---
title: "CRO agency or tool? How Australian teams decide | PostCliq"
description: "When an Australian marketing team needs a CRO agency, a tool or an in-house hire: a decision table, the traffic you need to test, and six questions to ask."
url: "https://postcliq.com/blog/cro-agency-or-tool"
---

# Do you need a CRO agency? How Australian marketing teams decide

You need a CRO agency when you have the traffic to test, fixes your team cannot build, or a deadline you cannot meet alone. Here is the decision table, the traffic arithmetic and the questions to ask before you sign.

By [Nirmal Gyanwali](https://postcliq.com/blog/author/nirmal-gyanwali) Updated 30 September 2026 10 min read

On this page

1. The short answer
2. What does a CRO agency actually do?
3. Tool, agency or in-house: which fits your situation?
4. How much traffic do you need before a CRO agency makes sense?
5. What does each option cost an Australian team?
6. When should a Melbourne or Sydney team hire a CRO agency?
7. What should you ask a CRO agency before you sign?
8. Where do PostCliq and WP Creative fit?
9. What to do next
10. Frequently asked questions

## The short answer

Key takeaways

- A CRO agency (a conversion rate optimisation agency) diagnoses why visitors do not enquire or buy, designs fixes and, at larger scale, runs A/B tests to prove which changes work.
- Testing needs volume: detecting a 20% lift on a 2% conversion rate takes about 39,000 sessions, which is about 20 months of traffic for a page with 2,000 visits a month.
- Below that volume, a tool-led audit plus a developer fixes most of what an agency would find, at a fraction of the cost.
- Hire an agency when you have the traffic for a testing programme, a rebuild or complex build work, or a fixed deadline such as EOFY or Click Frenzy.
- Whoever you choose, ask for the minimum traffic they need, how many tests they expect to win, and who owns the data and the test history.

Only [12% of experiments produce a statistically significant improvement on the primary metric](https://www.optimizely.com/insights/top-10-takeaways-from-running-127000-experiments/) , according to Optimizely's analysis of more than 127,000 tests. A CRO agency earns its fee when it can run enough experiments for that hit rate to pay off. If your site cannot, you are paying for a diagnosis you could get from a tool, and for testing you cannot statistically support. The decision table below sorts the situations.

## What does a CRO agency actually do?

A CRO agency finds where visitors drop out between arriving and enquiring, proposes changes, and proves which changes work. The work usually has three parts: research (analytics, recordings, surveys, an audit), design and build of changes, and experimentation to measure the effect. Some agencies do all three. Many do mainly the first.

It helps to separate the three, because they cost different amounts and need different conditions.

- **Diagnosis** (the CRO audit). Anyone with access and a method can do it. It needs no minimum traffic. The [CRO audit guide](https://postcliq.com/blog/cro-audit) shows how to run one in an hour.
- **Fixing.** Building the changes. This needs design and development capacity, which is where many in-house teams are short.
- **Testing.** A/B testing proves cause and effect, but only with enough traffic and enough conversions. This is the part an agency is most clearly needed for, and the part most small sites cannot use.

When an agency proposal bundles all three into one retainer, ask which part you are really buying.

## Tool, agency or in-house: which fits your situation?

Choose by traffic, capacity and deadline, not by how much the problem worries you. A tool is enough when the job is finding and ranking fixes. An agency is worth it when you need a testing programme, heavy build work or speed. In-house works when you already have an analyst and a developer who can give it regular time.

| Situation | Tool-led audit | CRO agency | In-house team |
| --- | --- | --- | --- |
| Under 10,000 monthly sessions, one marketer, no developer on staff | ✓ Start here | ◐ A one-off audit at most | ✕ Not enough to keep a hire busy |
| Obvious basics are broken (forms, speed, tracking, mobile layout) | ✓ Finds and ranks them | ◐ Useful only for build hands | ◐ If a developer is available |
| 40,000+ monthly sessions to key pages and a goal of continuous testing | ◐ Feeds the backlog | ✓ Testing programme, research, velocity | ◐ Needs an analyst, a developer and a testing tool |
| The findings point to a rebuild or redesign | ◐ Use it to write the brief | ✓ Or a web agency with CRO skills | ✕ Rarely the right team for a rebuild |
| A fixed deadline (EOFY, Click Frenzy, Black Friday) | ◐ Fast diagnosis | ✓ Capacity on a deadline | ◐ Only if the team is free |
| You already employ an analyst and a developer | ✓ Speeds up diagnosis | ◐ Specialist input only | ✓ Build the capability |
| The board wants proof that changes caused the lift | ✕ An audit cannot prove causation | ✓ If traffic supports testing | ◐ If traffic supports testing |

Source: PostCliq analysis. Session thresholds are working guides derived from the sample-size arithmetic below, not published cut-offs.

## How much traffic do you need before a CRO agency makes sense?

You need enough traffic to run tests that can reach a result in weeks, not years. As a working rule, a page needs tens of thousands of monthly sessions before a testing programme is realistic. Below that, spend on diagnosis and fixes, and measure before and after instead of running split tests.

Here is the arithmetic, so you can run it on your own numbers.

A common approximation for A/B test sample size (Lehr's rule, for 80% power at 5% significance) is: sessions per variant ≈ 16 × p × (1 − p) ÷ d², where p is your conversion rate and d is the absolute change you want to detect.

Take a service page converting at 2%. You want to detect a 20% relative lift, from 2.0% to 2.4%, so d is 0.004.

- 16 × 0.02 × 0.98 = 0.3136
- 0.004 × 0.004 = 0.000016
- 0.3136 ÷ 0.000016 = 19,600 sessions per variant
- Two variants: about 39,200 sessions in total

At 2,000 sessions a month, that test takes about 20 months. At 20,000 sessions a month, about two months. At 50,000, under a month.

Lehr's rule is a quick approximation. The exact two-proportion formula in our [A/B testing guide](https://postcliq.com/blog/ab-testing-for-marketing-teams) gives a slightly higher total, about 42,000 sessions, because it allows for the variant's higher rate. The conclusion is the same either way.

Months to detect a 20% lift on a 2% conversion rate

- 2,000 sessions a month 19.6months
- 5,000 sessions a month 7.8months
- 10,000 sessions a month 3.9months
- 20,000 sessions a month 2months
- 50,000 sessions a month 0.8months

Source: PostCliq analysis using Lehr's approximation (80% power, 5% significance, two variants)

And a 20% lift is a large effect. Smaller, more typical effects need far more traffic. Optimizely reports that [the median company runs 34 experiments a year](https://www.optimizely.com/insights/top-10-takeaways-from-running-127000-experiments/) , and that companies need about 200 a year to reach the top 10% for testing velocity. Those are Optimizely customers, mostly enterprise. A typical Australian mid-market service site cannot run 34 meaningful tests a year on one landing page. It can fix 34 things.

> Only 12% of experiments produce a statistically significant improvement on the primary metric.

Optimizely, analysis of 127,000 experiments

That is not an argument against testing. It is an argument against paying for a testing programme your traffic cannot support. Our guide to [when A/B testing pays for a marketing team](https://postcliq.com/blog/ab-testing-for-marketing-teams) goes further.

## What does each option cost an Australian team?

The honest comparison is time and money per fix shipped, not the headline fee. A tool costs little and ships nothing on its own. An agency ships fixes but costs a retainer. An in-house hire costs a salary before a single fix ships. Price all three against the enquiries they can realistically add.

We do not publish CRO agency price ranges, because we could not find any drawn from a disclosed sample. Agencies publish their own prices, and the ranges on comparison sites are rarely sourced. Ask for a fixed-fee quote for a defined scope instead.

The in-house number is easier to anchor. SEEK shows an average of [AUD 97,500 for digital marketing specialist roles advertised in Melbourne](https://au.seek.com/career-advice/role/digital-marketing-specialist/salary) in September 2026, based on salary ranges employers disclosed in job ads. That is a generalist marketing role, not a specialist CRO analyst, and it excludes super, tools and a developer. It still tells you a hire is a six-figure decision.

Two pressures make the decision sharper in 2026:

- **Budgets are flat.** Gartner's 2026 CMO Spend Survey found marketing budgets at [7.8% of company revenue, with 56% of CMOs saying they lack the budget to deliver their strategy](https://www.gartner.com/en/newsroom/press-releases/2026-05-11-gartner-2026-cmo-spend-survey-finds-cmos-allocate-15-point-3-percent-of-marketing-budgets-to-ai-but-only-30-percent-are-ready-to-scale-ai-capabilities) . The sample is large companies in North America and Europe, not Australian mid-market, but the direction matches what most marketing managers report.
- **Traffic keeps getting dearer.** Australian internet advertising grew 14% to [AUD 19.8 billion in FY26, with search at AUD 8.6 billion](https://www.iabaustralia.com.au/news/internet-advertising-records-strongest-growth-in-four-years-reaching-19-8-billion-in-fy26/) (IAB Australia). That is total spend, not unit price, but more money chasing the same buyers raises the value of converting the ones who already arrive.

Why teams under-use what they already have

**33%** of martech stack capability marketers actually use [Gartner Marketing Technology Survey, 2023](https://www.gartner.com/en/newsroom/press-releases/2023-08-23-gartner-survey-finds-63-percent-of-marketing-leaders-plan-to-invest-in-generative-ai-in-the-next-24-months)

**4.1%** of all websites run Microsoft Clarity, which is free [W3Techs, September 2026](https://w3techs.com/technologies/details/ta-microsoftclarity)

**34** experiments a year at the median company [Optimizely (customer base, mostly enterprise)](https://www.optimizely.com/insights/top-10-takeaways-from-running-127000-experiments/)

The Gartner utilisation figure is three years old, and it is the most recent Gartner has published. It points at something real: many teams already own tools that could answer their CRO questions and do not use them. Before hiring anyone, check what your GA4 and Clarity accounts can already tell you. The [best CRO tools guide](https://postcliq.com/blog/best-cro-tools) maps each tool to the question it answers.

## When should a Melbourne or Sydney team hire a CRO agency?

Hire when the work exceeds your team's capacity or your traffic supports testing, and when you can name the outcome you are buying. The scenarios below are illustrative, not real clients, but they are the situations that decide it for most Australian marketing teams.

**A Melbourne professional services firm, 4,000 sessions a month, one marketing manager.** No testing programme is viable: a 20% lift test would take around ten months. Run a tool-led audit, fix the three biggest problems with the web developer you already use, and measure enquiries before and after. A "CRO agency Melbourne" search will return agencies happy to take a retainer. The honest answer is that the firm does not need one yet.

**A Sydney ecommerce brand, 90,000 sessions a month, EOFY and Click Frenzy peaks.** This is where an agency pays. There is enough traffic to test checkout and product page changes in weeks, and a hard seasonal deadline the in-house team cannot meet alone. Buy a testing programme with a named test velocity and a research phase before the first test.

**A Melbourne B2B software company, 25,000 sessions a month, planning a rebuild.** The priority is not CRO testing. It is making sure the rebuild does not repeat the current site's conversion problems. Audit first, then write those findings into the rebuild brief. [The questions to ask before you rebuild your website](https://postcliq.com/blog/questions-before-you-rebuild-your-website) belong in that brief.

**A Sydney multi-location clinic group with an analyst and a developer on staff.** Build the capability in-house. Bring in a specialist for a short engagement to set up the testing process and review the first tests, then run it yourselves.

The same logic applies to conversion rate optimisation anywhere in Australia: Brisbane, Perth and Adelaide teams face the same traffic arithmetic.

## What should you ask a CRO agency before you sign?

Ask questions that force specific, checkable answers: what traffic they need, how many tests they expect to win, what they will ship in the first 30 days, and who owns the data. Vague answers to these are the clearest warning you will get. Put the answers in the contract.

1. **What is the minimum traffic you need to run tests on our site?** A credible agency will do the sample-size arithmetic with you. If they promise testing on a page with 3,000 visits a month, they are selling activity.
2. **What share of your tests win?** Anything far above Optimizely's 12% deserves scrutiny. It may mean they only count favourable results or stop tests early.
3. **What will you ship in the first 30 days?** You want named fixes, not a discovery phase with no end date.
4. **How will you measure success?** Enquiries or revenue, reconciled against your CRM, not engagement or "insights delivered".
5. **Who owns the test history, recordings and analytics access if we leave?** It should be you, in your own accounts.
6. **Who does the build?** If they hand designs to your developer, your developer's capacity is the real limit.

Brief them properly, too. A clear brief with evidence gets better proposals and fewer surprises. [How to brief a website fix](https://postcliq.com/blog/how-to-brief-a-website-fix) covers the format, and the [website fix brief template](https://postcliq.com/resources/website-fix-brief-template) is ready to fill in.

Run the audit before the pitch

Bring your own audit to the first agency meeting. It changes the conversation from "what do you think is wrong?" to "here is what is wrong, how would you fix it and what would it cost?" That is a much easier proposal to compare.

## Where do PostCliq and WP Creative fit?

PostCliq is a free website audit that covers the diagnosis step. WP Creative, the Australian website growth agency that built PostCliq, is one option for the fixing step if you want outside hands. You can use the first without the second, and nothing in the audit signs you up for the agency.

We have an interest here, so read this section with that in mind.

**What PostCliq does.** It reads up to 50 public pages the way a careful visitor would and checks seven areas: speed and page experience (Lighthouse lab data and Chrome UX Report field data), forms and calls to action, content clarity, search access, accessibility, trust signals and AI readiness. It gives you a private report with evidence and three quick wins to make first, and you can re-audit each site once a week to check the fixes held. The [CRO audit page](https://postcliq.com/cro-audit) sets out every check.

**What PostCliq does not do.** It does not run A/B tests, fix your site, log in, submit forms or guarantee results. It gives an estimated opportunity range from sourced traffic data or your own numbers, never a measured revenue loss. If your situation in the table above says "testing programme", you need a specialist, and PostCliq will not replace one.

**WP Creative.** WP Creative says it has been operating for more than 10 years, has built more than 1,200 websites and has more than 100 five-star Google reviews. If the audit points to work your team cannot do, it can take the report as the brief. So can any other agency. The [CRO audit for Australian businesses](https://postcliq.com/cro-audit-australia) explains the hand-off.

## What to do next

1. Write down your monthly sessions to your three most important pages and their conversion rate.
2. Run the sample-size arithmetic above. If a 20% lift test would take more than three months, skip testing for now.
3. Run a [free PostCliq audit](https://postcliq.com/#try-postcliq) or the one-hour [CRO audit method](https://postcliq.com/blog/cro-audit) to find and rank the fixes.
4. Check whether your team, or your current developer, can ship the top three fixes within a month.
5. If they cannot, or if the audit points to a rebuild or a testing programme, brief an agency using the six questions above.
6. Measure enquiries before and after each fix, reconciled against your CRM, whoever makes the change.

## Frequently asked questions

### What does a CRO agency do?

A CRO agency researches why visitors do not convert, designs and builds changes, and, where traffic allows, runs A/B tests to prove which changes work. Research usually includes analytics, session recordings, surveys and a site audit. The value is highest when the site has enough traffic for testing and the in-house team lacks design, development or analysis capacity.

### How much does a CRO agency cost in Australia?

We could not find a published price range based on a disclosed sample, so we do not quote one. Pricing usually takes three forms: a fixed-fee audit, a monthly retainer for fixes, or a testing programme priced by test volume. Ask for a fixed quote against a defined scope, and compare it with the salary cost of an in-house hire.

### Is CRO worth it for a small business?

Conversion work is worth it for most businesses. A full CRO agency usually is not, below tens of thousands of monthly sessions, because there is not enough traffic to run meaningful tests. A small business gets most of the value from a diagnosis, a handful of fixes and a before-and-after measurement of enquiries. Fix the basics first.

### How long does CRO take to show results?

Fixes to broken basics, such as a form that fails or a slow mobile page, can show in enquiries within weeks. A testing programme takes longer: each test needs enough sessions to reach significance, and Optimizely's data shows most tests do not win. Run the sample-size arithmetic in this article on your own traffic to see how many months one test needs.

### What is the difference between a CRO agency and a CRO tool?

A CRO tool collects or analyses data: analytics, recordings, heatmaps, audits or testing. A CRO agency supplies people who interpret that data, decide what to change, build it and test it. Tools are cheap and need someone to act on them. Agencies act, and cost more. Most teams need a tool first and an agency only at scale.

### Should I hire a CRO consultant or an agency?

A consultant suits a team that has build capacity and needs expert direction: an audit, a testing plan or a review of results. An agency suits a team that needs research, design, development and testing delivered together. If your developer is already busy, a consultant's recommendations will wait in a queue, so an agency with build capacity may be the better buy.

## Diagnose first. Free, in minutes.

PostCliq audits your website for what stops visitors becoming enquiries, so any agency brief starts with evidence.

Free. No credit card.

![Nirmal Gyanwali, founder and chief executive of WP Creative.](https://postcliq.com/authors/nirmal-gyanwali.jpg)

Written by

[Nirmal Gyanwali](https://postcliq.com/blog/author/nirmal-gyanwali)

Founder and CEO, WP Creative · Sydney, Australia

Nirmal Gyanwali has spent more than twenty years building websites, and most of that time watching them fail at the one job the business actually bought them for, which is turning a visitor into an enquiry. He founded WP Creative in Sydney in January 2016.

- [LinkedIn](https://www.linkedin.com/in/nirmalgyanwali/)
- [YouTube](https://www.youtube.com/@nirmalgyanwali)
- [Instagram](https://www.instagram.com/nirmalgyanwali)
- [TikTok](https://www.tiktok.com/@nirmalwpc)

## Keep reading

- [**How to write a website brief so the fix actually gets done**](https://postcliq.com/blog/how-to-brief-a-website-fix)
- [**Website redesign checklist: the questions to ask before you rebuild your website**](https://postcliq.com/blog/questions-before-you-rebuild-your-website)
- [**What makes a good website? Anyone can build a good-looking one now, and that is the problem**](https://postcliq.com/blog/good-looking-website-is-not-enough)

[More on Working with agencies and developers](https://postcliq.com/blog/topic/working-with-suppliers) · [All articles](https://postcliq.com/blog)