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

Google Ads › Expert

Capstone: A Google Ads Operating Plan for a Hot Tub Retailer

The closing project of the expert Google Ads level. You write a year's operating plan for an invented hot tub retailer whose survey campaigns starve on thin data, whose sale and a checkout release distorted bidding last year, whose platform figures run ahead of its orders and whose new head of digital wants search run from Search Ads 360. The plan sets the bidding rules, the conversion truth and offline uploads, Shopping beside Performance Max, a staged platform move and the way each step of growth is judged.
Capstone projectExpert levelModule 3: Account architecture for scale8 minUpdated

Brief

Felquarn Spas is an invented online retailer of hot tubs and swim spas with an installation survey service. Swim spas and larger hot tubs need a paid engineer's survey before ordering, and the campaigns selling those surveys each record few conversions a month.

Last spring's sale ran out of budget on its main day, and a checkout release once stopped the purchase tag for a day. Finance says platform figures run ahead of its orders and wants one search figure across both engines. The new head of digital wants search on both engines run from Search Ads 360, as the agency has proposed. The advanced Google Ads level set the account's values, tests and brand control; this plan runs it through a hard year.

Inputs and constraints

Illustrative example — not a real result

Felquarn Spas: the account over the last twelve months

CampaignBiddingSpendPurchase valueSurveys bookedStatus
Brand SearchMaximise clicks£14,000£696,00060Impression share 88%
Generic Search, hot tubsTarget ROAS 900%£88,000£930,00045Lost to budget 22%, to rank 14%
Standard Shopping, all productsManual CPC, one product group per category£8,000£104,000Serves on few products
Performance Max, all productsTarget ROAS 1,000%, no brand exclusion£150,000£1,700,000Spends 85% of its budget
Survey: swim spasTarget CPA £120£9,000152Limited by low conversion data
Survey: large hot tubsTarget CPA £120£7,500129Limited by low conversion data
Survey: installation areasTarget CPA £100£4,50077Limited by low conversion data

Microsoft Advertising, brand and generic search: £41,000 spend, £420,000 of purchase value, and a purchase tag that passes no order ID. Google Ads: one purchase action for every product, data-driven, 30-day click window, engaged-view conversions included; survey bookings on a 30-day click window, counted every time.

Illustrative example — not a real result

Felquarn Spas: orders, surveys and timing

Survey line, booking recordsSurveys bookedSurveys completedInstallations orderedAverage margin per installation
Swim spas14011841£3,900
Large hot tubs12010444£2,200
Installation areas, any model705218£2,300
  • The booking form stores the click identifier and consent state; installation orders carry their survey reference and follow the survey by two to eight weeks.
  • Hot tubs, click to purchase in last year's matched orders: 36% within 7 days, 36% at 8 to 30, 19% at 31 to 60, 9% at 61 to 90. Accessories: 88% within 7 days.
  • A closed month by conversion time: Google Ads 118 click-through purchases worth £284,000 and 9 engaged-view purchases worth £21,000; Microsoft 17 worth £46,000. Orders: 121 from paid search, worth £291,000 net of cancellations and returns.
  • Margins: hot tubs 30%, accessories 45%, with 10 points of revenue to be left as profit after ad cost. Next year's ceiling: £450,000 across both engines.
Illustrative example — not a real result

Felquarn Spas: last spring, the outage and the agency's proposal

Last spring's sale ran Friday to Monday, as this year's will, with 15% off hot tubs this time. The agency lowered the Generic Search and Performance Max targets to 700% by hand and restored them nine days later.

CampaignDaily budgetConversion rate, fortnight beforeConversion rate, sale days
Generic Search, hot tubs£2500.9%1.6%
Performance Max£4800.8%1.4%
Sale Saturday, share of daily budget spent byGeneric SearchPerformance Max
12:00100%, limited since 11:4083%
14:00100%100%, limited since 13:20
  • In June a Thursday release moved the order confirmation page to a new address, and the purchase tag, still set on the old one, recorded nothing from 10:20 until 13:00 the next day, missing 34 orders. Seeing a poor week, the agency raised the Performance Max target to 1,200%; no data exclusion was applied and the missing orders were never uploaded.
  • The agency proposes one March weekend, three weeks before the sale: both engines linked, every goal replaced with Floodlight, every campaign renamed and put in one portfolio at a target ROAS of 800%, engine reports retired. Its Floodlight tags, left from a display test, have never been checked against engine counts.

Constraints:

  • Team: the head of digital, a paid search manager, a part-time analyst and the agency; the developer releases on a fixed fortnightly day.
  • No change that would restart learning is made between sign-off and the read of the sale.
  • Uploads respect the stored consent state.
  • Any effect you expect is given as a direction and the reading that would confirm it, never as a figure.

Task

Write one operating plan, in this order.

  1. Calendar. Place the sale and the release days, and order the changes so each is read alone, logged in a measurement change log.
  2. Thin data and unusual days. Consolidate the survey campaigns in the search campaign build sheet and choose their bidding step with the few-conversions bidding guide. Plan the sale with the sale readiness checklist, and write the outage response, a daily comparison of orders with the purchases the account records and a release rule, from the sales and outages bidding guide and the outage response SOP.
  3. Conversion truth. Set each action's model and windows from Felquarn's own conversion lag with the conversion windows guide, and reconcile the closed month by identifier with the reconciliation tutorial, agreeing the ratio with finance.
  4. Offline outcomes. Put completed surveys and installation orders on the offline conversion upload SOP routine.
  5. Shopping beside Performance Max. With the Standard Shopping structure guide, decide which products Standard Shopping owns: the custom label in the feed that keeps each product in one campaign type, priorities and negatives routing generic and model searches, margin targets from the ROAS calculator and the brand exclusion. Check the rebuilt listing groups with the Performance Max launch checklist.
  6. Survey leads. Decide whether Performance Max runs for them this year, with the Performance Max for leads guide.
  7. Search Ads 360. Answer the agency, then stage the move with the tutorial on moving paid search into Search Ads 360, each stage a phase-gate review.
  8. Growth at the margin. Take the ceiling from margin as the advanced capstone did. Order the headroom from the lost-share columns, budget-limited campaigns first, then targets, then wider queries, as the guide to scaling spend does; say which survey and Shopping lines wait until their signal is reconciled; judge each step at the margin on reconciled orders.
  9. Summary. One page for the head of digital and finance: the changes and their order, how each is judged and what triggers a rollback.

Evaluation criteria

CriterionWhat a strong answer showsCommon gap
Bidding on thin data and unusual daysConsolidates the survey campaigns, bids towards a step that predicts installations with the outcome imported, and plans adjustments, budgets and frozen changes for the sale and the outage responseOne target per thin campaign, and targets raised by hand for the sale
Conversion truthSets models and windows per action from the retailer's own lag and reconciles a closed period with orders by identifier against an agreed ratioPlatform totals reported as sales
Offline outcomesUploads qualified survey and installation outcomes on a documented routine with consent checks, QA and reconciliationA one-off upload with no owner
Shopping beside Performance MaxUses Standard Shopping where control is needed, with priorities routing generic and model queries and negatives, and states how it runs beside Performance MaxBoth campaign types on the same products with no routing plan
Move into Search Ads 360Links engines and Floodlight, maps conversion goals, labels and names, and hands bid strategies over in stages with reporting kept comparableEverything moved at once with every bid strategy reset
Scaling at the marginFinds headroom in budget-limited campaigns first, loosens targets and widens queries in steps, and judges each step at the marginGrowth judged on the account's average return

Order decides most of the grade. Conversion truth comes first, because it decides what every target and SA360 portfolio learns from; the sale then fixes the calendar, with the platform move and survey restructure either side of it. The agency gets a staged move rather than a refusal: Felquarn meets some conditions the tutorial on moving paid search into Search Ads 360 sets out, a second engine with real spend and finance wanting one search figure. Because Smart Bidding predicts from the conversion rates it has seen lately, as the guide to bidding through sales and tracking outages explains, the outage response holds targets still and the sale uses an adjustment. On survey leads, credit goes to the reasoning, whichever way the decision falls.

Work the figures out yourself first, then compare them with the reference set below.

Illustrative example — not a real result

Reference figures for self-checking

  • Hot tubs: break-even ROAS 1 ÷ 0.30 ≈ 333%; target 1 ÷ (0.30 − 0.10) = 500%.
  • Value per completed survey: swim spas (41 ÷ 118) × £3,900 ≈ £1,355; large hot tubs (44 ÷ 104) × £2,200 ≈ £931; installation areas (18 ÷ 52) × £2,300 ≈ £796.
  • Hot tub purchases outside the 30-day window: 19% + 9% = 28%.
  • Closed month: £284,000 + £21,000 + £46,000 = £351,000 of platform value against £291,000 in orders, a ratio of about 0.83.
  • Microsoft's share of search spend outside Shopping and Performance Max: £41,000 ÷ (£41,000 + £14,000 + £88,000 + £21,000) = 25%. Last year's spend: £281,000 on Google Ads and £41,000 on Microsoft, £322,000 in all.

Submission checklist

0/10

Pages this lesson applies

Templates and checklists used

Terms used on this page

Also related

Related