the ads reach people who can't buy.

install crews have gaps in the calendar. spend is steady, but most of it lands on renters, apartments and postcodes where few homes have gone solar.

sound familiar?

install crews have gaps in the calendar
ad spend is steady but enquiries are thin
enquiries come from renters and apartments
you can't say which postcodes actually buy

check my funnel see how i’d fix it

how i’d fix it · residential solar and battery installer · reach

six steps, from the bottleneck to a system you own.

01
find where it stalls
i ask for one number at every stage and set it against where it should be. the stage furthest below its target is the bottleneck, and it's the only thing worth fixing first.
find the constrainta month of the funnel, against target · illustration
1. Rreach6,000right-fit homes reachedtarget 10,000the bottleneck
2. Aacquire180enquiries3% of the stage before · target 3%
3. Cconvert36signed installs20% of the stage before · target 20%
4. Eexpand9add-ons and referrals25% of the stage before · target 20%
conversion further down is healthy. there just aren't enough of the right households at the top to fill 60 install slots a month.
02
break it down three ways
one angle hides things. i take the problem apart with three different methods, so the fixes come from the cause, not the first idea in the room.
five whysask “why?” until the cause is something you can actually change.
1. problemenquiries are thin
2. why?the right households don't see the ads
3. why?targeting is a radius around the depot
4. why?nobody has mapped where solar sells
5. why?the data is public, but no one has pulled itthe root cause
inversionask how to make it worse, then do the opposite of every answer.
to make it worseso instead
1. show ads to every postcode equallyweight spend to high-uptake postcodes
2. run one creative until it tirestest new creative every fortnight
3. judge ads on clicks, not enquiriesjudge ads on cost per enquiry
4. never check who actually enquiredmatch enquiries back to postcode tiers
goal factoringstart from the goal and ask “how?” until every branch is something you can build this week.
more owner-occupiers in high-uptake postcodes see the offer
1. raise the budget
2. target postcodes by uptake
3. more creative, tested
4. add a channel
1. pull public install data by postcode
2. tier postcodes by uptake and home ownership
3. rebuild ad locations from the tiers
4. find the ads that already win on cost per lead
5. write five new hooks for each winner
6. create them paused, for review
6 concrete actions. most of them are a good fit for claude.
03
put every option on the table
the breakdowns turn into a long list of possible fixes. each one is plotted by how much it moves the number against how much work it takes.
every option on the tableimpact against effort · picked options in colour
less effort → more effort
more impact ↑
1. 1postcode tiers from public install data
2. 2a fortnightly creative expander
3. 3cost per enquiry by tier, every friday
4. 4double the budget
5. 5add youtube and display
6. 6letterbox drops in top postcodes
7. 7a new website
8. 8a partnership with local builders
04
pick what fits you
every option is scored on impact, effort and how much change it asks of your team. the top few go ahead; the rest wait, and i say why.
score every optionimpact counts twice · less effort and less change for your team score higher
impact ×2effortchange for your teamscore
1. 1postcode tiers from public install data19
2. 2a fortnightly creative expander17
3. 3cost per enquiry by tier, every friday16
4. picked above the line
5. 4double the budget15
6. 5add youtube and display12
7. 8a partnership with local builders11
8. 6letterbox drops in top postcodes10
9. 7a new website6
what i’d pick, for youthe top scores, and why they suit your team
1. 1
  postcode tiers from public install data
  uses the ad accounts you already run, and the data is free.
  no new tools
2. 2
  a fortnightly creative expander
  new ads are created paused, so nothing spends until you say so.
  you approve every ad
3. 3
  cost per enquiry by tier, every friday
  proves the shift is working before anything else moves.
  a report, not a change
the 5 we’d leave for now, and why
1. 4double the budgetmore of the same reach makes the waste bigger.
2. 5add youtube and displayworth it later, once tiers prove where to aim.
3. 8a partnership with local buildersgood long term, too slow to fill next month's crews.
4. 6letterbox drops in top postcodeshard to track against enquiries.
5. 7a new websitethe site isn't the bottleneck; the audience is.
05
build it and hand it over
claude does most of the building, inside the tools you already use. every part runs on its own, and you keep the files.
the systemwhat runs, in order · illustration
1. datapublic install data
2. scriptpostcode tiers
3. toolad account
4. claudecreative expander
5. youpaused ads, you approve
claude · the prompt behind it
pull every active ad from the last 90 days. rank them by cost per lead inside each postcode tier. for each top ad, write five new hooks and headlines in australian english, with no invented savings figures. create them as paused ads and summarise what changed and why.
runs without mein your own claude team account · yours to keep
1. 1st of the monthrefresh postcode tiersscheduled script
2. monday 7:00expand the winning adsclaude routine
3. fridaycost per lead by tierreport
client-workspace/
1. skills/creative-expander/SKILL.md
2. routines/monday-expand.md
3. data/postcode-tiers.csv
4. evals/creative-expander.csv
5. RUNBOOK.md
06
prove it, then repeat
we agree the numbers before we start and check them every week. when this stage is fixed, the next bottleneck shows up and we go again.
prove itbaseline → target, measured weekly · illustration
beforetarget
1. spend in top-tier postcodes
  35%
  80%
2. new creatives tested a fortnight
  2
  10
3. enquiries a month
  180
  260
evals ten runs of the same prompt, scored out of ten · on brand, no invented figures, within platform character limits
8979899879
average 8.3 / 10 · rerun after every change and every new model

want these in your funnel?

postcode tiers from public install data
a fortnightly creative expander
cost per enquiry by tier, every friday

check my funnel

what you get

fixed in your funnel, yours to keep.

a postcode tier file, refreshed monthly
the creative expander skill and its monday routine
a weekly cost-per-lead report by tier
the runbook and eval set, yours to keep

1send the funnel check below. it takes a minute.
2i reply with your bottleneck and what i’d build first.
3you get a fixed scope and price in writing before anything starts.

questions

do we need new software?

usually not. i build into the crm, ad accounts and inbox you already use, and say up front if something has to change.

who owns what you build?

you do. the skills, routines, sequences and runbook live in your accounts, and they keep running if we stop working together.

what if my bottleneck is somewhere else?

then the check will show it, and i'll tell you before you spend anything. fixing the wrong stage doesn't move the number.

how is it priced?

fixed scope, priced in writing before we start. no hourly billing and no lock-in.

the ads reach people who can't buy.

six steps, from the bottleneck to a system you own.

find where it stalls

break it down three ways

put every option on the table

pick what fits you

postcode tiers from public install data

a fortnightly creative expander

cost per enquiry by tier, every friday

build it and hand it over

prove it, then repeat

fixed in your funnel, yours to keep.

is this your bottleneck?