Landing Page Conversion Plan · October 2026

Fallowmere Commercial Cleaning Ltd

Contract cleaning for offices, schools and surgeries across Greater Manchester. 120 staff.

Worked example. Fallowmere Commercial Cleaning and its figures are fictional. All figures come from the worked-example spreadsheet: ALL CHECKS PASSED (with 1 note).
The headline

Five ideas worth about £61,879 a year, allowing for most not working

First-year net
£44,559
for £1,850 of work; £21,355 if the ideas do half as well
Expected
124
extra enquiries a year, at the chance each idea works
Per enquiry
£578
of gross profit at 75% of today’s quality, before £80 to handle it
Hard to see
4
ideas whose lift wouldn’t show in six months of figures

Verdict: Of the ideas (0 to test; 5 to just do; 1 parked), each extra enquiry is worth £578 of gross profit. Allowing for the chance each one works, they should bring about 124 extra enquiries and £61,879 a year (£161,190 if every one worked), for £1,850 of work: a first-year net of £44,559 (£21,355 if they do half as well). Worth most: “Cut the quotation form from 14 fields to 6”. Best for the money: “Cut the quotation form from 14 fields to 6”. 4 ideas’ lift wouldn’t show in six months of figures: judge by an earlier step (form starts, clicks to call).

The pages

A month’s visits and enquiries

PageVisits a monthEnquiries a monthEnquiry rate
Home page4,200421.0%
Office cleaning service page1,800271.5%
Google Ads landing page1,500302.0%
Quotation form (people who start it)9009911.0%
Idea by idea

Most expected gross profit first

IdeaPlanExtra enquiries a yearGross profit a yearCostFirst-year net
Cut the quotation form from 14 fields to 6
Quotation form (people who start it) · visitor feedback or recordings · 15% lift if it works, 40% chance
Just do it
visible in 2 months
71£35,462£300£26,296
Show prices: from £X a week for a typical office
Office cleaning service page · best practice · 15% lift if it works, 30% chance
Just do it
visible in 7 months
15£7,254£200£5,240
Phone number and live chat at the top
Google Ads landing page · visitor feedback or recordings · 10% lift if it works, 40% chance
Just do it
visible in 13 months
14£7,164£450£4,923
Client logos and reviews above the form
Google Ads landing page · test result elsewhere · 8% lift if it works, 50% chance
Just do it
visible in 21 months
14£7,164£100£5,273
Case studies by sector: schools, surgeries, offices
Office cleaning service page · best practice · 10% lift if it works, 30% chance
Just do it
visible in 15 months
10£4,836£800£2,827
New hero image of the team at work
Home page · hunch · 5% lift if it works, 20% chance
Park
visible in 38 months
0£0£0£0

The agency wanted to test three ideas, but none could finish in eight weeks: these pages don’t have the visitors. So they’re made as changes, one at a time, and the hero image (a hunch, worth little) is parked. Most of the lifts wouldn’t even show in six months of before-and-after figures: judge each by an earlier step (form starts, clicks to call). The shorter quotation form is worth most because every enquiry passes through it; it’s measured on the people who start the form, so its effect shows within two months. Page ideas and the form act on different steps, so their gains don’t overlap.

How confident are we?

What this rests on

Visits and enquiries come from the business’s analytics. Each idea’s lift counts only at the chance it works (50% if backed by a test elsewhere, 40% from visitor feedback, 30% best practice, 20% a hunch). Extra enquiries are worth 75% of today’s, at a 25% close rate and the first order × margin, less £80 each to handle, with work starting 2 months after the enquiry. A test needs enough visitors to see the lift (about 16 × rate × (1 − rate) ÷ (rate × lift)² each way); tests run no longer than 8 weeks, other changes take 1 month to go live. Lifts on the same page are added, which is optimistic; a page idea and a form idea act on different steps, so they don’t overlap.

If every idea worked the gain would be £161,190 a year: most won’t, which is why each is counted only at its chance of working.

The months after

What the agent flagged, and what was decided

At handover, the managing director agreed to make the changes one at a time, so each could be read from the figures. The agent was then set up in Fallowmere's own AI assistant account, with read-only access to the website analytics and the ad platform. On the first Monday of each month it reads each page's visits and enquiries for the last three months, compares each page's enquiry rate with its rate at the plan, and reports each change's result once the plan's months of figures are in. It uses the rule on Inputs: a fall of 20% in a page's enquiry rate is flagged. The agency records the date each change goes live. Nothing left Fallowmere, and no decision changed, without one of its own people approving it. These events continue the fictional example.

Monthly runs
5
November to March
Flags raised
2
a fall on the ads page, and a data problem
Results reported
1
the shorter form, once its two months of figures were in
Too early to tell
1
prices on the office page: about 7 months needed
DateWhat the agent foundFlag?What people decided
2 NovOctober's figures: every page within the line. The shorter form had been live a week, too early to tell.Note onlyNothing needed.
7 DecThe Google Ads landing page's enquiry rate fell 32%, to 1.35%: the campaign had moved to broad match, bringing many more visits but no more enquiries.FlagThe campaign went back to phrase match, and the next change on that page waited so its effect could be read on its own.
4 JanNo home page figures for December: a site update had removed the analytics tag.FlagThe agency put it back; December was left out of the window.
4 JanThe shorter quotation form, after the two months of figures the plan said it needed: 12.9% of people who started the form sent it, against 11.0% before.Note onlyKept, and the next change scheduled: prices on the office cleaning page.
1 FebThe ads page back within the line at 1.63%, with November's broad-match month still in the window.Note onlyPhone number and live chat went on that page next.
1 MarPrices on the office cleaning page: too early to tell. The plan says about 7 months of figures are needed on a page with 27 enquiries a month.Note onlyThe result is due in the August note.

At the check-in after the first monthly run, on 30 November (about six weeks after handover), Geoff and the managing director read the log together, and Geoff's upkeep of the set-up ended there. They kept the 20% line, and agreed the agency would record each change's live date the day it happens.

What the agent added. The method is unchanged: same pages, same ideas, same rule for how many months of figures each change needs. The agent made sure the figures were read every month, so the broad-match problem was found within a week of November's figures, and each change's result was reported only once there were enough figures to read it. The figures above are fictional, and no £ figure is claimed for the agent.

Fictional and labelled: these events continue the worked example. The same six events are the sample rows in the Agent log of the worked-example spreadsheet. The agent is part of the guided and done-for-you routes; the self-implement kit does not include it.