Skip to content
Results

Every number here will arrive with its denominator

This is where client results get published. RampFunnels is pre-launch, so the first one is not on it yet. What is on it is the standard the first one gets held to: the eight things a case has to contain, the numbers that count, and the eight things that will never appear. Written down now, while agreeing to it is free.

Score your funnelSee what gets built

Not open yet. When it opens, the score and the report come back to you, not to a sales team.

The standard, in one sentence

A named company that signed something, their starting position taken from their own analytics before anything was built, two matched windows with the dates printed, every figure sitting on the count it came from, and the metrics that refused to move next to the ones that did.

Published after the company in it has read every word.

The format

A published result has eight parts and names all eight

Two of those eight are the ones a flattering case study quietly drops. They are not optional here, and the card below cannot be filled in without them.

The shape every published result takes

Placeholder

Eight fields, in this order, and none of them optional except the last one. RampFunnels is pre-launch, so all eight are empty. They are printed anyway, because a reporting standard written before there is anything to report is the only kind worth trusting.

  1. 01. Who

    The company, named, with written permission on file and the date it was given.

  2. 02. Where they started

    Relevant visitors a month, what converted, and what that cost, pulled from their analytics before anything was built.

  3. 03. What was built

    The instrument, the scoring model, the entry pages, the follow-up and the routing, itemised.

  4. 04. What was measured, and when

    Two windows of the same length, the same metrics in both, with the start and end dates printed.

  5. 05. What moved

    Every figure with the count it is a share of. No percentage travels on its own.

  6. 06. What did not move

    The metrics that stayed flat or went backwards, printed next to the ones that did not.

  7. 07. What we would do differently

    The decision we would take again and the one we would not, named.

  8. 08. Their words

    Printed only if the client wrote them. Nothing is drafted here for anyone to approve.

Nothing on that card is a preview and nothing on it is filled in with an example. An example in a template is a fabrication with a shorter shelf life: readers remember the numbers and forget the label that said they were made up.

The numbers

A completion rate is not a result. A booked call is

Three tiers, published together. Most reporting stops at the first one, because the first one is the easiest to make look good.

What it did

The easy tier, and the one to be suspicious of. Every one of these can look excellent while the pipeline stays exactly where it was.
  • People who started, as a share of the people who reached the page
  • People who finished, as a share of the people who started
  • Where the drop-off sat, by question number

What it identified

The middle tier. This is where the answers start doing work the contact record never could.
  • Named companies and roles, as a share of the traffic that arrived
  • Responses landing in the qualifying band, as a share of all responses
  • The segments the answers produced, and how many people fell into each

What it was worth

The tier that decides whether the money was well spent. It is the slowest to arrive and it is the only one that settles the argument.
  • Calls booked from the qualifying band, and calls that actually happened
  • Deals opened from those calls, against deals opened from the old form
  • Spend per qualified lead, before and after, on the same traffic

Every one of those is a fraction, and the bottom half gets printed.

The comparison

You will see the starting position, not only the finish

A figure with no starting position is not a result. It is a number, and a number can be walked to from anywhere.

So the 90 days before we start get pulled out of the client’s own analytics and written down while they are still just a quiet quarter, before anybody has a reason to be generous about what they were. Then the same 90 days again, the same metrics, the same report.

Before

The 90 days before anything was built.

Their analytics, written down first, so nobody chooses the baseline afterwards.

After

The same 90 days, measured the same way.

Published with the dates, the counts, and whatever refused to move.

The refusals

A number without its denominator is a number we invented

Eight things a results page is normally allowed to do. This one is not, and each of them costs us a sentence that would have read better.

  1. A percentage without its count

    A percentage on its own can be four leads becoming five. The count it came from is printed beside every figure or the figure does not go up.

  2. A composite client

    One case per company. Never several blended into a representative example, however well the blend reads.

  3. A logo without a signature

    A company mark appears here when that company has signed something saying it can, and not when somebody said yes on a call.

  4. A number borrowed across categories

    A recruitment firm’s figures are a recruitment firm’s figures. They are not evidence about your accountancy practice and they will not be presented as though they were.

  5. A conversion-lift statistic

    The interactive-content literature was checked. The most cited figure measures marketer opinion rather than performance, and the punchiest multiple traces back to a research organisation with no independent record of existing.

  6. A quote we drafted

    A sentence we wrote with somebody else’s name under it is our sentence with somebody else’s name under it, whoever approved it.

  7. A case the client has not read

    Nothing publishes until the company in it has read every word and said yes to the words, not to the idea of the words.

  8. Only the metrics that went the right way

    What stayed flat gets printed next to what moved. A case with nothing disappointing in it has had the disappointing parts removed.

That list is longer than the one most agencies could sign, which is the only argument for publishing it. It is easy to agree to now. The test is whether it is still here when there is a case that would read better without it.

Meanwhile

The proof that exists today is on three other pages

None of it is a client result. All of it can be checked without speaking to anyone, which is more than a case study from an industry you are not in would give you.

The research

174 templates classified on 23 fields, 40 B2B categories scored on 14 dimensions, 25 buyer segments ranked, finished September 2026. Original work, method attached, most of it free to read.
Read the Atlas

The method, in public

What gets built, in what order, who has to be in the room and what happens when the scoring model turns out to be wrong. Enough to run it yourself if you would rather.
How the work runs

This site

If the question is whether we can design something that converts the traffic you already pay for, you are currently standing inside the answer.
Why this exists

And the one you cannot check yet

Our own scorecard. It is being built, by the method and on the software we would use for yours, and it is the only item on this site that we will not describe as finished until it is. When it opens it becomes the fourth thing on this list and the best of them, because using it is the nearest you can get to reading a case study about yourself.

What people ask before they ask for a call

Your turn

Find out what your own funnel would score

A short set of questions about the traffic you already pay for, and a band, a number and a report at the end. The same instrument the first case on this page will be about.

Score your funnel nowTalk it through first

Not running yet. Nothing on this page changes when it opens, except that one of the cards above stops being empty.