Guides

Automate the Steps You Already Send Away

Megan Daniels
Megan DanielsCEO
· Updated

A study can pass through a programming vendor, a data-processing team, coders, and a separate resource for the deck. Each handoff needs a brief, review, and sometimes another round of corrections.

If your agency outsources these steps, they are a practical place to start with automation. You already have a process and an invoice to compare against.

The steps already leaving the building

Look at the production work your firm sends to other teams:

Scripting. The survey gets sent to a programming vendor to be built and tested.

Data processing. Raw data goes to a DP shop for cleaning, tabulation, and table production.

Open-end coding. Verbatims get shipped to an offshore team to be read and categorized.

Charting and production. The deck build goes to a production resource to be templated and formatted.

The MX8 Labs Research Platform connects programming, QA, fieldwork, analysis, and reporting. You can generate a survey from a supplied questionnaire, review coded open ends, and work with report drafts in the same project.

Compare the cost of a step

Start with a task you currently commission, such as programming a questionnaire or coding open ends. Compare the vendor invoice with the platform cost and your team's review time. That shows where bringing the task into your workflow saves money.

Your team reviews the programmed logic, coding frames, and report output before using them. Include that work in the comparison.

Count the handoffs too

A late questionnaire change may mean briefing the programmer again and waiting for a new link. In one workflow, you can make the change, inspect the updated survey, and test it yourself.

Record the time spent briefing, waiting, reviewing, and making corrections. Fewer handoffs can shorten the study even when the production task itself takes a similar amount of time.

Start with one step

Pick the outsourced task that causes the most delay and run it on the platform for a study or two. Compare the cost, turnaround, corrections, and final output with your usual process. Use that experience to decide what to bring in next.