Skip to content
Faisal Mahmud TusherDigital marketing, SEO & analytics
with Faisal Tusher

Google Ads › Advanced

Capstone: A Google Ads Growth Strategy for an Office Furniture Seller

The closing project of the advanced Google Ads level. You write a year's growth strategy for an invented office furniture seller whose account treats every conversion as equal and whose reported return is inflated by brand traffic, covering the conversion signal plan, the test agenda, brand control, a scaling path with stop rules and the operating rhythm that holds it together.
Capstone projectAdvanced levelModule 3: Automation, brand control and multi-engine scale6 minUpdated

Brief

Swivelgable Office Furniture is an invented seller of desks, chairs and storage to small businesses and home workers. Home workers buy online; small businesses usually request a quote for a fit-out, which a sales adviser follows up by phone. Both flows land in the CRM. The account runs brand search, generic search and Performance Max, with every conversion action primary and unvalued, so an online chair order, a fit-out quote request and a brochure download count the same.

The owner wants to grow spend substantially next year. She suspects much of the reported return is brand traffic and that quote requests, which close into the largest orders, are undervalued by the way the account counts them. She has also been offered a search management platform and wants to know whether to buy it.

Inputs and constraints

Illustrative example — not a real result

Swivelgable Office Furniture: account over the last twelve months

CampaignBid strategySpendConversions reportedOnline ordersQuote requestsDownloads and sign-ups
Brand SearchMaximise conversions£12,0002,4001,500300600
Generic SearchTarget CPA £40£96,0003,6001,5007001,400
Performance MaxTarget CPA £25, no brand exclusion£72,0003,0002,100200700
  • Performance Max search term insights: brand name and variants carry about 35% of its conversion value.
  • Brand search volume index over two years: roughly flat.
  • Lost impression share to budget on Generic Search: 30%. Brand Search impression share has fallen this year.
Illustrative example — not a real result

Swivelgable Office Furniture: CRM and margins

Quote stageShare of quote requests reaching itTypical time after the request
Qualified by an adviser60%Two to five days
Quote sent45%One to two weeks
Won25%Three to eight weeks
  • Average won fit-out order: £6,000 at a 30% contribution margin.
  • Online orders: average £320; contribution margin chairs 35%, desks 25%, storage 20%.
  • The CRM stores email and phone on every quote request; it does not store the click identifier.
  • Budget ceiling for next year: £300,000, against £180,000 this year.

Constraints:

  • One in-house marketer runs the account; a CRM administrator and a developer are available a few days a month.
  • The consent banner gates advertising tags; uploads must respect the stored consent state.
  • No tests across the January buying peak, when a stalled change costs the most.
  • The account runs on Google Ads only.

Task

Write one strategy document with the sections below, in this order. State expected effects as direction with a way to check them; do not invent uplifts.

  1. Baseline. Score the account with the Google Ads account audit scorecard, gathering the evidence with the account audit checklist. Record the failed gates and say which findings the strategy must fix before anything else.
  2. Signal plan. Value online orders from margin by product line and quote stages from close rate and fit-out margin. Choose the primary action for each flow and demote the rest. Plan capture and upload through enhanced conversions for leads, given that the CRM holds email but no click identifier, with consent checks, an upload rhythm and a monthly reconciliation against the CRM. Use the CPA calculator and ROAS calculator to derive the break-even and target ROAS on the new values.
  3. Test agenda. Design a geo split for the move from cost targets to value-based bidding, and a budget increase incrementality test for the first step of scaling. For each, write the hypothesis in the experiment hypothesis template with matching, duration, learning exclusion, guardrails and a decision rule, and place both in the calendar around the January peak.
  4. Brand control. Exclude the brand from Performance Max, give brand search its own budget and target, and follow the Performance Max structure guide on the customer acquisition goal. Define the reporting split between brand, returning customers and new demand.
  5. Scaling path. Set the maximum marginal cost per incremental order and per won fit-out from margin, then lay out budget steps towards the ceiling, each gated by the test result and a stop rule, following the guide to marginal returns and forecasting. Use the marketing budget calculator for the scenario totals.
  6. Operating rhythm. Assign daily, weekly, monthly and quarterly work with owners from the paid search account management playbook, name the review points for each strategic change, and write the first decision log entries.
  7. Platform question. Answer the owner on the search management platform in a short paragraph.
  8. Summary. One page for the owner: what changes, in what order, how each change will be judged and what would make you stop.

Evaluation criteria

CriterionWhat a strong answer showsCommon gap
Signal planValues orders and quote stages from margin and close rate, with imports and validationKeeps optimising to all conversions equally
Test agendaDesigns geo tests for the bid strategy and budget changes with guardrailsRelies on before-and-after comparisons
Brand controlSeparates brand from new demand in structure and reportingReports blended return as growth
Scaling pathScales in steps against marginal cost per incremental conversion with stop rulesRaises budgets on average return
Operating rhythmSets weekly and monthly routines and owners from the account management playbookA plan with no review points

A strong strategy also sequences its parts. The signal plan comes first, because a geo test of value-based bidding means nothing until the values are trusted, and scaling comes last, because it needs the Incrementality reading from the budget test rather than the platform's attributed return. It treats the brand exclusion as a change to read at account level, since campaign figures will shift when brand queries move, and it keeps a cost view such as cost per qualified quote (CPA) alongside the return targets for the fit-out flow. On the platform question, a strong answer notes that a SA360 portfolio earns its place when bidding spans several engines; an account on one engine gains little from it.

Check your arithmetic against the reference figures below after working them out yourself.

Illustrative example — not a real result

Reference figures for self-checking

  • Fit-out margin per won order: £6,000 × 30% = £1,800.
  • Value per quote request: 25% × £1,800 = £450. Value per qualified quote: (25% ÷ 60%) × £1,800 = £750.
  • Online order margin at the blended product mix lies between £64 (storage) and £112 (chairs) on a £320 order.
  • Downloads and sign-ups: 2,700 of the 9,000 reported conversions, nearly a third, carrying no value at all.
  • Generic Search reported cost per conversion: £96,000 ÷ 3,600 ≈ £27; cost per quote request alone ≈ £137.

Submission checklist

0/12

When you have finished, compare your signal plan with the conversion values exercise for a solar installer, your test design with the geo split exercise for a machinery dealer and your brand split with the Performance Max exercise for a candle maker. For the account foundations the strategy assumes, revisit the intermediate Google Ads level.

Pages this lesson applies

Templates and checklists used

Terms used on this page

Also related

Related