The most valuable outcome of the three was the concept that stopped. Learning inside five weeks, for a four-figure media spend, that a fixed hardware price will not convert is worth more than an engineering programme that discovers it later.
Back to all proofs
Three product concepts to a go, a kill and a price test in five weeks
Client
Date
$2B+ US connected-home technology manufacturer
12/2024
The challenge
A $2B+ US connected-home technology manufacturer holds a backlog of roughly two thousand product ideas and no consistent way to decide which earn engineering time. Ideas advanced on internal conviction, because nothing existed for putting a concept in front of a real buyer at a real price before committing to build it.
The solution
FifthRow ran three internal product ideas through one validation sequence end to end: market research and synthetic personas in, a priced concept on a live landing page with real paid traffic behind it out, and a weekly read on which propositions were actually converting. The output is a decision on each concept, not a report on all three.
5 weeks
from internal idea to live priced market test3 of 3
concepts resolved to a decisionConcepts validated3, from internal idea to live market test
Cycle time5 weeks, idea to live traffic
Pricing evidenceConsumer price points estimated per concept from personas
Decisions produced1 advanced, 1 closed, 1 taken to further price testing
Three ideas had sponsors but no evidence. Each was built out into a proposition specific enough to put a price on and to show a stranger, using synthetic personas to establish who would buy it and what they would pay.
Core Activities
- Expanded each sponsored idea into a full concept with a defined buyer
- Built synthetic personas for the customers each concept would have to win
- Ran simulated interviews to pressure-test desirability before any spend
- Estimated a consumer price point per concept from the persona evidence
- Generated further solution options inside the same problem spaces
- Sized the opportunity and tested feasibility on each concept in turn
The Results
Three concepts with a named buyer, a proposition and a defensible price attached, and two pricing calls already made by the owning teams before any traffic ran.
2
pricing decisions taken directly off the simulated persona evidenceA priced concept is still an assumption until someone is asked to act on it. Each concept went onto its own live landing page, with paid traffic pointed at it and a real sign-up as the thing being measured.
Core Activities
- Designed a landing page per concept, then built and published each one live
- Presented each concept at its estimated price, so the offer was the thing being tested
- Ran paid social campaigns against each page inside a fixed client media budget
- Reported per-concept performance back weekly, on one dashboard the teams read directly
- Optimised creative and targeting inside the campaign window rather than after it
The Results
Three propositions in front of real buyers at real prices inside five weeks, each producing its own conversion evidence rather than an internal opinion about how it would land.
3
product ideas taken from backlog to live paid trafficThe point of the tests was to make the three concepts decidable. Each one was read against its own conversion evidence and closed out with a call, including the call not to continue.
Core Activities
- Interpreted each concept's conversion data against its price and its audience
- Optimised cost per click, click-through rate and cost per sign-up across the window
- Took one concept to an approved A/B test on price rather than guessing at elasticity
- Confirmed one concept as good to go on the strength of its own results
- Recommended closing one concept where the fixed hardware price would not convert
- Benchmarked the approach against the consumer-research platforms the company already used
The Results
A decision on every concept: one advanced, one taken into price testing with approval, and one closed on the evidence, all inside the same five weeks and the same four-figure media budget.
$49.57 to $40.15
cost per sign-up, optimised inside the test windowWhat it proved
Three sponsored ideas went in. Each came out as a full concept with a named buyer, a persona-derived price and a live landing page carrying paid traffic. Across the four-week campaign window, cost per click fell from $1.03 to $0.50, click-through rate rose from 3.81% to 5.21%, and cost per sign-up came down from $49.57 to $40.15, all inside a $3,600 client media budget. Every concept ended with a decision: one confirmed good to go, one approved for A/B price testing, and one closed because the hardware price it was locked into would not convert.
What the company got was a way of resolving a concept rather than a way of describing one. The value is clearest in the concept that stopped: the team learned within weeks, for a four-figure media spend, that its price point would not work, rather than discovering it after an engineering commitment. Against a backlog of roughly two thousand ideas, having a repeatable route from internal idea to real-buyer evidence at a known price is what makes the backlog triageable at all.
Results
| Metric | Benchmark | Result | Improvement |
|---|---|---|---|
| Concept validation route | No mechanism for testing a concept on a real buyer before building it | Idea to live priced landing page in 5 weeks | A concept can be tested before it is committed to |
| Pricing evidence | Price set on internal judgement | Consumer price point estimated per concept from persona evidence | Price decided on buyer evidence |
| Decision yield | Concepts advanced on internal conviction | 1 advanced, 1 closed, 1 into price testing | Every concept resolved, including the one that stopped |
| Cost of finding out | An engineering commitment before market feedback | $3,600 of media across three concepts | A wrong price found for four figures, not a build |
| Conversion efficiency | No prior in-market conversion data on any concept | Cost per sign-up optimised from $49.57 to $40.15 | Sign-up cost cut by a fifth inside the window |
A landing page that does not name a price tests interest, not demand. Presenting each concept at its estimated price meant the sign-up measured willingness to buy rather than curiosity.
Synthetic personas produced a defensible price to go to market with, and live traffic said whether it held. The two work as a sequence: one makes the test possible, the other makes it true.
Creative and targeting were adjusted while the campaigns ran, not reviewed afterwards. Cost per sign-up fell by a fifth during the same four weeks, which is conversion evidence a single static test would never have produced.
One concept failed on a hardware price the company could not move. A test that identifies which constraint is binding has answered a more useful question than one that only reports whether people liked the idea.
Two thousand ideas cannot be argued through one at a time. What makes a backlog triageable is a repeatable path from internal idea to real-buyer evidence, cheap enough to run on a concept that might not survive it.
The intelligence function six business areas never had to staff
Client
$30B+ global electrification and automation group
Date
08/2026
596,505
analyst-hours equivalent, 150 FTEsFrom one-off research requests to a standing new-business intelligence practice
Client
$500M+ global speciality chemicals manufacturer
Date
08/2026
382,406
analyst-hours equivalent, 155 FTEsFor Enterprises
ProofSee all
From a standing start to an AI venture serving small businesses
From 72 patents to 4 commercialisation-ready technologies
Cutting a four-month research cycle to six weeks across five validation workstreams
From a weekly newsletter to a self-running innovation radar
A full year's validated-concept target, met in the first quarter
A utility innovation lab that answers its own questions in a day
Finding the next high-growth US metro in five days
FifthRow – The Agentic Consultancy
Copyright 2026 FifthRow Inc., Privacy policy