The Marketing KPIs Nobody Tracks

There's a particular type of marketing report that gets sent around agencies every month. It's got MQLs in it. Maybe a traffic graph going up and to the right. Cost per lead. Open rates. A pie chart, if someone's feeling fancy.

And it tells you almost nothing useful.

Not because those metrics are wrong, exactly. But because they measure activity, not intent. They tell you things happened. They don't tell you whether those things mattered. When you're selling £50k+ projects with six-month sales cycles and three people on the buying committee, "we got 200 leads this month" is about as useful as knowing how many people walked past your office.

The KPIs that actually matter for agency marketing are the ones almost nobody tracks. They're harder to measure. They don't fit neatly into a HubSpot dashboard. And they're worth ten times more than your open rate.

Here are three categories of metrics that will tell you what's really going on.

This is the big one. Most agencies track whether someone opened an email or clicked a link and call that "engagement." It isn't. It's the digital equivalent of someone glancing at your business card before putting it in a drawer.

Real engagement looks different, and you have to look harder to spot it.

Are they replying to your emails, not just opening them?

An open is nothing. A reply is a conversation. If your marketing emails are generating replies (even short ones, even "thanks, interesting") you're doing something most agencies never manage. You've said something a busy person felt was worth responding to. That's a signal worth paying attention to. If nobody ever replies to anything you send, your content is wallpaper.

Are they forwarding your emails to colleagues?

This is the one that should make your ears prick up. When someone forwards your email internally, they're doing your selling for you. They're saying to a colleague: "look at this, it's relevant to that thing we're dealing with." That's a buying committee signal. It means your content has landed with one person and they think it's important enough to bring others in. Most email platforms can track this. Almost nobody bothers.

Are they opening the same email multiple times?

Someone who opens your email once might be curious. Someone who opens it three times over a week is considering it seriously. They're coming back to it. Maybe they're showing it to someone in a meeting. Maybe they're weighing up whether to get in touch. Either way, repeat opens on the same piece of content is a much stronger signal than a single open on ten different emails.

Are they visiting multiple pages on your site?

One page visit is nothing. It's a bounce that hasn't happened yet. But if someone reads your article, then looks at your case studies, then visits your about page, then checks your services page, that's a research pattern. They're evaluating you. And if your analytics are set up properly (most agencies' aren't, but that's another article) you can see exactly which pages and in what order.

Are they coming back repeatedly over weeks or months?

This is the research phase in action. Someone who visits your site in January, comes back in March, and returns again in May is in a long buying cycle. They're not ready yet but they keep coming back to you. If you're only looking at new traffic, you'll miss this entirely. And it's one of the strongest intent signals you'll ever get, because it means you've earned a place in their consideration set without them telling you.

None of these are hard to track technically. Most CRMs and email tools can surface this data if you set them up right. The problem is that nobody thinks to look for them because they don't show up in the standard reporting templates.

Here's a question that should be simple but almost nobody can answer: which of your marketing channels produces leads that actually close?

Not leads. Not MQLs. Closed deals.

Most agencies can tell you where their leads come from. Very few can tell you which source produces leads that convert to opportunities, let alone revenue. And the difference matters enormously, because the channel that drives the most volume is very rarely the channel that drives the most value.

Which channel drives MQLs that actually convert to opportunities?

This is the first filter. You might get loads of leads from paid LinkedIn but if they all stall at the first sales conversation, that channel is expensive noise. Meanwhile, the five people who found you through a specific piece of long-form content might convert at 40%. The volume looks bad. The quality is exceptional.

Which channel drives MQLs that close?

One step further. Converting to an opportunity is one thing. Actually signing is another. Some channels produce leads who are happy to have a conversation but never commit. Others produce fewer leads who arrive with genuine intent and budget. You need to know which is which, and you can't know that without connecting your marketing data to your actual sales outcomes. This usually means getting your CRM in order, which is annoying but necessary.

Which channel drives the highest deal value?

This is where it gets really interesting. You might find that organic search drives your highest volume of closed deals, but events drive your highest average deal value. That changes how you allocate budget and effort. A channel that produces two deals a year at £150k each is worth more than one that produces ten deals at £20k. But if you're only measuring lead volume, you'd invest in the wrong one.

Which channel has the shortest sales cycle?

Time matters. If referrals close in two months but paid media leads take eight months, that's not just a cashflow issue. It tells you something about the quality of the relationship and the level of trust at the point of entry. Channels that produce faster closes are usually channels where the buyer arrives with some pre-existing confidence in you. That's worth understanding, because you can try to replicate those conditions elsewhere.

Right. Now here's the caveat that should stop you making a terrible decision with the metrics above.

A huge amount of what actually influences a buyer to get in touch with you is completely invisible to your tracking. Someone shares your article in a Slack group. A founder mentions you on a podcast. Your post gets screenshotted and sent in a WhatsApp thread. A buyer reads three of your pieces over six months in a private browser window and then types your URL in directly when they're finally ready to talk.

None of that shows up in your attribution data. None of it. As far as your CRM is concerned, that person appeared out of thin air. "Direct traffic." "No source." Maybe you give the credit to whatever they clicked last, which was probably a Google search for your company name that they only did because someone recommended you in a DM three weeks ago.

This is dark social. Dark activity. And it accounts for way more of your pipeline than most agencies realise.

The danger with source quality metrics is that someone looks at the data, sees that a particular channel "isn't generating leads," and kills it. You stop posting on LinkedIn because the attribution says it only produced two MQLs last quarter. You stop writing long-form content because nobody fills in the form at the bottom. You stop going to events because the leads don't convert fast enough.

And then six months later your pipeline dries up and nobody can work out why, because the stuff that was actually warming people up, building familiarity, making your name the one that comes to mind when a project kicks off, was all happening in places your tracking couldn't see.

Last-touch attribution is particularly dangerous here. It gives 100% of the credit to whatever happened right before the conversion, which is almost never the thing that actually did the heavy lifting. It's like giving all the credit for a goal to the striker and pretending the midfield doesn't exist.

Multi-touch attribution is better in theory. In practice it's expensive to set up, hard to maintain, and still can't see into Slack channels and WhatsApp groups. So even the "good" model is incomplete.

The honest answer is this: track your source quality metrics because they're useful directional signals. Use them to understand patterns and make smarter bets. But do not use them as a kill list. If you culled every channel and activity that your tracking says isn't generating leads, you'd eventually have no marketing at all. And the channels that are "working" in your reports would start working less too, because they were quietly being fed by all the dark activity you just switched off.

The way to handle this is to ask one simple question at the start of every sales conversation: "What made you get in touch?" Not a dropdown menu. An open text box, or better yet, an actual conversation. The answers will terrify your attribution model and illuminate your actual marketing.

This is the one agencies are worst at measuring. And it's arguably the most important, because for high-value, long-cycle work, the nurture is where deals are actually won or lost.

Most agencies have some kind of email nurture sequence. Maybe an automated drip. Maybe a newsletter. Maybe both. But very few can answer basic questions about whether any of it is working.

How many warm leads are in active nurture right now?

Start here. If you don't know the number, you don't have a nurture programme. You have a mailing list. There's a difference. Active nurture means you know who these people are, you know where they are in their buying process (roughly), and you're sending them content that's relevant to their situation. If everyone gets the same emails regardless of where they are or what they care about, that's a broadcast, not a nurture.

What percentage of nurtured leads eventually convert to opportunities?

This is your nurture conversion rate and it's the number that justifies the entire investment. A reasonable benchmark for agency marketing (selling complex, high-value work to informed buyers) is 15-25% over twelve months. If you're well below that, either your content isn't landing, your leads weren't qualified to begin with, or your nurture is too generic to be useful. If you're above it, protect whatever you're doing and do more of it.

What's the average time from first touch to opportunity?

Track this. It will probably be longer than you think, and that's fine. What matters is that you know the number, because once you know it you can start trying to shorten it. Not by being pushier. By being more useful at the right moments. If your average is nine months, you can't run a marketing programme on quarterly cycles and expect it to show results. The KPI and the strategy have to match the reality of how your buyers actually buy.

Which nurture content drives the most replies and engagement?

Not opens. Replies. Engagement. The pieces that make people respond, click through, forward to a colleague, or book a call. Knowing which content actually moves people through your nurture is how you make the whole thing better over time. Most agencies never review this because they set up the nurture sequence once and let it run forever like a slow drip of lukewarm water.

What's your re-engagement rate on cold leads?

Here's a number that's quietly powerful: of the leads that went cold (stopped engaging, didn't respond, dropped off), what percentage eventually warm back up? A reasonable target is 10-15%. That means one in seven to one in ten of the people you thought were dead come back to life. Usually because their situation changed. A new budget cycle. A project that suddenly got approved. A competitor relationship that fell apart. If your content is good and your nurture keeps running, you're there when that happens. If you stopped emailing them three months ago because they "went cold," you're not.

Average nurture time: track it, then try to shorten it.

Similar to the first-touch-to-opportunity metric, but specifically for your nurture sequence. How long does someone typically stay in nurture before they convert or drop off? This tells you whether your sequence is the right length, whether you're losing people at specific points, and where the drop-off happens. Most agencies don't know this. The ones that do have a significant advantage because they can tune the system instead of just hoping it works.

The reason most agency marketing feels like it isn't working is because people are measuring the wrong things. They're watching vanity metrics tick up while the actual commercial signals go unnoticed.

When you sell high-value, complex work, the buyer journey is long, messy, and involves multiple people. Standard lead gen metrics were designed for a simpler world. A world where someone clicks an ad, fills in a form, and buys something. That's not your world.

Your world needs KPIs that reflect how your buyers actually behave. They research for months. They share content internally before anyone picks up the phone. They go quiet and come back. They need nurturing that respects their timeline, not yours.

Track the things that tell you whether real buying behaviour is happening. Ignore the things that just tell you activity is happening. The dashboard will look less impressive. The pipeline will be more honest. And honest pipelines are the only ones worth having.

Waye helps eCommerce agencies and SaaS companies build marketing that generates and converts the right leads. If your reporting tells you everything is fine but your pipeline says otherwise, we should probably talk.