What marketing automation really costs

The price on the vendor's page is the smaller part of the bill. Over three years, small companies typically see only 30 to 50 per cent of total cost in the licence. This piece counts the other items.

Two glowing masses of different size hold each other in balance at the ends of an implied axis

In short

  • The licence is the smaller part: over three years, setup, upkeep and working time usually add up to more than the subscription.
  • Four items appear in no quote: data cleanup, getting people up to speed, ongoing upkeep, and getting out again.
  • Pricing tiers based on contact count are the most common cost trap — they grow with the list, not with the benefit.
  • The investment has paid off when saved working time plus additionally closed enquiries exceed total cost — usually after nine to eighteen months.

There is an honest answer to what marketing automation costs, and a useful one. The honest one is: it depends. The useful one is to list the items missing from almost every calculation — which together often exceed the licence.

Four translucent planes float one above the other, the lowest glowing, the upper ones fading
The visible price is one of four layers. The other three appear in no quote.

The four cost layers

Layer 1 – Licence

The advertised price. Almost always tiered by contact count, by users, or both. Visible, plannable, and the only item that usually features in the decision at all.

Layer 2 – Setup

Data cleanup, import, connecting website and inbox, building the first processes, templates. One-off but substantial: for small companies, 20 to 60 working hours is realistic — either as your own time or as an invoice.

Layer 3 – Ongoing upkeep

Keeping content current, adjusting processes, cleaning lists, reading reports. The most underestimated item: two to six hours a month, permanently. Fail to plan for it and after a year you have a system nobody trusts.

Layer 4 – Lock-in and exit

What does it cost to get out again? Can contacts, histories and content be exported in full — or only addresses without history? This item rarely arises, but when it does, it arrives at full size.

Worked example for three sizes

The figures below are experience-based estimates, not a survey. They show orders of magnitude and ratios — not one particular vendor's quote.

ItemOne-person business
1 person, 500 contacts
Small firm
5 people, 3,000 contacts
Mid-sized
20 people, 25,000 contacts
Licence per year300–9001,800–6,0009,000–36,000
Setup, one-off20–30 h own time3,000–9,00012,000–40,000
Upkeep per year24–40 h own time60–120 h0.3–0.8 of a role
Add-on services per year0–300300–1,5001,500–8,000
3 years, totalapprox. 1,500–4,000
+ 100–160 h
approx. 12,000–30,000
+ 180–360 h
approx. 50,000–150,000
+ headcount

The licence share of total cost sits below half in all three cases. For one-person businesses doing everything themselves, the largest item is their own time — and that appears in no invoice, even though it is the scarcest resource they have.

Worth knowing

Pricing tiered by contact count is the most common reason costs rise faster than benefit. It rewards nothing connected to results — only the growth of a list.

Never clean up and you pay year after year for contacts that have not opened anything in three years. A quarter to a third of a grown list typically consists of such entries. They cost money, lower your delivery rate and distort every report.

What appears in no quote

A small bright shape above a dark plane, beneath it a much larger, barely visible mass
The part below the line is bigger than the part above — and rarely counted.

Data cleanup before you start. Duplicates, dead addresses, inconsistent spellings. One to three days that nobody plans for and that happen anyway — before the import, or afterwards at twice the cost.

Getting everyone up to speed. Not the training, but the weeks afterwards when every action takes longer. One to two months of reduced pace is normal.

Add-on services billed separately. Send volume above the allowance, SMS, telephony, additional connections, storage. Small individually, a noticeable item together.

The exit. If histories, reporting records and built processes cannot be exported, switching is not a move but a fresh start. That question belongs before signing, not after.

From practice

A recurring pattern: a company picks an extensive system, sets up three processes in the first weeks, and is still using exactly those three two years later. It pays for the full scope throughout.

The check is uncomfortably simple: which features from the quote have you actually used in the last three months? In most cases the answer is the ones in the smallest tier. A smaller package used fully is almost always the better deal than a large one used a quarter.

When it starts to pay

The investment has paid off once two quantities together exceed total cost.

  1. Saved working time. Recurring manual steps that fall away: maintaining lists, writing follow-up emails, assembling reports. Realistically two to eight hours a month, depending on size.
  2. Additionally closed enquiries. The larger lever. Enquiries nobody followed up before and that now get answered. One extra closed enquiry per quarter covers the annual licence of a small system at four-figure order values.

For one-person businesses and small firms, the point at which it pays typically falls between nine and eighteen months. Stop after three months and you have paid for the setup and not waited for the return.

Tip Before deciding, work out what the current state costs — hours on recurring manual steps plus enquiries not followed up. Without that comparison figure every quote looks expensive, because it is being compared against zero. But the existing state is never free; it is just not invoiced.
Prompt
Work out for me what marketing automation would cost us over three
years, including the items that do not appear in the quoted price.

Our situation:
- People in marketing/sales: [number]
- Contacts in the list: [number], of which active: [number or
  unknown]
- Enquiries per month: [number]
- Average order value: [amount]
- Close rate: [per cent or estimate]
- Recurring manual steps that cost us time today: [bullet points]
- Internal hourly rate: [amount]

Tasks:
1. Set out the four cost layers for us: licence, setup, ongoing
   upkeep, exit. Give a range for each.
2. Count against that: saved working time and additionally closed
   enquiries. Make the assumptions visible.
3. Name the month from which the investment pays off, and the two
   assumptions that change that result most.
4. Tell me under what conditions the answer is: not yet.

Do not invent vendor prices. Where you do not know a figure, give
a range and mark it as an assumption.

In closing

Marketing automation has rarely failed on the licence. It fails on the three items nobody planned for: the tidying beforehand, getting people up to speed afterwards, and the upkeep permanently.

Count those three in and you arrive at a higher figure — and a decision that still holds two years later. Compare subscription prices alone and you will reliably choose the package that is too large and used a quarter.

Common questions

What does marketing automation cost per year?

For a one-person business with around 500 contacts the licence runs at roughly 300 to 900 a year; for a five-person firm at 1,800 to 6,000. On top come one-off setup and ongoing upkeep, which over three years usually add up to more than the licence itself.

Which costs are most often overlooked?

Four: cleaning the data before import, getting everyone up to speed, ongoing upkeep of two to six hours a month, and the exit — that is, whether contacts, histories and reporting records can be exported in full.

When does marketing automation pay off?

For small companies, typically after nine to eighteen months. It has paid off once saved working time plus additionally closed enquiries exceed total cost. The larger lever is nearly always the enquiries nobody used to follow up.

Why is pricing by contact count a problem?

Because cost grows with the length of the list rather than with the benefit. A quarter to a third of a grown list usually consists of contacts that have not opened anything in years. They cost licence fees, lower the delivery rate and distort every report — so regular cleaning is directly worth money.

Is a cheap tool or a large system the better buy?

In most small companies, only the scope of the smallest tier of a large system ends up being used long-term. A smaller tool used fully is therefore almost always the better deal — provided your own data can be exported completely.

Marketing that sets itself up

The Studio Engine beta is live. Claim your spot and help shape it from the start.

Join the beta →
← Back to overview