How incident.io made R&D tax claims effortless and audit-ready

incident.io is a fast-growing incident management platform used by modern engineering teams. Like many UK tech companies, they invest heavily in R&D — and file annual R&D tax relief claims.

Company

Series B SaaS company, industry leader in incident management

Team Size

40+ engineers

Region

United Kingdom

Challenge

Annual R&D claims required manual reconstruction of project context, senior engineer involvement, and repeated narrative reviews

Solution

Taxnova connected to incident.io's engineering systems to automatically identify, evidence, and structure R&D activity

Result

~80% reduction in preparation effort, clearer documentation, and materially improved evidential trail in line with HMRC standards

The problem: great engineering, painful paperwork

Each year, the R&D claim followed the same pattern:

  • Engineers were asked to reconstruct work done months earlier

  • Technical leads had to translate in-depth concepts and technical language into HMRC-friendly language

  • Large questionnaires circulated back and forth

  • The process was very disruptive, despite only requiring a short period of focused effort

The claim process meant weeks of senior engineers' time every year — time they'd rather spend on actual R&D.

And as HMRC requirements increased over the years, the pressure shifted from better wording to better evidence.

A different approach: start from real engineering data

incident.io first encountered Taxnova through a conversation with its founders. What caught their attention wasn't automation for its own sake, but a simple idea:

What if R&D claims started from the systems where engineering work actually lives, not from reconstructed memories?

Instead of relying on memory, interviews and rewritten narratives, Taxnova connects directly to tools like GitHub, Linear, Notion, Slack, and even blog articles to understand:

  • What work was actually done

  • Who worked on it

  • How projects evolved over time

  • Where evidence already exists — and where it doesn't

incident.io decided to trial Taxnova alongside their existing advisor, focusing initially on the AI engineering team where most R&D activity was concentrated.

Why Taxnova was selected

incident.io was open to a different approach. Rather than automating questionnaires or writing narratives faster, Taxnova proposed something simpler: start from real engineering data.

The key factors:

  • Traditional interview-based approaches weren't scaling with the size of the company

  • Manual spreadsheets lacked traceability

  • Evidence quality mattered more than claim breadth under increased scrutiny

What changed

1

The process became dramatically lighter

Compared to the previous year, preparation effort dropped by around 80%.

  • No long interview cycles with engineers

  • No chasing people for historical context

  • Only a small number of targeted follow-up questions, where the data alone wasn't enough

  • All while their existing advisor stayed in control of final review and submission

The process was described simply as "super easy".

2

The output quality exceeded expectations

incident.io was impressed by the results:

  • Clear, structured R&D project narratives

  • Explicit assumptions and uncertainties (rather than vague claims)

  • Strong traceability from narrative statements and R&D% calculations back to real tickets, commits and documents

Taxnova made it much easier to see which projects were genuinely robust — and which weren't worth including.

3

Less risk, more confidence

incident.io could see exactly:

  • What evidence supported each claim

  • Where gaps existed

This made it much easier to see which projects were genuinely robust — and which weren't worth including. That clarity reduced anxiety and made internal decision-making simpler.

The bigger insight: the annual scramble is the real problem

The most positive reaction came when discussing what happens next.

Taxnova's roadmap focuses on continuous R&D data collection — capturing context while projects are active, not a year later. Lightweight prompts, ongoing evidence, and early visibility into weak or strong projects.

For incident.io, this clicked immediately. Instead of an annual fire-drill, R&D claims could become something that's quietly prepared in the background, ready when needed.

"I just don't know why someone hasn't already done this" — Senior Engineering Manager at incident.io

incident.io has also shared they can't see a scenario where they wouldn't continue preparing claims with Taxnova.

The takeaway

By grounding R&D claims in real engineering data, Taxnova helped incident.io:

  • Cut preparation time by ~80%

  • Reduce disruption to senior engineers

  • Produce clearer, more defensible R&D documentation

  • Approach HMRC scrutiny with far greater confidence

Not by replacing advisors — but by finally fixing the hardest part of the process.


Ready to see similar results for your company?