Form Submissions as the Only Number That Matters

The thought experiment

Suppose you were allowed one figure about your website and nothing else at all.

Not visits, not rankings, not time on page, not bounce rate.

Almost everybody, given that constraint, would choose how many people asked to be contacted, because it is the only one that corresponds to something happening.

Which raises the question of what the other eleven figures on the report are for.

What one number can answer

• Is the site working at all.
• Was this month better than last year.
• Did the change in March do anything.
• Is the advertising producing anything.
• Is something broken.

All five, from one series of monthly counts. The fifth is the one people assume needs monitoring and it does not: a figure that falls to zero is the loudest possible signal that something has broken.

Why this one and not visits

Since visits are the figure everybody reports.

Visits move for reasons that have nothing to do with the business: a mention somewhere, a bot, a seasonal search pattern, a page ranking for something irrelevant.

Submissions move because somebody decided to contact you, which is the thing the site exists to produce.

A month where visits fall and submissions hold is a good month reported as a bad one.

And a month where visits rise and submissions fall is a problem that a visit-led report actively conceals.

One of those figures describes the outcome and the other describes the weather.

A worked example

A business replaced their five-figure monthly report with one number and a note.

Submissions, the previous year's figure for the same month, and one sentence saying what happened.

Within a quarter the conversation about the website had changed entirely, because there was nothing to discuss except the thing that mattered.

Two decisions came out of it in a year: stopping a piece of advertising that had never moved the number, and rewriting a service page that had.

The previous reports had never produced a decision in four years.

The change was removing figures rather than adding any.

Add the source, not more metrics

Which is the one useful extension.

Knowing that eleven people submitted is useful; knowing that four came from search, five from the listing, and two from a referral is considerably more so.

That is a breakdown of the same number rather than a second number, which is the distinction that keeps a report short.

Resist adding anything that is not either the count or its composition.

Rate, percentage, and average are all derived figures that make a small number look more precise than it is.

Include the ones that arrive another way

Since the point is the count of people asking rather than the mechanism they used.

An enquiry by email, through a listing, or by a message on a social platform is the same event as a form submission and belongs in the same number.

Splitting them across four systems is how a business ends up unable to say how many enquiries it received, which is the question the whole exercise exists to answer.

One column, one total, with the source noted beside each, keeps it a single figure while preserving the breakdown.

That also means the number survives you changing your form, your platform, or your provider, none of which affects a count you keep yourself.

Count it yourself as well

Since a tracked figure can fail quietly.

Your inbox contains the submissions, so counting them takes two minutes and is not dependent on any tool.

Where the tracked figure and your inbox disagree, the inbox is right and the tracking needs attention.

That check also means the switch happening at the end of this month cannot break your reporting, since the number you actually use is one you count.

Businesses that made this change in advance are having a considerably quieter July than those that did not.

What you give up

Stated honestly, since it is not nothing.

You lose the ability to diagnose why a change happened, which the other figures help with.

You lose the early signal, since traffic moves before submissions do.

And you lose the ability to compare a page against a page, which needs page-level data.

The answer is that those are available when you need them, in the tool, on the day you have a question.

What changes is that they stop arriving monthly whether or not anybody has a question, which is the entire proposal.

Look deeper only when the number moves

Which is the discipline that makes this work.

A month where submissions are within the usual range needs no investigation at all.

A month where they fall by a third is when you open the other figures and find out which pages, which sources, and what changed.

That is the same investigation you would have done anyway, done when there is a reason rather than every month.

It also means the deeper look gets proper attention, since it is not a routine.

Show it against last year, not last month

Since a single number needs a comparison to mean anything.

Month against month is dominated by seasonality, so a quiet February compared with January reads as a decline when it is simply February.

The same month last year removes that entirely and is the only comparison a small business can read reliably.

That requires two years of the figure before the report becomes useful, which is an argument for starting the count now rather than when it feels important.

In the first year, note the figure and resist drawing conclusions, which is difficult and correct.

The counter-case

One number hides things.

A business whose enquiries arrive mostly by telephone is measuring a minority of its outcomes and should count both.

Anybody spending meaningfully on advertising needs more than a monthly total to manage it.

And a shop needs orders, revenue, and the checkout funnel rather than a submission count.

Count submissions monthly from your own inbox, add the source of each, and open anything else only when the number moves.

The report

1. Count submissions from your inbox.
2. Note the same month last year.
3. Break it down by source.
4. Write one sentence.
5. Add nothing else.
6. Investigate only when it moves.
7. Count calls too if they matter.

Step five is the hard part, since every report grows by somebody adding a figure that seemed interesting and nothing ever gets removed.

The original argument for this is covered in counting enquiries instead of visits.

Frequently asked questions

West Coast Media Solutions Inc. provides web design, web development, hosting, digital marketing, and business consulting to organisations across Canada, drawing on more than twenty-five years in the field.