Backlog assessments your clients can’t argue with.
Scored by a neutral engine, delivered under your brand. For agile consultancies, coaches, and delivery trainers: Vindex scores every story in a client’s backlog against INVEST, names the gaps in plain language, and turns the findings into a white-labeled assessment. Baseline on day one, re-score after the engagement, and show the change.
The free evaluator scores any CSV in about 30 seconds, with no signup and nothing installed. Or see how an engagement runs.
“Acceptance criteria are missing or untestable on 14 of the 20 highest-priority stories; recommended fixes are included per story.”
Stop defending your assessment.
Let the score say it.
Without evidence, telling a client their stories aren’t ready is your experience against their habit, and you’re the one who looks difficult.
Your word against theirs
You walk the backlog with the team and most stories aren't ready as written. The product owner has shipped from stories like these for years. Without a neutral measure it's opinion versus opinion, and the outsider loses.
No baseline, no proof
Engagements start without a measurable picture of the work product, so the renewal conversation runs on anecdotes and survey scores instead of what actually changed in the backlog.
Training that can't be shown to have stuck
You ran the INVEST and definition-of-ready workshop. Six weeks later nobody can say whether the real backlog got better, and the sponsor asks what they paid for.
Same conversation, different ending: a neutral score names the gap, the team can act on it, and it isn’t your opinion.
Here’s what it catches before the team commits to it.
Four stories from the sample backlog in every workspace, scored by the same engine that scores your client’s stories. Findings come worst-first, in plain language.
- 12%NW-110“Support other things the customer might need later”
No concrete outcome and nothing a tester could verify. Recommended: Remove or replace with concrete, testable needs.
- 15%NW-106“Fix bugs”
Unbounded scope, impossible to estimate or call done. Recommended: List the specific defects with reproduction steps.
- 20%NW-102“Improve the system”
Vague scope with no measurable acceptance criteria. Recommended: State which system and which outcome is expected, then add measurable acceptance criteria.
- 85%NW-101“As a dispatcher I want to see live truck positions”
Clear actor, outcome, and testable behavior. Recommended: Consider splitting map rendering from data refresh.
Assessments ask the team. Vindex scores the work.
Maturity surveys measure how people feel about the process. Vindex reads the backlog itself, story by story, so your assessment rests on the work product rather than on self-report.
The score does the pushing back
Telling a client their stories aren't ready sounds like criticism. Vindex names the gap with an objective, per-story score from a neutral engine, so it's evidence the team can act on, not your opinion.
Your methodology on every page
Every report is white-labeled: your logo, your colors, your framing. Clients never need a Vindex account and never see ours; they see an assessment from the firm they hired.
A baseline and a delta you can sell
Score the backlog on day one, coach or train, then score it again. The report shows the change round over round: proof the engagement worked, and the case for the next one.
Jira, Azure DevOps, Asana, and Monday exports, with no file-size or row limits; 100,000+ stories score unattended.
Prepaid credits with the exact cost shown before every run: a unit cost you can build into a fixed-fee engagement.
Every story is de-identified before any AI model sees it, and Vindex never trains on your client's data.
Delivery agencies and dev shops use the same workspace to gate fixed-bid scope before the SOW is signed. See how to catch scope creep in the requirements.
Five steps, one deliverable.
A project moves through a clear lifecycle, draft → scoring → review → reporting → delivered, and you can re-run it any time the client revises their backlog.
Set up the project
Create a client workspace and a project for the engagement. Everything for that client stays separated.
Upload the backlog
Drop the client's CSV or Excel export. Vindex detects the sheet and suggests a column mapping automatically.
Score it
Confirm the exact credit cost, then let the job run unattended; every story gets a health score and plain-language findings.
Review and curate
Work the findings worst-first: read the reasoning, exclude junk rows, add analyst notes, curate recommendations.
Deliver the report
Build a white-labeled report and hand off the file. Re-score later to show improvement round over round.
A report that looks like it came from you.
The output isn’t a Vindex dashboard your client has to log into; it’s a polished, self-contained report you deliver under your own brand. Compose the sections, set the branding, and generate an immutable version you can hand off with confidence.
- Your logo, colors, and about-us copy on every page
- Executive summary with backlog-health scores at a glance
- Per-story findings in plain language, worst-first
- Curated recommendations your analysts approved
- Round-over-round improvement when you re-score
- Delivered as PDF, self-contained HTML, and an Excel data export
Re-score after the client revises their stories and the report shows the delta: progress a stakeholder can see, and a natural reason for the engagement to continue.
Pay for stories scored. Nothing else.
No per-seat fees, no subscription. 100 free credits when your account is set up, then $2.50 per story, bought on demand and refunded when a score fails, so the cost of an assessment is always something you can build into a fixed-fee engagement. See pricing and the margin on a 500-story backlog.
Every re-scoring round is included: score the backlog, the client fixes their stories, score it again: round two of the same stories costs 0 credits.
A story you've already scored in a project never bills again, and if a story can't be scored you're not charged. You only ever pay for new stories that get results.
Prepaid balance carries across projects and clients. A run only starts when your balance covers it, and the exact cost, split into previously scored and new stories, is shown before you confirm.
Making the case to a client? Put their team size and ticket volume into the savings calculator and walk them through what at-risk stories cost in rework: real numbers for the pitch, editable in front of them.
The playbook, in the open.
How consultancies, coaches, and delivery firms run backlog assessments, present hard findings, price the work, and protect fixed-bid margins.
What's in the report I deliver to a client?
The white-labeled backlog assessment report, section by section, with the real sample PDF as the reference: what each section shows, why it is there in a coaching or training engagement, what you can change, and what the client receives.
What does a low-scoring story actually cost?
One story from vague to shipped defect: what a story that looks ready hides, what scoring flags in it, and who pays when nobody acts. The guide to send a product owner who says the team has always worked this way, and the pre-read for a definition-of-ready workshop.
How do I audit a backlog when the project is already slipping?
The mid-flight assessment, where the room is already taking sides: score what remains rather than re-litigating delivered sprints, split the slip into named causes a neutral engine can back, and turn the findings into a quality gate and a re-baselined date.
The evidence behind the assessment.
Case studies and field guides on what bad requirements cost, the same evidence base your reports draw on.
How a Healthcare Platform Cut Rework 68% by Shifting Quality Left
A healthcare technology company was losing 340 engineering hours a month to unclear requirements. After scoring story health before code, they cut rework 68%, recovered $564K a year, and caught HIPAA gaps in refinement instead of UAT.
Bad Scores Cost Real Money: A $1.2B Automotive Recall Study
Inside a pilot where an automotive manufacturer ran 400,000 Jira tickets through Vindex. The lowest-scoring requirements traced to $1.2 billion in software-related recall costs. The highest-scoring traced to zero.
Stop Bad Requirements: Why Small Dev Teams Need Vindex
A story looks ready on Friday and blocks a developer on Monday. Here is how small teams catch vague, oversized, and untestable work before it burns their scarcest resource: developer time.
Make requirements assessment part of every engagement.
We’ll set up your firm’s workspace, invite your first admin, and add 100 free starter credits so you can baseline a real client backlog and deliver a report before any payment conversation.