TL;DR
"Can we start with something smaller before we sign a year?" Yes, and you should. The real question is how you will know, in 90 days, whether this partner is worth a year. An audit tells you what is broken and what to fix first. A pilot tells you whether one channel can bring in qualified leads at a cost you can afford. Either should end with working tracking, accounts you own, written findings and a clear decision.
Audit or pilot: which one do you need?
They answer different questions, so pick the one that matches the question you actually have.
| Marketing audit | Pilot | |
|---|---|---|
| Question it answers | What is broken, and what should we fix first? | Can this channel bring in qualified leads at a cost we can afford? |
| Choose it when | You do not trust your numbers, or you are spending and not getting leads | Your tracking already records real buyer actions |
| What you get | Written findings and a ranked list of fixes | A measured cost per qualified lead from one channel |
| What it does not tell you | Whether a channel will work at scale | Why your other marketing is not working |
If you are not sure, do the audit first. A pilot run on broken tracking produces numbers you cannot use, and you will have paid for a test that tested nothing. A common finding in first audits is tracking that does not record the real business action. If your ad platform reports conversions but your sales team sees no leads, read how to find the tracking break before you spend on a pilot.
A combined version works well for many companies: fix the tracking and the pages in the first month, then run a small, narrow pilot on top of it in months two and three.
What a useful marketing audit covers
A useful audit is done with access to your accounts, not only from the outside. It should cover these areas:
- Tracking and conversions. Is the conversion the real business action (a demo request, a sample request, a booked meeting) or something softer like a page view? Does it fire once per real action, and does it reach your CRM?
- Accounts and ownership. Who holds admin access to your Google Ads, analytics, Search Console, domain, website and CRM? If the answer is a former agency or a former employee, that is the first fix.
- Landing pages and forms. Do the pages your ads and emails send people to load, say the same thing as the ad, and have a form that reaches a person? Is there a thank-you page that tells the buyer what happens next?
- Claims. For a device company, do your pages and ads say only what your clearance supports? The audit should flag lines for your regulatory lead to review, not rewrite your labeling for you.
- Search demand. How many people actually search for what you sell? In a narrow clinical category the number can be small, and that limits how much paid search can usefully spend.
- Past spend. Where did earlier ad money go? Spend that ran mostly on mobile apps and partner sites, or on broad keyword matches, can often be fixed in the campaign settings.
- Competitors. What are they running and saying? Google's Ads Transparency Center and a competitor's own site will show you a lot. This competitive audit guide walks through the steps.
What the audit should hand you: written findings, a list of fixes ranked by impact and effort, and a note of which fixes were already made during the audit. Keep the whole plan visible even if you only pay for the first step of it.
What does not work: an audit that is only the output of an automated tool, or one that ends in a slide deck of observations and a retainer proposal. If the findings could apply to any company, they are not findings.
What a pilot should prove
A pilot is a narrow, measured test. Keep it to one product, one audience and one or two channels, and write down what it has to show before it starts.
- Tracking works end to end. Submit a test lead yourself and follow it from the form to the CRM to the rep's inbox.
- What a qualified lead means. Agree on it in writing with sales before launch. A form fill from a student or another vendor is not a qualified lead.
- A real cost per qualified lead. Not cost per click, and not cost per conversion if the conversion is softer than a lead.
- Whether there is room to grow. If the pilot already buys most of the relevant searches, doubling the budget will mostly buy weaker clicks.
Start with the highest-intent searches only, the ones a buyer types when choosing a product like yours. Broader, informational searches are usually better served by content than by ads.
Ad spend should be paid by your company directly to Google, LinkedIn or whichever platform you use. You control it. Decide the amount after tracking is in place, when you can see real data, rather than fixing it in the contract before the work starts.
Do not change the ads every week. If everything changes at once, you do not know what you tested. Check the account often, and change one thing at a time.
An example of the break-even arithmetic
The numbers below are made up to show the method. Replace them with your own.
- Gross profit from a new customer's first year: $6,000
- Share of qualified leads that become customers: 1 in 10
- So a qualified lead is worth up to $600 to you before it loses money in year one.
If the pilot spends $4,800 on ads over its three months and brings in 12 qualified leads, the cost per qualified lead is $400. By this example's numbers, that channel is worth continuing. If it brings in 4, the cost is $1,200 per qualified lead, and you need either a cheaper channel, better pages, or a higher-value customer to make it work. Add the partner's fees to the spend if you want the full cost per lead.
What you should have at day 30, 60 and 90
Ask for this schedule in writing before you sign, and ask what decision you will be able to make at each point.
| Day | What you should have | The decision you make |
|---|---|---|
| 30 | Admin access to every account in your company's name. Tracking fixed and tested. Audit findings in writing. A written lead definition. A pilot plan naming the channel, audience and a spend range. | Run the pilot as planned, change it, or stop. |
| 60 | The pilot live for several weeks. Spend, conversions and qualified leads, checked against what sales actually received. A review of the searches that triggered your ads. Landing page changes made from what you saw. | Keep the budget, move it, or narrow the pilot. |
| 90 | A written report: cost per qualified lead, how many leads became sales opportunities, what worked, what did not, and a plan for the next six to twelve months with options at more than one budget level. | Sign a longer agreement, extend the pilot, or part ways with everything handed over. |
For many medical devices, 90 days will not show closed revenue, because the sales cycle is often longer than that. Judge the pilot on qualified leads and sales opportunities, and keep tracking those leads after day 90 to see what closes.
Questions to ask before you sign a year
Ask these before you sign anything longer than the first phase:
- What exactly will I have at day 30, 60 and 90? Get the list in the agreement.
- Who owns the accounts and the work? Your company should hold admin access to the ad accounts, analytics, domain and CRM. In Google Ads, Admin access is the level that can give other users access and change their access levels. Everything built for you should be yours if you part ways.
- Can I stop after the first phase? A partner confident in the first phase should not need to lock you into a year to keep you.
- Who does the work each day? Ask for the name of the person who will be in your accounts, and whether that is the person you met.
- How often will you check the accounts, and how often will you report to me?
- How is ad spend paid? Directly by you to the platform, or through the partner's invoice? If through the invoice, is there a markup?
- Who decides what counts as a lead? It should be you and your sales team, in writing.
- What will you not promise? A good partner will tell you plainly. Be wary of guaranteed lead counts or rankings; here is why guaranteed-lead offers are a poor deal for medtech companies.
- How do you handle claims review? For a regulated product, ask who checks copy against your clearance and at what step.
For the contract itself, the marketing agency SLA guide lists the terms worth writing down, and the retainer models guide covers how longer agreements are usually structured.
Warning signs in a first-phase proposal
- A year-long contract before any findings exist.
- Guaranteed leads, rankings or return on ad spend.
- Ad accounts opened in the partner's name, not yours.
- An audit described only as a tool report or a score.
- Reporting on clicks and impressions, with no count of qualified leads.
- No written definition of a lead, or a definition the partner sets alone.
- A pilot so broad (every product, every channel) that no single result means anything.
How Buzzbox Media starts new work
Buzzbox Media usually starts new engagements this way: a defined first phase that fixes tracking and the pages that matter, ends in written findings and a ranked plan, and leads to a decision about what comes next. The client owns the accounts, and everything built in that phase is handed over.