Which tools help teams plan, test, and collaborate on growth-driven design?

Assume a marketing site drawing 12,000 sessions a month, one page under test taking 4,000 of them, and a team of four. Those three inputs are chosen to make the arithmetic visible, not measured from any site, so substitute your own numbers.

The HubSpot route: $495 a month

Content Hub Professional lists at $450 a month billed annually and includes three core seats, with additional seats at $45 a month, so a four-person team pays $495 [13]. That subscription is the line that buys the test step, since page A/B testing is documented as Content Hub Professional and Enterprise only [14]. Contentsquare's free plan carries the research step, because 12,000 sessions sits well under its 200,000-session ceiling [3][4]. Its 10,000-replay cap binds first: if one session produces one replay, about 2,000 sessions a month go unrecorded [4]. Google Analytics adds nothing to the bill [11] and Jira Free holds the sprint record for four users, inside its 10-user limit [20]. The total is $495 a month, or $5,940 a year, which is about $228 for each of the 26 fourteen-day sprints a year allows.

The WordPress route: not computable from vendor pages

On a WordPress site Clarity and Google Analytics still cost nothing [9][11], and the line a sprint cannot skip is the experiment tool, which is the line the vendors leave blank. Nelio's free tier allows 500 tested page views a month [15], which is 12.5% of the 4,000 monthly views on the page under test, so one test covering every visitor to that page would take eight months on the free quota. Its Basic tier raises the quota to 5,000 tested page views, enough for a single sprint's traffic, and no price for it appears in the pricing page text [16]. Optimizely and VWO publish no prices either [17][18]. Producing a total for this route requires a sales conversation, which is information in itself when you are comparing two routes on paper.

What neither total includes

Hosting, domain registration, developer hours for anything a theme's modules cannot express, and the salary of whoever runs the sprint sit outside both figures.

Contentsquare's free plan runs out on quota and Clarity's runs out on time, and the two failures land in different parts of a sprint. Contentsquare's 200,000-session ceiling sits far above most marketing sites, so the caps a team meets first are 10,000 replays and 100 surveys a month [4]. Clarity has no traffic limit at all, and its playback data is deleted after 30 days while click data and sessions you label or favorite are held for 9 months [9][10].

The 30-day window changes one habit, because a sprint that ships a change, waits two weeks for data, then reviews the evidence in the retrospective is close to the edge of it, and a decision revisited six weeks later has the heatmap and no recording. Labeling the sessions that informed the hypothesis during the sprint keeps them for 9 months [10].

Google Analytics has the same shape of limit on a longer clock. Standard properties keep user-level and event-level data for 2 or 14 months, large and extra-large properties are capped at 2 months, and expired data is deleted rather than archived [12]. A year-over-year read on the focus metric therefore depends on the 14-month setting being chosen before the year passes, not after.

An exit rate establishes that some share of sessions ended on a page; a replay names the field where the last click landed. That second observation is what a hypothesis gets written from, and on a free plan it is the piece of evidence with the shortest retention [10].

No source cited on this page compares these products by the outcomes they produce. Optimizely's web experimentation page advertises 446% platform ROI attributed to Forrester and 78% faster testing with agents, without a sample, a method, or a link to the underlying study on the page itself [17]. Customer counts and adoption figures fail the same test in the other direction: a number of installations establishes that a product sells, not that a site improves after installing it.

What experiment software does is narrower than proof: a statistical test sets a significance level, which NIST's engineering handbook defines as the risk of rejecting the null hypothesis when it is true, and carries a second risk of failing to reject a false null that the user does not choose [21]. The handbook frames the intent as determining whether there is enough evidence to reject a conjecture, and notes that failure to reject can mean insufficient data [21]. A platform reporting 95% confidence has held the first risk at 5%. It has not established that the variant is better, and it says nothing about whether the difference is worth the build.

That leaves task fit and plan terms as the criteria this evidence can support. Both are documented on the pages cited above: which step of the sprint a tool covers, and what its own documentation says it will not do at the price you would pay.

Two different things get asked under the heading of growth-driven design tools, and mixing them produces the wrong purchase.

A recommended tool and a required one

The tool list on growthdrivendesign.com is a blog post from January 2018 with a named author [2], and the stage documentation beside it sets out durations and steps without naming software [1]. A team running the method on Webflow with Clarity and a spreadsheet is running the method. Reading the post as a requirement also means inheriting its age, since four of its links now resolve to other companies [3][5][6][7].

A plan quota and a usable sample

A plan that covers your traffic is not a plan that settles your test. Nelio's free tier counts 500 tested page views a month [15] and Contentsquare's free plan counts 200,000 sessions [4], and both figures describe what the vendor will process, not what a comparison needs to separate two variants. HubSpot makes the gap visible by documenting the 50/50 split and the manual winner selection while giving no sample-size guidance anywhere in the article [14], which leaves the stopping decision with the person reading the dashboard.