Custom questionnaire routing and research design software
Appfront builds software for research agencies, universities and research departments that design questionnaires with complex logic: routing and filters that depend on earlier answers, piped-in answers in question text, randomisation of order and conditions, experimental designs, and a test that runs every route before the questionnaire goes into the field. With versions, a codebook and an export that exactly matches what the respondent saw, so a routing error doesn't surface only in the data.
What is questionnaire routing and research design software?
A questionnaire of forty questions with routing can quickly contain hundreds of possible paths. If someone answers in question three that they don't have a car, one block is skipped; if someone names a brand in question twelve, that brand comes back in question eighteen; half of the respondents see condition A and the other half condition B. Questionnaire routing software captures that logic, lets you test it, and ensures the data is understandable afterwards.
In much research, the logic is described in a Word document, translated into a survey tool by someone else, and tested by filling it in a few times oneself. A filter that is set up slightly wrong excludes a group of respondents from a block, and this only becomes apparent when the analyst asks why there are so few answers. Randomisation that isn't recorded makes an experiment impossible to verify afterwards. And a change made halfway through fieldwork isn't documented anywhere.
We build custom solutions because the design must fit your research: which question types and designs you use, how specifications are currently created and approved, which fieldwork tool or panel the questionnaire runs on, and which statistical software reads in the data. A standard survey tool can do a lot; if you design experiments or need to demonstrate hundreds of routes, you run into its limits.
Logic that is readable
Routing, filters and conditions recorded as rules a researcher can read, with a diagram of the paths, rather than hidden inside a tool.
Every route tested
Simulated respondents work through the questionnaire along all paths. A question nobody reaches, or a filter that excludes a group, is flagged before fieldwork begins.
Design that can be demonstrated
Randomisation, conditions and experimental designs recorded, with the allocation per respondent stored in the data, so the analysis matches what was actually done.
How we build your questionnaire routing and research design software
We start with the most recent complex questionnaire: how it was specified and built, how it was tested, and which error only surfaced in the data.
Your question types and designs, how specifications are created and approved, the fieldwork tool or panel, and the statistical software that reads in the data.
The question bank with question types, recording routing, filters, passed-on answers and randomisation, and the diagram of the paths.
Automated testing of routes with simulated respondents, experimental designs, and version control with approval.
The integration with the survey tool and panel, the codebook and the export to statistical software, followed by management and expansion.
What questionnaire routing and research design software actually does
The components below come up in almost every team that designs complex questionnaires. Which ones you need depends on your designs and your fieldwork tool.
Question bank
Questions and answer scales reused across studies, with fixed codes, so that trend questions remain comparable.
Routing and filters
Conditions based on earlier answers, blocks that are skipped, and answers that reappear in later question wording, written as readable rules.
Randomisation and conditions
Random ordering of questions or answer options, allocation across conditions, and the allocation per respondent stored in the data.
Route testing
Simulated respondents that work through all paths, with a report of questions that are not reached, filters that exclude too much and empty blocks.
Versions and approval
Every change as a new version with who made it and why, approval by the project lead, and the version a respondent saw recorded in the data.
Codebook and export
A codebook with variables, labels and routing, and an export to statistical software that matches the version and the conditions.
Who we build questionnaire routing and research design software for
The software is intended for teams designing questionnaires where a standard survey tool no longer provides enough certainty.
Market research agencies
Many questionnaires a year, with complex routing and tight deadlines. The route test and the question bank are at the core.
Universities and research institutes
Experiments with conditions and vignettes, and demands for reproducibility. Designs and per-respondent allocation matter most.
Statistics bureaus and government bodies
Long questionnaires that must remain comparable year after year. The question bank, versions and codebook are what's needed.
In-house research departments
Customer and employee research with fixed trend questions and changing modules. The question bank and export are at the core.
Test your idea first: a working prototype in 1 day
With OneDayBuild, we turn your idea into something tangible in one day for €1,150, so you can see whether further development is worth the investment. Decide to go ahead with the full build? Then we credit the full cost.
Explore OneDayBuild →Technology and integrations
This page covers the design and logic of the questionnaire. For fieldwork, including offline use in the field, see our page on an survey app. For managing respondents, see our page on panel management and respondent incentives. For the resulting data, see our page on a research data management app. You can read about how we work under building software.
Why choose Appfront for your questionnaire routing and research design software?
A routing error becomes expensive the moment the first respondent starts. We build on that: logic that is readable, tested along every route, and data that matches what the respondent saw.
Errors caught before fieldwork
The route test finds the question nobody reaches and the filter that excludes a group, before any respondents are lost.
Designs you can verify
The allocation to conditions and the order per respondent are recorded in the data. The analysis can check exactly what was done.
One specification, no translation
The researcher records the logic directly in rules that the software executes. No Word document sits between design and fieldwork any more.
Security and privacy in questionnaire routing and research design software
The software holds questionnaires, some confidential to clients, and test data. Access is set up per project: researchers edit their own questionnaires, the project lead approves, and the client sees only what is shared. Every change is logged with name, time and version.
The software runs in a European data centre, with encrypted storage and daily back-ups. Respondent answers remain in the fieldwork tool or are supplied separated from identifying details.
Frequently asked questions about questionnaire routing and research design software
Questions researchers ask before they get started.
It records questions, routing, filters, carried-forward answers and randomisation as readable rules, supports experimental designs with conditions, tests all routes with simulated respondents, keeps versions with approval, and delivers a codebook and an export that match what each respondent saw.
Not necessarily. The questionnaire can be designed and tested in the software and then rolled out to the fieldwork tool you use, provided it supports an import. For designs the tool cannot handle, the software can also field the survey itself.
Simulated respondents complete the questionnaire with answers that systematically hit every condition. The report shows which questions are not reached, which filters exclude a group, and which blocks remain empty.
Allocation across conditions, vignettes with varying characteristics, and choice experiments with sets composed according to a design. The designs you use determine what we build; the allocation per respondent always appears in the data.
The change becomes a new version with a reason and approval. The data records which version each respondent saw, so the analysis can take that into account.
The export contains the variables with labels and value labels in a format that common statistical packages can read, along with a codebook. We'll look at which packages you use in the first step.
Ask that first. Modern survey tools offer routing and randomisation, and for most questionnaires that is enough. Custom development makes sense when your designs are more complex than the tool can handle, when routes must be demonstrably tested, or when a question bank needs to remain comparable over years.
Find a routing error before the first respondent?
Tell us which questionnaires and designs you build, which fieldwork tool you use and how you test now. We'll show you what the logic, the route test and the export look like.