How to run ads for a service with a long sales cycle
How to optimise paid media when the sale closes months after the click. Covers proxy conversions cohort reporting and when not to run ads at all.
The platform cannot see your sale
An ad platform improves by learning which clicks turned into conversions. It learns from conversions that arrive inside its reporting window. If your service closes four months after first contact then the platform never sees the outcome and so it never learns from it. It optimises towards whatever cheap event you did give it.
This is the central problem for consultancies equipment suppliers industrial services and anything sold through a proposal. The machinery works exactly as designed. It is fed the wrong signal because the right signal arrives too late.
What this approach costs and what it excludes
The cost that catches businesses out is the lag. You will fund several months of spend before you can honestly assess it. That is not an agency failing. It is arithmetic and it has to be budgeted as a period rather than a monthly line you judge each month.
Excluded from any media budget is the customer record keeping that makes this measurable. If enquiry source is not captured and retained through to the closed deal then none of this reporting is possible and you are buying clicks blind. Nurture material and follow-up sequences are excluded too and a long cycle depends on them. Ongoing management across these moving parts is available as a published scope under performance marketing management. Service fees carry GST at 18%.
Who should run this and who should not
This suits a business with enough deal volume for patterns to emerge and capital to fund months of spend before revenue appears. It also suits anyone whose buyers research before they enquire because there is something to influence during that research.
Do not run this if you need revenue this quarter. Paid media against a long cycle cannot deliver it and starting now will produce cost now and pipeline later. A business under cash pressure is better served by direct outreach to people it already knows. Do not run it without a working customer record. If nobody will reliably record where an enquiry came from then skip the channel because you will not tell a good month from a lucky one. And if you close only a handful of deals a year then accept that measurement will never be conclusive. Either run paid media as a presence channel judged on qualitative evidence or put the money into relationships and referrals which at that volume usually return more.
Optimise on a proxy you have checked
Pick one event earlier in the cycle that reliably precedes a sale and that happens often enough for the platform to learn from. A booked consultation a submitted requirement or a downloaded specification document. Then verify the correlation in your own records rather than assuming it. Many obvious proxies turn out to attract people who never buy and optimising towards one of those is worse than optimising towards nothing.
Where the platform supports importing offline conversions send the later stages back. Feeding a qualified opportunity or a closed deal into the account with its original click identifier is the only way the system learns what a good lead looked like. Weight the values so a closed deal counts for more than an enquiry.
Build long remarketing windows because your buyer is in the market for months. Keep brand search in its own campaign so it does not flatter the account. And leave the learning period alone. Restructuring a campaign every fortnight prevents a long-cycle account from ever stabilising.
Report by cohort not by month
Cost per lead flatters badly here and platform return on ad spend is meaningless because the revenue it needs sits outside its window. A monthly report comparing this month's spend with this month's closed revenue describes two unrelated things.
The measure that matters is cost per qualified opportunity and pipeline value created by the cohort of spend that produced it. Group enquiries by the month they arrived and follow that group forward. The month that looked poor at the time frequently looks strong a quarter later. Nobody can tell you in advance how long your own cycle runs so measure it from your existing closed deals before you set any expectation.
The common mistake
The expensive error is cutting the channel in month three. Spend has accumulated closed revenue has not and the decision looks obvious. What is actually happening is that the first cohort is only now reaching the proposal stage. The budget stops precisely when the evidence was about to arrive and the business then carries the cost with none of the return.
The related error is switching the optimisation target every few weeks in search of a cheaper figure. Each switch resets the platform's learning and an account reset repeatedly never accumulates enough signal to perform. Decide the proxy carefully then leave it.
The discipline this depends on
All of this rests on one unglamorous habit. Enquiry source has to be recorded at the moment of first contact and it has to survive through to the closed deal. That means whoever answers the phone asks and records the answer and whoever updates the deal does not overwrite it. Businesses attempt this reconstruction months later from memory and the result is worse than useless because it is confidently wrong. Fix the recording before raising the budget. Without it every other decision here is guesswork.
This guide explains how something works. It is not legal, tax or financial advice, and rules change. Confirm anything regulatory with a qualified professional before you act on it.