Primary research · Survey data · QC
From raw survey data to research-ready tables
Cleaning, validation, weighting, tabulation and QC for quantitative market research. Your format, your template, your deadline.
What slows research projects down the most?
For agencies of 1–10, data processing is the biggest bottleneck — 61% vs 34% at larger agencies.
n=47 · demo dataset · updated Sep 2026
What I do
Everything between raw data and the deck
Eight things I do, usually as one pipeline — but you can take any single piece.
-
01
Data cleaning
Duplicates, straightliners, speeders, missing values and open-end tidying — with a short QC note on exactly what was removed and why.
-
02
Formatting & labelling
Clean value labels, consistent coding and a variable structure that opens correctly in SPSS, Excel, Q or your platform export.
-
03
Validation & QC
Quota checks, logic checks, outlier flags and inconsistency reports — before anything gets analysed.
-
04
Logic & routing testing
Skip patterns, branching and piping tested against the questionnaire, so every respondent took the path they should have.
-
05
Weighting
RIM and post-stratification weighting to your target profile, with the efficiency and extreme-weight checks that come with it.
-
06
Cross-tabulation
Full banner sets with significance testing, nets, means and indices — in your house table format, not mine.
-
07
Chart packs
Excel and PowerPoint charts built to your template and brand, ready to drop into the client deck.
-
08
Report preparation
Summary tables, topline decks and the written read-through that turns the numbers into a finding.
Live data
The same data, cut three ways
This is the part clients actually pay for — not the chart, but the cut underneath it. Switch between views to see it move.
Shaded band = 95% CI · gap vs category labelled at the latest wave
Demo dataset, synthetic. Built the same way I'd build yours.
How it works
Four steps, no surprises
You'll know what's happening at each stage, and what you're getting at the end of it.
-
Brief & audit
You send the data and the banner plan. I open the file, check the structure against the questionnaire, and come back with anything that looks off — usually within a day.
-
Clean & validate
Cleaning, logic checks and QC. You get the clean file plus a short note listing what was removed, flagged or recoded.
-
Weight & tabulate
Weighting to your targets, then the full banner set with significance testing — in your table format.
-
Charts & handover
Chart pack in your template, final files, and a walkthrough of anything worth flagging in the data.
About
One person, the whole pipeline
I'm Amjad, a research analyst based in India. For over six years I've worked on quantitative primary research — data processing, tabulation and reporting for agencies running trackers, U&A studies, segmentations and brand work.
I work the way an in-house DP team works, but as a single named person. You brief one person, that person does the work, and that person answers your questions about it.
Most of my clients are small agencies and independent consultants who have more fieldwork than processing capacity. I work white-label: your template, your branding, and your client never needs to know I was involved.
- 0
- Years in quant MR
- 0
- Tools in the stack
- 0
- Typical first reply
- 0
- Person on your file
- Agency-standard QC on every deliverable
- Your table format and template, not a generic one
- White-label by default
- Overlap with UK, European and Gulf working hours
- NDA before anything starts, if you want one
- SPSS
- WinCross
- Q
- Decipher
- Power BI
- Excel
- PowerPoint
FAQ
Common questions
Which tools do you work in?
SPSS, WinCross, Q, Decipher, Power BI and Excel. I deliver in your format — if your tables come out of a specific system, send me an old set and I'll match it.
What's your turnaround?
Small jobs — cleaning, a short banner set — usually 2 to 3 working days. Full studies with weighting and a chart pack, 5 to 10. Rush work is possible if you tell me early.
How do you handle messy data?
Audit first. I'd rather spend an hour understanding the structure than fix the same problem four times later. You get a written note of every decision made on the file.
Do you work white-label?
Yes, by default. Your template, your branding, no mention of me anywhere in the deliverable.
What about confidentiality?
Your data stays on one encrypted machine. It never goes into any cloud AI tool. I delete project files 30 days after handover, and I'm happy to sign an NDA before you send anything.
How do you price?
Per project, not hourly. Tell me the number of completes, the banner size and what you need at the end, and you'll get a fixed quote — usually the same day.
What do you need to start?
The data file, the questionnaire and the banner plan. If the banner plan doesn't exist yet, I can draft one from the questionnaire.
Get in touch
Send me a file and a deadline
Tell me what the study is and what you need at the end of it. You'll get a clear scope and a fixed quote back, usually within a day.