Conversion rate optimisation built on evidence, not opinions about button colours: session recordings, funnel analysis and surveys show where buyers hesitate and why. Then experiments where traffic can power a reliable test, or direct fixes where it can't.
PixelCrayons' CRO service turns existing traffic into more revenue: session recordings, funnel analysis and surveys feed a ranked backlog of hypotheses, specific changes the evidence says should lift conversion, each tested where traffic allows. Where traffic can power a reliable A/B test we run one; where it can't, we say so and ship evidence-led UX fixes instead of fake statistics. Every experiment is reported win, lose or inconclusive, tied to revenue. Direct for brands, white-label for agencies.
Session recordings watched, funnels analysed step by step, and on-site surveys asking buyers what nearly stopped them. Then a heuristic review of the buying path by a senior analyst: form length, price surprises, proof placement, mobile friction. Every hypothesis we test starts here, in what your visitors actually do, not in a best-practice listicle.
First
02
A scored hypothesis backlog
Every candidate change written up with its evidence, expected impact and effort, then ranked. You see the whole queue and the reasoning behind the order. That way testing follows a plan rather than whoever argued loudest in the last meeting.
Maintained
03
Honest experiments
A/B tests run only where your traffic can reach a reliable verdict (statistical significance) in a sensible window, with the maths shown. Results reported as win, lose or inconclusive. Losing tests teach us where buyers surprise us. Pretending otherwise just corrupts the backlog.
Per test
04
Evidence-led fixes when tests can't run
Smaller sites deserve the truth, not theatre: below a certain traffic level an A/B test is a coin toss with a dashboard. There, the research still runs. The clearest problems it finds are fixed directly, measured before-and-after, without dressing the result up as an experiment.
As found
05
Design, build and reporting in-house
Winning variants are designed and implemented by our own designers and developers, so there's no handover gap where results go to die. A monthly report ties every test and fix to revenue.
Monthly
30
First 30 days
Evidence to verdicts, one test at a time.
01
Days 1 to 2
Instrument & baseline
Analytics audited, recordings and funnel tracking switched on, and a revenue baseline agreed. A lift you can't measure against a baseline is a story, not a result.
02
Week 1
The research week
Recordings reviewed, funnels analysed, surveys launched, heuristic website review done. The output is evidence: where buyers hesitate, backtrack and leave, and the likeliest reasons why.
03
Weeks 2 to 4
Backlog & first moves
The hypothesis backlog scored and agreed. The first real test launches where traffic allows. The clearest friction fixes ship directly where it doesn't.
04
Month 2+
Test, learn, compound
The cadence holds: run, read out, ship winners, log losers, re-score the backlog. Each verdict sharpens the next hypothesis. That's how CRO compounds.
Inside Prism
Your engagement, week to week, in one workspace.
Where the engagement includes it, marketing runs in a Prism workspace you log into: requests, approvals, tasks and the weekly review, with the cost of each channel beside the leads it produced.
Each decision recorded, with the expectation attached.
03
Cost per qualified lead
Fully loaded. Unmeasured spend shows as unknown, never zero.
How we report it, in Prism
The experiment ledger, including the ones that lost.
What you receive in month one
Conversion work is reported as an experiment ledger with every verdict recorded. In the first month you receive the analytics audit showing where the funnel data can and cannot be trusted, the research notes from session recordings and user interviews, and the hypothesis backlog ranked by evidence. Each test entry records its sample, its duration and its result, whether that was a win, a loss or inconclusive.
01Analytics audit: funnel events verified, gaps and double counting named
02Research notes: recordings, heatmaps and interviews, summarised with evidence
03Hypothesis backlog: ranked by evidence and effort, with the page each targets
04Experiment ledger: sample, duration, verdict, and what shipped as a result
05Weekly note in Prism: tests running, tests concluded, next tests queued
CRO multiplies what the system brings in. Pair it with PPC, where each conversion point makes the budget buy more, and web design, when a better page is the fix. Email and lifecycle keeps the buyers a better site wins.
Before you sign
Our commitments
Each one says where it applies and links to its full terms.
Response and start times
We typically respond within four business hours. Eligible engagements can typically start within two to three business days after scope, payment and required access are confirmed.
Business hours are Monday to Friday, 9:00 am to 6:00 pm IST, excluding published company holidays. Typical expectations, not guaranteed service levels.
Before work starts we agree how often you get updates, what the reports cover and who sends them. You know who coordinates your engagement, who owns the work and how to escalate an issue. For hourly and retainer work, time is recorded in Workstatus, so you can see the time records behind your invoice.
Who owns the ad accounts, the data and the work?
Your materials, data and accounts remain yours, and we prefer to work in advertising and analytics accounts you own, with access delegated to our team. Agreed ownership rights in bespoke deliverables transfer under the engagement terms and applicable payment; third-party software and assets keep their licences. We do not withhold customer-owned data or accounts over a payment dispute.
What if we want to stop?
Cancellation, notice and any renewal terms are written into your proposal or statement of work before you accept it. If an engagement ends early, we hand over completed, paid-for work and list any unfinished work separately. For an eligible specialist trial, you can stop within the trial window and pay nothing for eligible trial hours.
Can you guarantee first-page rankings?
No. Search engines decide rankings, and we do not guarantee rankings, AI-search citations, leads or revenue. We agree the work, how progress will be measured and how often we report it, and we explain what the measures show and what we will change next.
No. Treat any agency that does with suspicion: a guaranteed test result is a contradiction in terms. Conversion depends on your offer, your prices and your market as much as on the site itself. What we commit to contractually is the work: research delivered, a scored backlog maintained, experiments run properly and reported win, lose or inconclusive, winners implemented. The record we ask to be judged on is the aggregate one, not a promise about your next test.
Then we'll tell you, before you pay for testing theatre. Below a certain volume of conversions, an A/B test can run for months and still tell you nothing. Agencies that run them anyway are selling dashboards, not answers. For smaller sites the research still works: recordings, surveys and heuristic review find friction just as well. The clearest problems get fixed directly and measured before-and-after. When your traffic grows into proper testing, the backlog is already scored and waiting. That order of operations matters: research is useful at any size, testing only above a threshold. A CRO engagement worth paying for knows which side of it you're on.
Four strands, run together. Session recordings show where real visitors hesitate, rage-click and abandon. Funnel analysis shows which step of the journey loses the most people, segment by segment and device by device. On-site surveys ask buyers and near-buyers what almost stopped them, in their own words. And a heuristic review checks the buying path against the failures we see most: forms that ask too much too soon, shipping costs that appear at the last step, proof buried below the fold. Each hypothesis in the backlog cites which of the four produced it.
As many as your traffic supports. That might be four concurrent experiments on a high-volume store or one carefully chosen test on a leaner site. We'd rather run fewer tests with decisive readouts than pad a report with underpowered ones that read as noise. The cadence is agreed upfront from your actual numbers. Evidence-led fixes ship alongside whatever the testing calendar can carry. You'll always know which mode a given change is in, experiment or fix, and why. Mixing the two up is how CRO reporting turns into fiction.
Engagements are scoped to your traffic and stack: a research audit with direct fixes for a leaner site, or a rolling research-and-testing retainer for a high-volume one. Every proposal itemises what ships each month: research delivered, hypotheses added, tests run, fixes implemented, reporting. Recording and testing tool subscriptions are quoted separately, never hidden in the fee. Once we have a complete brief, we confirm the date you will receive your written, priced scope. The initial conversion audit is free.
Yes. Research, testing and implementation all ship under partner agencies' brands: your templates, your client calls, our analysts, designers and developers. It's NDA-backed. We do not approach the clients you introduce to us for direct business. That protection runs through the work and for twelve months after the last assignment for that client ends, and it is written into the agreement. We have served 500+ agencies.
Yes, as part of the same research-led process. Recordings, funnel analysis and surveys show where visitors hesitate on a landing page, and the backlog ranks what to change. Where traffic allows we test the change; where it doesn't, we fix it directly and measure before and after. Our own designers and developers build the winning version.
Send us your site. We'll come back with a free conversion review: where the funnel leaks, what the evidence suggests, and whether your traffic can power reliable tests, priced and itemised. Keep it either way.
