What are growth-driven design templates?

The Deep Dive resource center on growthdrivendesign.com is the only place the method publishes files, and its list runs twelve items [1]. Four are Google Slides canvases (Goals Document, Persona Template, Fundamental Assumptions Template, Journey Mapping Template), three are Google Sheets (GDD Service Implementation Process, GDD Dashboard, Rainbow Spreadsheet), one is a Google Doc (Action Item Document), and the remaining four are a hierarchy cheat sheet JPG, a "What is GDD" slide deck in PDF and PowerPoint, a methodology graphic archive, and a logo archive [1]. None of them is a page layout, a theme, or code. Each link opens the file directly, with no form in front of it [1][2][3].

The method's stage pages name no CMS and no template either. The how-it-works page lists Strategy at "10-14 Days", Launch Pad at "60-90 Days" and Continuous Improvement at "14-Day Sprints" [9], and the continuous improvement page sets out a four-step cycle of plan, build, learn and transfer in which "Every plan starts with a focus metric that you want to improve" [10]. The documents exist to carry that cycle, and none of them touches the pages themselves.

The dashboard, the goals deck and the action item card all carry a 2018 Growth-Driven Design copyright and the line "Built by Luke Summerfield", who created the method at HubSpot [2][3][7]. Two of the canvases credit a second origin on the slide itself, "Inspired by Ash Mauyra's Lean Startup Canvas", misspelling Ash Maurya's name in the original [4][5]. The site hosting the files is a HubSpot property: the footer on the how-it-works page reads "2017 Growth-Driven Design (dba of HubSpot). Made by the Lean Labs GDD Team", while the resource center footer credits Media Junction [1][9]. Lean Labs is the contributor on this page, which is why that footer is disclosed here as well as in the perspective section below.

The GDD Dashboard holds both, on two tabs that answer different questions [2]. The Scorecard tab is the measurement side. Under each KPI metric it stacks four rows, Goal, Stretch, Actual and "Difference on goal", laid out across twelve months and four quarter columns, with a cell beside each metric for a link to the report in your analytics tool [2]. It records whether a number moved against a target you set in advance. It does not say what to build next.

The "Wishlist + Current Sprint" tab is the ranked backlog [2]. Its columns run Action Item Name, KPI Metric, Impact, Effort, Owner, Start, Stop, Prediction, Actual, Outcome and Notes / Learning, and Effort splits into Total, Strategist, UX, Developer and Other [2]. Rows sit in named bands: Current Sprint, High Priority, Medium Priority, Low Priority, then a Completed Action Items archive with a block per month [2]. The example row shipped in the template reads "[web] Interactive Product Graphic on HubSpot.com", impact 13, effort 3 split one strategist and two developer, owner Luke, start 5/18/2015, stop 5/22/2015, prediction 16% CTR, actual 18% CTR, outcome Success [2].

Impact and effort are typed in by hand, since the template carries no scoring formula, no scale definition and no rule for what makes an item impact 13 rather than impact 5, so the ranking is a judgment the team has to defend on its own [2]. Agency write-ups of the method fill that gap with their own rules. An Australian HubSpot partner agency describes brainstorming "50-150+ ideas", cutting to the 20 percent that drives 80 percent of the impact, then running a must-have versus nice-to-have pass [13]; IMPACT's version of the method applies the same 80/20 cut and works it in monthly sprint cycles [14]. Both are agencies that sell growth-driven design, so treat those rules as practice, not as part of the published template.

The blank Action Item card has fourteen fields: Action Item Name, card number, Owner, Impact, Metrics, Funnel Focus, Total Effort, User Job to Be Done, Objective, Hypothesis, Experiment Design, Results, What We Learned, and Future Actions / Recommendations [3]. Below is one filled in for a pricing page. Every number in it is invented for this example and none of it is a measured result.

The assumed inputs

The pricing page took 4,000 sessions last quarter. Of those sessions, 2.0 percent ended in a demo request. The page shows three plan tiers and the top tier says "Contact us" where the other two show a price.

The card

Action Item Name: pricing page, replace "Contact us" on the top tier with a starting price.

Card #: 41. Owner: one strategist and one developer.

Impact: 8, on the same free-entry scale the template's own example uses at 13 [2].

Metrics: demo requests from the pricing page as a share of pricing page sessions.

Funnel Focus: decision.

Total Effort: 3, split one strategist and two developer, matching the effort breakdown the dashboard asks for [2].

User Job to Be Done: "As a buyer comparing three vendors, when I am building a shortlist for my manager, I want a number I can put in the budget line so I can keep this vendor in the running." The wording follows the four-blank sentence the card prints [3].

Objective: find out whether a visible starting price on the top tier increases demo requests from the pricing page.

Hypothesis: buyers drop out at the top tier because "Contact us" gives them nothing to put in a budget. A starting price lifts demo requests from 2.0 percent to 2.6 percent of pricing page sessions, a rise of 0.6 percentage points, which at 4,000 sessions a quarter is 24 extra requests.

Experiment Design: run the change as a single 14-day sprint, the sprint length the method specifies, and compare against the previous quarter's rate [9].

The two assumptions doing all the work

The first is that 2.0 percent is a stable baseline. One quarter of traffic on one page may not establish that, and the card has no field that forces anyone to check.

The second is that price visibility, and not something else changing on the page or in the market that fortnight, causes any difference. The arithmetic makes that assumption hard to test at this volume: 4,000 sessions a quarter is about 615 sessions in a 14-day window, 2.0 percent of 615 is roughly 12 demo requests, and the predicted 0.6 point lift adds about four. A four-request difference on a base of twelve sits inside ordinary week-to-week variation, so a single sprint here will not settle the question.

One agency write-up states the hypothesis formula for this method as "Expected Impact + Effort Required + Metrics Measured + Definition of Complete = Hypothesis Statement" [13]. The official card covers the first three in its header row and leaves the fourth to the dashboard, where the Outcome cell closes the item and the Stop date sets the deadline [2][3]. Filling the card in is still worth the twenty minutes, because the prediction is written before anyone sees the number, which is what stops a team from deciding afterward that it expected whatever happened.

Page templates come from the platform, and the method has no opinion about them. HubSpot defines a theme as "a portable and contained collection of developer assets designed to enable a flexible content editing experience", holding templates, modules and style fields, with two JSON files at the root, and it scopes the word portable to movement between HubSpot environments and accounts [15]. WordPress block themes are "built entirely out of blocks" and store their templates as .html files, against .php in classic themes [16]. Neither platform ships anything branded growth-driven design, and the method's pages require no particular theme [1][9].

What a theme does change is the price of a wish-list item. An improvement is editor work when the layout and module already exist in the theme and developer work when they do not, which moves items between the Strategist and Developer effort columns the dashboard asks teams to fill in [2][15]. That is a build-cost question about your platform, not a question the method answers.

Themes sold in a marketplace are vendor material, and their listings publish template and module counts, which describe what is in the package and establish nothing about how pages built from it perform. This page names no preferred theme.

A template supplies a starting structure, and none of the four limits below is fixed by filling the file in more carefully.

The files are dated, and the method's own instructions tell you to change them

The dashboard is v1.0 with a 2018 copyright, its Scorecard month grid is labeled 2017, the goals slide asks for a 2016 goal, and the assumptions canvas is still version 0.8 [2][4][7]. The dashboard's Instructions tab says "Take your time to customize this dashboard to work best for your process and clients" and adds that "you don't have to use this dashboard" because "you can simply get ideas from it to add to your current reporting system" [2]. The service implementation sheet carries a similar step telling agencies to check that pricing still matches the process [1].

A filled-in canvas records a belief, not a finding

The Fundamental Assumptions Template sorts its blocks from "RISKIEST" to "LESS RISKY" and prints "QUESTIONS TO ANSWER" under the high-risk and low-risk website columns, which is a prompt for research, not a substitute for it [4]. Nothing in the file tests an entry. The Persona Template will accept whatever a room believes about a buyer's 6AM, because the day-in-the-life grid is blank cells [6].

The ranking is unscored

Impact and Effort on the wish list tab are free-entry numbers with no formula and no published scale [2]. Two teams can rank the same backlog in opposite orders and both be following the template correctly, which is why the ordering rule has to be written down somewhere the template does not cover.

The launch pad rule is narrower than it usually gets quoted

The method's launch pad page says sprints focus on "the 3-5 pages that can drive the greatest impact" and, in the same sentence, tells you to "put a 'fresh coat of paint' on the rest of the site while you migrate it into your new website" [11]. Three to five is the sprint focus, not the size of the site you launch, and the stated window is a 60 to 90 day build against three to six months for a traditional redesign [11].

The Slides canvases and the dashboard tabs are two copies of the same artifact

The GDD Dashboard carries tabs named Goals, Personas, Fundamental Assumptions and Journey Map [2], and four standalone Google Slides files on the same resource page carry the same four artifacts [1][4][5][6][7]. The dashboard version is a cell grid built for weekly editing; the Slides version is built to present, and the assumptions canvas appears in both the Fundamental Assumptions and Journey Mapping decks [4][5]. The method does not say which copy governs when they disagree, so a team that fills in both ends up maintaining two versions of its persona. Pick one home per artifact before anyone starts typing.

The free certification is training, not a template pack

The certification page describes the course as free, hosted by HubSpot with Luke Summerfield teaching, and lists eight modules including "Creating your GDD Tool Stack" [12]. Searches for growth-driven design templates often land there, and the course explains how the artifacts are used without being the place they are published. The files sit on the Deep Dive resource center and open without a course login [1][2][3].