130Leads generatedUS decision-makers who submitted their information
46Intro calls booked9 in the launch period, 37 once scaled
136K+Professionals reachedSenior US finance and technical leadership
Budget scale-upClient confidence to grow spend within two months
3Landing pages shippedUS, New York and France market entries
6Conversion blockers foundHalf already fixed, all documented

Two consecutive reporting periods between 29 January and 27 March 2026. Client identity, personnel and logos withheld under confidentiality.

Booked intro calls in the sales team's calendar, each showing date, time, two participants and the call agenda
Figure 1. Booked intro calls straight from the sales team's live calendar — each one a real event with a date, a time, two participants and a fixed agenda: an introduction to innoscripta's R&D Tax Service and Platform. Participant names and email addresses are masked; nothing else is altered.

The client

One of Europe's most established R&D tax credit specialists. Munich-based, 2,100+ companies served across manufacturing, consumer goods, banking, logistics, food and life sciences.

Their platform runs raw project data through an innovation test built by Master's and PhD-level experts, and produces an audit-ready claim file. In Germany, the incumbent. In the United States, unknown.

The mandate: build US pipeline from zero — in a category where the buyer's dominant emotion is fear of an IRS audit, and the dominant objection is "we already have a CPA."

The challenge

1

The product is invisible

Nobody wakes up wanting R&D tax credit documentation software. They want money back without risk. The offer had to be reframed from software to outcome.

2

The buyer is scared, not curious

CFOs and Controllers had all heard the same rumour — that claiming the credit invites an audit. "Maximise your claim" triggers exactly the wrong instinct.

3

Volume alone was worthless

The client's economics only work above a certain company size. A funnel full of five-person startups looks excellent on a dashboard and produces nothing.

Three market entries, three arguments

Each market got its own front door, matched to how that market actually thinks about the incentive.

United States

The audit-anxiety angle

"How confident are you in your R&D documentation if the IRS audits tomorrow?"

Leads with Form 6765 and the IRS four-part test. Compliance-first promise: we'd rather under-claim a little and keep you 100% safe. The Head of US Strategy fronts the consultation, so the call has a real person attached.

New York

The emotional variant

"Revolutionize Your R&D Tax Credit Process — IRS Safe and Stress Free"

Aerial New York hero video, with CTAs written as feelings rather than functions: Feel the Joy of IRS Compliance. Three pillars under the fold — accurate calculation, audit-proof compliance, easy integration.

France

The CIR angle

"Sécurisez votre déclaration CIR dès aujourd'hui."

Different incentive, entirely different frame. Centralises CIR scientific and fiscal evidence, with an audit export showing live project statuses. The France Consulting Director fronts the booking.

New York market landing page variant, above the fold, with aerial hero video
Figure 2. The New York variant — emotional headline, feeling-led CTAs, and an aerial hero video carrying the market signal before a single word of body copy is read.
French CIR landing page, above the fold
Figure 3. The French CIR entry point. Same platform, entirely different argument — the incentive, the vocabulary and the proof visual are all localised. Personnel name and product name redacted.

The creative engine

We ran a wide test rather than betting on one idea, then let spend follow the winners. Two ideas separated from the pack almost immediately — and one came back from the dead.

Ad creative: Claim R&D and they will pay, showing a $380K estimated credit marked APPROVED
Winner

"Stop Leaving Money on the Table"

Loss aversion
47Leads

"80% of companies leave R&D tax credits unclaimed because documentation feels impossible."

$0 upfront IRS audit-ready Zero engineer work
44% of total creative spend · top performer in both periods
Ad creative: Claim up to $750,000 in innovation funding, with a phone mockup of approved claims
Turnaround

"750K Claims Processed"

Proof of scale, rebuilt
21Leads

"Claim up to $750,000 in innovation funding — strong project stories, less tax consultant time."

Concrete ceiling Cost objection Phone mockup
Launched as one of the weakest · rewritten, not rebuilt · returned top-rated
28 Leads · highest of launch period
Winner

"$7.6 Million in Claims"

Proof of scale

No cleverness — just a number large enough to make a finance leader stop scrolling.

Creative not shown
35% of launch-period creative spend

The workhorse ad is the one on the left: a visible $380K estimated credit stamped APPROVED, and three risk-removal badges — $0 upfront, IRS audit-ready, zero engineer work — dismantling the three reasons a CFO says no before they got the chance to say it.

Two creatives produced 52 of the launch period's 81 leads — 64% of all results from two ideas.

The instructive failure

A Reddit-style video, a "Zero Hassle" simplicity angle and a "$0/Month Risk-Free" offer all lost and were cut. The last one is the one worth studying: for this buyer, removing price risk is far less persuasive than removing audit risk. That insight now shapes every new creative brief on the account.

The full test matrix

Creative conceptAngleLeadsVerdict
"Stop Leaving Money on the Table"Loss aversion47Winner · 44% of spend
"$7.6 Million in Claims"Proof of scale28Winner · 35% of launch spend
"750K Claims Processed"Proof of scale, rebuilt21Turnaround · worst to best
"Reddit-Style Video"Social proof5Cut
"Track Your R&D"Product / control3Efficient but under-funded
"Runway Ad"Brand / awareness2Efficient but under-funded
"Zero Hassle"Simplicity1Cut
"$0/Month Risk-Free"Price-risk removal0Cut · the useful failure

Creative-level attribution across both periods. "Track Your R&D" and "Runway Ad" produced leads at genuinely efficient rates — they simply never got enough budget to prove it. They are the first two candidates for the next scaling round, and they only surfaced because the test stayed wide.

Results

46 intro calls booked across the two periods. Reach nearly doubled between them.

Period over period

Launch (29 Jan – 25 Feb) compared with scaled (1 – 27 Mar 2026)

Professionals reached

Launch46,027
Scaled90,670

Intro calls booked

Launch9
Scaled37

Each panel is scaled to its own metric — the two are not on a shared axis.

These weren't dashboard numbers. Every booking landed in the team's live calendar as a named intro-call event with two participants and a fixed agenda — shown at the top of this case study.

Two campaigns, two different jobs

CampaignPeriodLeadsAppts
Primary Lead Gen (CBO, US)Launch384
Secondary Lead Campaign (US)Launch435
Primary Lead Gen (CBO, US)Scaled2410
Secondary Lead Campaign (US)Scaled254
Qualified totalBoth periods13023
All bookingsBoth periods13046

Scaled-period appointments are qualified only — 37 booked, 14 passed qualification. The launch period had no qualification filter in place.

This produced the account's clearest strategic signal: in the scaled period the CBO campaign delivered fewer leads but more than twice the qualified appointments — 10 against 4. Volume and quality were coming from different places, and that now drives how budget gets split.

We built a filter, not just a funnel

The most valuable thing shipped in the second period wasn't an ad. It was a qualification layer that scores every inbound lead and routes it automatically, before it reaches a salesperson.

14 Accepted & routed to sales Above the 20-employee threshold where the client's model works
23 Correctly filtered out Almost all below the minimum viable company size
Read this number the right way

Twenty-three rejections isn't twenty-three failures — it's twenty-three hours of senior sales time that never got wasted.

No ad platform can stop unqualified people entering a funnel. What separates a working account from a leaky one is what happens in the ninety seconds after they do.

Beyond paid social

Meta was the primary engine, but never the only one.

LinkedIn Advertising

Impressions351,808
Clicks659
Campaigns6
Ad sets / Ads16 / 30

Lead gen and accelerator groups clearly beat video views on click-through — so budget was consolidated instead of keeping three formats limping along.

Manual Prospecting

Tracked events1,264
Profiles viewed835
Connections sent312

Named decision-makers only — warming the exact same audience paid social was already reaching.

Outbound Email

Contacted629
Opened288
Clicked83
Open rate41% → 47%

The sequence got stronger as it scaled rather than decaying — the opposite of what usually happens when outbound volume goes up.

LinkedIn Campaign Manager showing campaign-level performance
Figure 4. Campaign-level performance across the LinkedIn account. Account name, identifiers and all spend columns redacted.

The stand-out figure sits at the bottom of the outbound funnel. In the scaled period, 85% of everyone who clicked actually landed and stayed on the page — the clearest signal that the message and the landing page were finally saying the same thing. Open rate rose from 41% to 47% while volume quadrupled.

The work nobody puts in a case study

A meaningful share of the second period went into finding out why good traffic wasn't converting. Six issues found. Three fixed.

Fixed

Booking page broken on mobile

The calendar widget sat below the fold on mobile — where most ad traffic lands. A large share of visitors never saw a calendar at all.

Fixed

CAPTCHA before the booking step

Leads who had already decided to book were asked to prove they were human, and abandoned at the final click. Removed.

Fixed

Calendar availability was lying

Showing as little as one available hour per day while the sales team had far more open. Fixed through a sync rebuild with sales.

Open

No appointment reminders

Nothing automated between booking and the call itself. Documented in the client report.

Open

No real-time lead notifications

Sales isn't alerted the moment a qualified lead lands, so speed-to-lead is left to chance.

Open

No owner for marketing-tech issues

The real bottleneck. Issues fixable in a day currently take a week — and closing that gap costs nothing in media.

The headroom

Projected next period

40–60% more leads from the same budget. Two concepts are proven at volume, two more at efficiency and under-funded, four are cut. Moving spend out of the cut concepts into the winners — plus closing the three open operational blockers — is a compounding gain, not an incremental one. No increase in media investment required.

What we'd tell the next client

Client name, personnel, logos and commercial figures withheld or redacted under confidentiality. Reporting periods: 29 January – 25 February 2026 and 1 – 27 March 2026.

Free 2-Hour Digital Diagnostic

Ready to see similar outcomes in your account?

Book a 30-minute call — we'll cover your current stack, the gaps, and what a first 90 days would look like.

Book a Call →