What’s going to happen next?
Marketing is never a clean line that only goes up. But it does give off signals — the ones that warn you a dip is starting, and the ones that show where money is leaking. The only real question is when you hear about them: two or three days in, while there is still something to fix, or four to five weeks later, when the quarter has already gone.
What it looks like when it works
Imagine finding out in days, not weeks.
A marketing team that tells you a drop has started while there is still a month left to do something about it. One that reads your real figures — not the flattering ones — and tells you plainly which change grows the business next.
The two jobs
Where reporting and forecasting fall apart
Two different jobs with two different failures. Most businesses have been sold one of them and are missing the other entirely.
Reporting tells you what already happened. Forecasting tells you what is likely to happen next, and what to do about it. You need both, and most businesses have only ever been sold the first.
They fail in completely different ways, which is why fixing one does nothing for the other.
- Reporting is usually present, and usually useless. It arrives on time and counts the wrong things.
- Forecasting is usually absent altogether. Nobody is reading the month while it is still running.
The result is a business that knows a great deal about last month and nothing at all about this one. Every decision gets made on figures that were already history by the time they arrived, and the only lever anybody has left is to spend more.
The two tabs alongside take them one at a time.
A report can arrive on time, look professional, and still not answer the question you asked.
Where forecasting does exist, it breaks in one of two ways.
How it usually arrives
One figure
An exact number for next month, offered with confidence. When it misses, the client stops believing any of it.
What it should be
A direction
Which way a number is moving, and what has to change to move it — read while the month is still open.
The second way it breaks is quieter. It gets run on data nobody cleaned first, which is a pattern read out of noise rather than out of the business.
Both failures come from the same place: forecasting is treated as a document to produce rather than something to do every week. A forecast that is not corrected as the month runs is just another report, written earlier.
Handré du Preez
Role
Operational Manager at LDP Digital
What I do
I spend my days working out why good traffic stops short of becoming revenue — and then fixing it.
Specialities
- Marketing
- Website Optimisation
- Deep Funneling
Qualifications
- 2nd strongest Revenue Engineer in the group
- Degree in Business Commerce
Reporting and forecasting
Two different jobs. You need both.
These two words get used as if they mean the same thing, and they do not.
Reporting tells you what happened.
Forecasting tells you what to do about it.
Reporting — what happened
The record of the month that just closed. Done properly, it is honest rather than flattering.
- Tracks progress against what you set out to do
- Shows the real figures, not the flattering ones
- Measures the points that actually count as success
- Highlights the bottleneck that needs attention
Forecasting — what happens next
The same figures, acted upon while there is still time to change and prevent a drop in performance.
- Says what has to change to lift performance
- Says when a dip is likely to ocurr, and how to prepare for it
- Says where budget should move for better cost efficiency
- Plans the customer journey around trends, actions and what the market leads with.
Don’t only measure what has happened start measuring what is coming.
Said plainly
A forecast is a direction, not a promise
So here is exactly what you get from us, and what you never will.
- What has to changeThe targeting, the assets, the page the traffic lands on — whatever the figures say is holding the result back.
- How long it takes to showChanges do not land the week they are made. The platforms have to relearn, and that takes as long as it takes.
- Which way it should moveQuality up. Cost per sale down. The direction the figures should travel once the change is in, and what we are watching to confirm it.
- What we are correcting nextEvery call gets checked against what actually happened, and the next one is tighter for it. The forecast is never finished.
Forecasting was never a promise. It is a reading of what the data is showing right now, and a plan to improve on it — optimising what is running, lifting conversions, and getting more sales and better-quality leads out of the same budget.
And it needs figures to read. Markets move, trends move, and without real data behind an account there is nothing to read a pattern from. Where that history does not exist yet, we will say so rather than hand over a number that was never going to hold.
A conversation we have often
Is my marketing truly working?
Impressions went up. So did the interaction rate. Neither one tells you whether the marketing sold anything.
Handré du Preez
Role
Operational Manager at LDP Digital
What I do
I spend my days working out why good traffic stops short of becoming revenue — and then fixing it.
Specialities
- Marketing
- Website Optimisation
- Deep Funneling
Qualifications
- 2nd strongest Revenue Engineer in the group
- Degree in Business Commerce
The short list
The two numbers that settle it
Sales, and qualified leads. Every other figure on a report is context for those two. If a report cannot put them next to what you spent, it cannot tell you whether the marketing worked — however good the impressions look.
“It is a capital mistake to theorize before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts.”
Arthur Conan Doyle, A Scandal in Bohemia, 1891
Business owners are consistently surprised by how obvious the next move becomes once the right numbers are in front of them. The decision was never the hard part. Seeing clearly enough to make it was.
The mechanism
What we actually do
Four steps, in order.
- 1Measure what you already runWe put proper measurements on the marketing you have now, so that every figure after this comes from your business instead of an assumption.
- 2Separate the numbers that matterSales and qualified leads on one side. Impressions, clicks and engagement on the other. Most reports lead impressions, clicks, etc. but it is qualified leads and ultimately sales that grow your revenue.
- 3Act while the month is still openWhen a number starts moving the wrong way, you hear about it that week, together with a plan on what should change and why.
- 4Check the change, then tighten itEvery change gets measured against what it was meant to do. What worked stays. What did not gets corrected, and so your marketing gets optimised
None of that is a report arriving faster. It is a report that gives the right information.
The obvious objection
We already get monthly reports. What does forecasting add?
A report and a forecast are built to answer different questions.
A monthly report is a record. It arrives once the month has closed, because that is the earliest it can be complete and by then the budget has already been spent.
Forecasting reads the same figures while the month is still running. When something starts moving the wrong way you hear about it with part of the month still left, and the budget that has not been spent more strategically than where it was heading.
Four ways to tell which one you are being sent. None of these are about a report being late or badly designed.
- You ask what is changing next month. You get a shrug, and a chart of last month.
- You ask how something is performing. The answer is “well”. No number, no comparison, nothing you could act on.
- You ask what changes when the market shifts. Nobody has an answer, because nobody was watching for it.
- You ask what the plan is for the off-season. There is not one. The same marketing runs in a dry month as in a peak month, and the same budget goes with it.
Proof
The forecast said 90 days. They stopped at 30.
Not a number for next month. A read on what had to change, and how long the platforms would need to relearn before it showed.
The account arrived with the wrong targeting in it. Not slightly wrong — wrong enough that the platforms had learned the wrong buyer entirely, and everything they did next was built on that.
So we told the client what had to change and what to expect after the changes had been made. The quality would be good early. The volume would not. The account did not yet know who it was looking for, and it had to relearn from clean data before it could go and find more of those ideal clients.
That is exactly what happened. Good leads came in from the start. There were not many of them, and the client wanted volume.
Under 30 days in, they pulled the account.
The relearning stopped where it was. When they came back there was nothing left to build on. Everything, back to square one. Our changes returned to old methods and that ideal client vanished. So we started again from scratct. The same 90 days, from the beginning.
The second run went the full 90. Same budget, and the same work going into it: targeting rebuilt, assets changed, the website changed, the account corrected against what the data was actually showing.
What finishing the 90 days produced
What ran to the end, and what it returned
Lead volume
3×
Three times the leads arriving each day by the end of the 90, on the monthly budget the account started with.
Leads worth calling
4–5×
The share of them the sales team could actually do something with, which is the number the client came to us about.
Cost per quote
↓ 82%
What it cost to get to a real quote, once the targeting and the landing page were doing their job instead of working against each other.
None of that came from spending more. It came from letting the changes run long enough to be measured, and correcting them against what the figures were actually showing. The forecast did not predict those numbers — it said what had to change, and how long before anyone could fairly judge it.
A second case
Emmarentia Guest House
Emmarentia Guest House, Johannesburg. The case is hospitality-specific. The principle behind it is not.
Emmarentia priced its rooms the way most guest houses do — flat seasonal rates, adjusted by feel. Once real booking-pace data and a monthly measurement framework were in place, pricing could move with actual demand instead of with a fixed calendar.
The figures alongside are revenue against the same months a year earlier, when the flat seasonal rates were still running.
The growth did not come from cutting prices to fill rooms. It came from seeing the pattern early enough to act on it.
Get your blueprint today
Ready to see what’s actually happening?
15 minutes, no obligation. You’ll leave the call knowing exactly what your current setup can and can’t tell you.
Handré du Preez
Role
Operational Manager at LDP Digital
What I do
I spend my days working out why good traffic stops short of becoming revenue — and then fixing it.
Specialities
- Marketing
- Website Optimisation
- Deep Funneling
Qualifications
- 2nd strongest Revenue Engineer in the group
- Degree in Business Commerce
See all Revenue Engineering posts →
Questions About Forecasting & Reporting
Reporting tells you what happened: last month’s revenue, your cost per sale, which channel produced what. Forecasting tells you what is likely to happen next and what to do about it. Reporting is the evidence; forecasting is the decision made from it. Most agencies deliver the first and describe it as the second.
Days rather than weeks, once the reporting layer is clean and the right leading measures are being watched. The delay in most businesses is not the analysis — it is that nobody looks at anything until the monthly report lands.
No. You do not need to understand attribution models any more than you need to understand an engine to drive a car. We handle the analysis; you get a plain answer to what should happen next, and the reasoning behind it if you want it.
That is the normal starting point, not a disqualifier. Getting the data clean and consistent is usually step one — and it matters far more to an accurate forecast than any clever method layered on top of messy numbers.
Usually as part of step one, and it is less about software than about agreeing what counts. Once everyone measures a lead, a sale and a cost the same way, the argument about whose number is right tends to stop — and the forecast finally has something solid to stand on.
Three months is what it took in that specific case — long enough for lead quality to settle and a real pattern to appear. The exact timeline depends on your sales cycle and how much history already exists, but it is always long enough to see a genuine pattern rather than a lucky week.
“Seasons” here means any predictable up-and-down in your business — a slow month, an end-of-quarter push, a weekly rhythm. Every business has some version of it, even when it is not calendar-driven.
The whole method, in sixteen pages
This pillar is one of seven, and they only hold together when they are read together. The Revenue Engineering brochure walks the whole path — where the engine leaks, what each leak costs, and what has to be measured before any of it can be fixed. Have it emailed to you instead, or pass it on.