to provide feedback

- the Ask the workers dashboards which are used by workplaces and stakeholders to review results, identify good practices and monitor the impact of any improvements or remediation steps taken

- the Ask the workers advisor which can be used by stakeholders to get automated, in-context specific advice on potential root cause and improvement steps given worker response patterns

- the Ask the workers analysis tool which is used by stakeholders to dig into the data in order to recognise developing risks and trends

Key definitions we use:

- Worker: someone who is working at the workplace (see other answers for how this is checked)

- Anonymously: means that it is not possible to find out what any one individual worker might be saying, feedback is confidential

- Collectively: means that feedback is collective and not individual - it comes from the workforce as a whole built up from individual responses

- Workplace or site: the place where a worker works that is responsible for the day-to-day activities of the worker

- Sponsors: a third party such as a customer, brand, property or site manager, landlord, franchisor, NGO, consulting firm, trade union that has an interest in making sure workers are treated fairly.

Using the app - answering questions

Every day, workers have the opportunity to provide answers to questions via the Ask the workers app about how they are treated. Questions are in the language of the phone that the worker is using. It takes maybe one minute to cycle through the app giving answers. Workers can do this any time they wish and from anywhere, as long as they have internet access.

The total question set is approaching 80 questions (and can be increased). Since workers can visit the app multiple times and every day, a worker only sees 10 questions each time pulled randomly from the library of available questions. Since all the workers are included all the time, it means all the questions are answered every day and individual workers see all the questions over time (typically across a calendar month).

The app covers all the workers, all the time.

It is not a survey. It is continuous. Workers can use the app as often as they like but not more than once per day.

Continuous operation is cheaper to set up and run than a survey - reflected in the pricing of the Ask the workers plaform.

Using the app - sending messages (grievance)

If enabled for a given workplace, the app provides a grievance channel that workers can use to send messages. This is free format text and messages can be about anything that workers want to say.

Our recommendation is that the grievance channel is only used if there is a suitable triage process in place to receive and review the messages. Our development team is currently working on an AI driven system to interpret and triage messages - which, given the dangers of messages being inappropriately handled, requires a careful approach and a lot of testing (so this will not be appearing for some time).

Prize draws

We believe that workers should be compensated for providing information that is valuable to sustaining ethical supply chains. Direct payments being commercially and administratively challenging, we believe that prize draws are a feasible step in the right direction and provide workers with a further incentive to cooperate in reporting.

Sponsors can (and should) set up a compensation program for workers in the form of a prize draw that provides small monetary prizes to workers in return for workers using the app.

This is typically a low cost (from US$0.01 to US$0.10 per worker per month).

The prize draw budget can be changed from month to month depending on usage rates of the app or to drive periods of intensive reporting (like a survey) followed by periods of monitoring in between. During an intensive period, prize draws are increased, reduced during periods of pure monitoring.

Each time the worker runs the app and answers the questions (taking one minute or so), the worker has a chance to win a prize - with the award being notified on the next occasion the worker uses the app.

The prize draw is automatically managed by the Ask the workers platform.

We support two approaches:

The open access model (location code):

There are situations where workplaces (or stakeholders) do not have easy access to a worker list. This can be where the worker population on site is changing frequently because of the use of casual, agency or sub-contracted labour. In these situations, there is no easy way to establish and maintain a worker list.

In this situation, we offer "location code" based access:

1) We generate, for each location within each workplace, a unique location code and this is put on a noticeboard or other location in the workplace which workers have access to (but the general public does not).

2) Workers download our app. We still take zero permissions on the mobile phone in order workers can be confident that their privacy is protected and that they are not being spied on.

3) Workers enter the location code on first use, tying them to a location. Workers enter basic demographic information about themselves that we need to support reporting (eg: age range, nationality, gender, employment type etc). and then they are good to go.

4) We still do not ask for worker names, email addresses, telephone numbers or other personal information. There is zero invasion of privacy.

5) Workers report on conditions using the app in the usual way. Periodically the location code will expire and workers will need to get an updated code from their location to continue. This helps to ensure that workers who leave the location are no longer able to report on it.

The worker list model:

1) The workplace uploads a list of workers to our system. We expect this to be all the workers including sub-contractors on site, temporary, casual and agency workers. This is typically quite easy to audit - but see the "open access model" below, if this is not feasible.

2) We do not know any names, emails or phone numbers. For each worker we request their ID number (eg: payroll or security pass) and their date of birth via the worker list provided by the workplace (or sometimes the stakeholder or sometimes the agencies and sub-contractors involved).

3) When a worker downloads our app, he picks his workplace and then enters the correct ID number and date of birth and he is registered. He only has to do this once.

4) In this way, we ensure that workers can only report on the workplaces where they actually work.

Periodically, based on turnover, the workplace will upload new lists of workers to our system representing the worker population at that moment. This will automaticlaly add new workers, and remove workers who leave (who will find that they can no longer use our app to rate their former workplace).

We do not ask for worker names, email addresses, telephone numbers or other very personal information (except date of birth).

The open access model has a lower level of authenticity when compared to the worker list model, but it has been proven to be a strong approach - because workers feel truly anonymous since they do not provide any details to the app which could link back to them.

The open access model (location code) is the recommended model to use if possible.

Continuous operation

The Ask the workers platform runs all the time for all the workers - and for stakeholders, can run for all the suppliers (1,000s of suppliers, 1,000,000s of workers).

As we all recognise, little in life or business is constant; conditions today may change tomorrow and due diligence should ideally track any changing conditions for workers day to day, week to week, month to month.

Workers report issues as they arise, allowing swift remediation and limiting the risk to workers and stakeholder reputation.

Workers report on changing conditions, allowing tracking and verification of remedy direct from workers themselves and removing the need for workplaces to send stakeholders evidence of remedy – substantially reducing the administrative burden and expert resources involved in tracking whether remedy has been delivered when surveys and social audits are used.

This enables:

- The platform to be set up once, and then to run continuously thereafter.

- Trends to be established, showing issues as they arise and the impact of remedies.

- Workplaces see the data themselves via their dashboard providing the opportunity and incentive for them to self-remediate without requiring stakeholder intervention (and we have evidence from case studies that this works).

- Issues with sampling and potential opportunities to skew results via sample selection and coaching to be avoided.

- Coaching to be detected by comparing how responses from workers are changing over time and how responses from individual workers compare to the population as a whole.

- Turnover in the workforce is monitored, for example, if there are arrivals and departures or the use of migrant workers, casual workers or agency workers.

Survey model

Ask the workers can be used like a survey with periods of intensive monitoring followed by checkpoints and formal reporting. Our clients get the best of both worlds - the discipline of a survey approach combined with the much lower cost and greater authenticity of continuous monitoring.

These responses are aggregated and analysed against standard pillars (we use the ETI base codes) to create a score for the workplace as a whole, for each base code, and ultimately for each individual question. Stakeholders can see the aggregated results (but not the answers of an individual worker).

This "sound-bite" level is useful to track trends and to rank workplaces, but we recommend that other methodologies are applied to detect whether systemic harms are occuring in the workplace, especially if there are minority groups of workers or a mix of migrant and local workers involved. This is discussed further below (dangers of using averages).

Stakeholders (and workplaces) are able to drill through the data to isolate specific demographic cohorts of workers and see their responses over selected time periods - enabling trends to be seen.

Moreover, this picks up immediately when workers raise an issue (lines dip) and when there is an appropriate remedy applied (lines recover). We can show you live examples of this from our datasets:

- Workers reporting issues with sick pay, over time, harassment, safety

- Remedies being applied by the workplace (without interference from stakeholders) and responses reverting to expected norms

Cohorts can be created across gender, nationality, employment type (employed, agency, migrant worker) and job role. NOTE: we automatically suppress the data if the number of responses in a given cohort becomes too small so as to avoid individual workers being identified. We call this "gating".

The danger of using averages

A common question from clients is how we manage the data if some workers report more frequently than others, because this skews the average result that is obtained.

The key point here is that averages are dangerous. We do provide "average results" but these are based on the average response of each worker, not the average of all the responses (ie: it is an average of the responding workers). So this deals with the potential for results to be skewed towards overly-active workers.

But there is another important point. Averages are dangerous. If 95% of workers are reporting that they are not being harassed - this sounds good. Especially if it was 90% last month. Yes, we support that, it is a move in the right direction. But if there are 1000 workers, it means 50 workers are still reporting harassment. This is not a comfortable result. The idea that a certain average level of response is a "pass" or a result that implies no remedial action is required is usually wrong.

Our dashboards also offer a number of tools that enable the edges of the response distribution to be investigated properly - because usually, if worker rights are abused, it occurs in marginal populations (small cohorts) not systemically across entire businesses.