How to Get Participants For Your Study
This guide outlines the range of options available for getting respondents for your experiments and helps you choose the best one for your experiment.
Buy responses from our panel network. Target them by location, demographics, and profiling questions.
Target specific pre-defined respondents (including those who are pre-defined for your company).
Use a list of your customers or leads, or source respondents from elsewhere.
How we source respondents?
Conjoint.ly partners with leading panel partners, such as Cint and Dynata, to deliver quality respondents straight into your experiment. With access to thousands of panels worldwide, we can provide you with numerous options to match your specific needs. Our partners offer broad targeting and reach that covers a diverse range of deeply-profiled audiences to ensure you receive relevant and insightful responses. If you have additional requirements for your panel, our expert team is available to support you.
Panel recruitment process
We work diligently with our panel partners to direct quality respondents to your survey through the following process:
Respondents sign up to one of our partners who ask them an extensive list of profiling questions, including (but not limited to) demographics, shopping/spending habits, and nature of household
We determine feasibility and cost by evaluating profile respondents to create time and cost efficiencies.
For example, if you require 100 iPhone users, we will send 100 survey invitations directly to people who have already identified themselves as iPhone users, rather than inviting a larger sample size (e.g. 200) with hopes that enough will fit the requirement. I.e. the more profiling we do, the cheaper the sample will be and the more accurate our feasibility estimates will be.
Once you launch an experiment with a pre-defined panel, we inform our panel providers to start inviting respondents from their panels to our survey link, using profiling where possible. Aside from receiving a survey link via email from the supplier, respondents may also find surveys on websites/apps that the supplier operates.
Respondents launch the experiment through the survey link, where they are taken to our screening questions (see more below). Respondents who do not qualify for the survey (e.g. an Android phone user) are redirected to the supplier. Qualifying respondents are redirected to Conjoint.ly to complete your experiment. If our system detects that they are providing low-quality responses (e.g. answering questions too quickly), they will be redirected to the supplier and identified as a “low-quality response”.
Upon survey completion, respondents are redirected to the supplier to claim their incentive. Following this process, we will sometimes review the actual observed cost and feasibility of the pre-defined panel and adjust the price or sample size accordingly.
Panel costs are calculated per respondent and will vary depending on country, age, and other characteristics. Our platform quotes costs in real-time as you complete the panel selection process. Once you launch your experiment, we will confirm feasibility and immediately start data collection, which normally takes less than two days.
Our panel partners manage and provide incentives for the panellists, which may include monetary payouts, coupons, points, vouchers, charity donations, and lottery draws. Respondent participation is voluntary – recruitment involves a double opt-in procedure and confirmation of personal information. As with other respondents, we monitor the quality of panel responses (e.g. how much time they spend per questions) and you will only pay for complete, quality responses.
Option 1: Target specific panel respondents
Use “Target specific panel respondents” if you would like to define your own profiling for respondents that are invited to complete your survey. This option allows you to target respondents by country, region, age, gender, and deep profiling characteristics. To select this option:
- Open the “Choose participants” tab.
- Select “Target specific panel respondents” from the “Participants module”.
- Choose your desired demographics from the drop-down menus.
Deep profiling: You can also filter respondents based on specific questions they have previously answered with our panel provider (these questions vary by country). For example, to survey people in a certain income bracket, choose a question about income and select all the answers that correspond to the people you’d like to survey. To add this filter:
- Open the “Choose participants” tab.
- Click the “Add filter” button in the “Panel responses: Profile of respondents” module.
- Select your filtering question and desired answer/s from their respective drop-down menus.
If you require more specific criteria that is not listed here, please check our list of pre-defined panels as we may already have a price and sample size for what you are looking for.
Option 2: Select from a set of pre-defined panel
This option provides you with a selection of pre-defined sample definitions. You can select either:
- A general-purpose pre-defined panel (e.g. “USA general population”), or
- A definition of panel that we have recorded from your company’s past experiments. Once you have entered the required number of responses, the system will provide an estimated cost, timeframe, and feasibility.
For all pre-defined panels, you can review the screening questions that are used to profile respondents;
Open the “Choose participants” tab.
Select your pre-defined panel from the drop-down menu in the “Pre-defined panel options menu”.
Click the “Review screening questions” button.
If you have a specific sample requirement that is not listed in the existing screening questions, please get in touch with us and we will set up pre-defined panels tailored to your company’s needs.
Option 3: Bring your own respondents
By default, each survey generates a link which you can share with your own respondents which becomes actives once you launch your experiment.
Tip: Conjoint.ly works with other survey platforms. Learn how to use URLs to integrate Conjoint.ly with survey platforms, such as Decipher, SurveyGizmo, or SurveyMonkey.
How many respondents do I need?
The system will automatically provide a recommended minimum sample size based on the settings of your experiment. This minimum sample size is acceptable for exploratory research. It is recommended to aim for a higher sample size if:
You need to achieve higher confidence in the findings (e.g. when significant policy or managerial implications are attached), or
You are aiming to explore segments. For example, if you are looking to explore two or three pre-defined segments in your experiment (e.g. people split by age: under 20 years old, 21 to 40 years old, 41 and over), apply the recommended sample size to each of the three pre-defined segments.
It is fine to oversample if you keep the proportions of your groups (e.g., if you are aiming at 50%/50% female/male split, your sample should remain 50%/50%) – in fact, the more respondents the better for statistical robustness.
Does length of interview affect cost of panel?
What is the optimal survey length?
What demographic information is included with the purchased panel?
How many additional questions can I include before, during, and after the conjoint portion?
Can I source a combination of my own respondents and a Conjoint.ly pre-defined panel?
Yes, you can include respondents from both a pre-defined panel as well as those you bring yourself.
The first step is to launch the experiment using data from a pre-defined panel — shown above in Option 2. Once you have launched your experiment, you can share the survey link directly with your own respondents, as shown in Option 3.
What is the difference between a hard and soft quota?
We achieve hard quotas to within a 5% margin of error. For example, a hard quota of 100 male and 100 female respondents will have at least 95 male and 95 female.
For soft quotas, we will attempt to fill the quota as best we can, but will prioritise keeping the speed of the data collection. Soft quotas are often used in regard to even spread of age, gender, income, or region. For example, we will attempt to spread the responses across all the income brackets, but we do not guarantee any specific numbers for any of the brackets.