How to Tell If Your Marketing Is Actually Working

You are four or five months into a marketing engagement. You are paying every month. Traffic looks like it might be up. Somebody sends you a report with green arrows on it, and you still cannot answer the one question you actually care about, which is whether any of this turns into revenue.

I have been on the other side of that question more than once. The most useful thing I can do is tell you how I answered it the last time somebody asked me directly.

The conversation I had at month four

Redbridge hired us in 2023. About four or five months in, someone on their team asked me flatly: how big of an impact do you think you have actually made?

I could have pointed at impressions. I could have shown them a chart of interactions climbing and let them draw their own conclusion, which is what a lot of agencies would have done and what the report format quietly encourages. Instead I told them I could show them the number of conversations happening and the number of people engaging, but I could not show them closed revenue yet, and it would probably be six to eight months before they saw that needle move.

Their average sales cycle at the time ran over 300 days. Nothing we did in month four was going to close in month four. The deals we were influencing had not finished being deals.

So we leaned into an education and information based approach to their sales cycle. Case studies, articles, campaigns that handed their buyers something useful before anyone asked for a meeting.

Their average deal size when we started that work was around $280,000. It is now over $650,000, and it is still climbing.

At month four, the honest answer was that I could not prove it yet. The number that eventually moved was not one anybody had been watching.

Why B2B breaks the way you are used to measuring

If you have ever sold to consumers, you got used to fast feedback. Someone becomes aware of you, bounces around the consideration phase for a while, gets hit on a few different platforms, and converts. The whole arc can run in weeks. By month two or three you know, because people have either added to cart and checked out or they have not.

B2B does not behave that way. Some software companies close fast. Plenty of other businesses do not, and if you are selling something expensive to a committee, you are in the second group. Redbridge sat above 300 days when we started, and that is not unusual for considered purchases with several people signing off.

Then there is the part that makes this genuinely hard to measure.

Which means most of the work that decides whether you win happens somewhere your analytics cannot see it, months before the revenue lands in your reporting. You are trying to measure a room you were never allowed into.

Here is what I think is fair to expect.

And if your marketing is new, months two and three go mostly to A/B testing and gathering data so you know what to lean into, which feels like a stall from the outside and is not one.

What I look at first when I get access to a new account

People assume I open Google Analytics first. I do not. There are four things I want to see, roughly in this order.

Frequently none of this exists yet, and clients have no idea where to find any of it. That happens more often than not, and it is not a mark against anyone. It just means the first stretch of work is building the measurement before we start making claims about results.

The numbers I end up talking clients out of

It is social media metrics, almost every time.

People are still caught up in followers, likes, and engagement counts, and I understand why. Those numbers are visible, they update constantly, and they feel like progress.

Here is what the actual analytics show. Paid social clicks to a website tend to be the worst performing audience on the entire site. Nine times out of ten they carry the highest bounce rate of any source. Someone taps an ad, lands on your page, and leaves.

Social still earns its place for specific jobs. It puts your product in front of people who have never heard of you. It works for retargeting someone who already visited your site. And it genuinely performs for consumer goods, things like clothing and jewelry where the purchase is visual and quick.

What I redirect people toward is where the engaged traffic is actually coming from. In the accounts we manage, AI search referrals and organic search consistently produce the highest engaged sessions of any source. Those visitors show up having already done some of their homework. They stay longer, read more, and convert better than anyone who clicked an ad on a social feed.

That is the trade I try to get clients to make. Fewer clicks that flatter the report, more attention paid to being findable by people who are already looking.

When there is no clean answer to “where did this lead come from”

Sometimes there just is not one, usually because the tracking was never set up to answer the question.

The fix is unglamorous. Set up UTMs through Google so your analytics can tell you definitively which landing page someone hit and how they got there, whether that was an email campaign, a mailer, a social post, or organic search. Once that exists, the question stops being a debate.

Some of it is also reasonable inference, and that is fine. If we built a location specific landing page for a client and never ran a single ad to it, traffic showing up there is almost certainly organic. You do not need perfect attribution to make a good decision.

What you should not accept is an agency treating “we cannot really tell” as a permanent condition. If nobody can tell you where your leads come from six months in, that is a setup problem, and setup problems get fixed.

How to read a report you did not write

We took over an account where the client had been paying a dedicated SEO company for a while. Every month they were told their rankings were climbing, and the reports backed it up.

The number they were being shown was an average rank across all searches. The client’s own brand name ranked first or second, which is normal, because someone searching your company by name and finding you is not an achievement. Their actual products and services ranked somewhere around 46th. Blend those together into one average and the line on the chart goes up and to the right.

Nothing in that report was technically false. It just was not useful, and it was hiding the only thing worth knowing.

I have inherited good reports too, and what makes them good is never the design. It is that somebody sat down with the client at the start and agreed which numbers mattered to that specific business, so the report has a defined goal and something to compare against. Even when I think the chosen metrics are slightly off, that structure gives us something to work with.

The question worth asking your agency is short: what exactly is being averaged here, and what does this number look like broken apart?

If you only have ten minutes a month

Then you need one dashboard, in one place.

If you are paying an agency, building it is their job. It might be a project checklist showing what is in flight, what is upcoming, and what shipped. It might be a live view of ad spend, traffic, engagement, leads, and open tasks. The format matters far less than the fact that there is a single place you can open without emailing anyone for a status update.

If you are running your own marketing, get Google Analytics configured properly with the dashboards built ahead of time, so you can walk in cold and read it in ten minutes. Selling to consumers, you want the source of your sales. Selling to businesses, you want the source of your highest engagement and your conversions.

This does not need to be sophisticated. It needs to exist, and it needs to be the same place every month, so you are comparing against yourself instead of against a feeling.

The uncomfortable version

Marketing built to compound looks unimpressive at month three and obvious at month twelve. That gap is where most engagements die, and usually it dies because nobody set the expectation honestly at the start.

So set it. Ask for the baseline before the work begins. Ask what is being averaged in any number somebody shows you. Ask for one dashboard you can open on your own. Then give the work long enough to turn up in the number you actually care about.

And if someone promises you a clean revenue result at month three on a business with a 200 day sales cycle, they are either measuring something that does not matter or building a report designed to look good rather than be true. I would rather tell you at month four that I cannot prove it yet.

How this was made: this article draws on Haven’s direct work managing marketing for B2B clients with long sales cycles, including Redbridge, and on Spencer’s own notes from those engagements. The client examples and patterns here come from real accounts. Outside statistics are cited inline.