Website Visits Are Not Sales
As part of our ongoing series on growth and measurement for CPG brands, we just spent a post taking foot traffic down a peg, while also giving it credit where it's earned for categories like QSR and retail. Today we're staying in the same neighborhood with a metric that shows up constantly in performance TV and CTV pitches: the website visit.
If you've ever bought TV media for a brand that's sold primarily in stores, you've probably heard some version of this pitch. Look how many people visited your website after seeing the ad. That's the KPI. That's the proof it worked.
It's not proof. It's interest, and interest isn't the same thing as a sale.
The setup: a KPI built for the wrong kind of brand
This one matters most for a specific kind of brand: the ones bought primarily in store, not online. Think grocery, mass, drug, club. If your product mostly moves off a physical shelf, a website visit sits several steps removed from the actual transaction, not one click away from it.
Performance TV and CTV vendors love this metric because TV has historically been hard to tie to outcomes. A website visit after ad exposure gives them something concrete to report. Something that updates fast, looks like a funnel, and feels like proof.
To be fair, it's not nothing. Someone got curious enough to look you up. That's a real signal. It's just not the signal being sold.
Curiosity vs. conversion: two very different events
A website visit tells you someone was curious. It does not tell you what happened next.
Most grocery and mass purchases are low-consideration and habitual. Someone doesn't typically pull up a laundry detergent brand's website on their phone before grabbing the bottle they always buy in the aisle. When a website visit does happen after an ad runs, it's more often someone checking a coupon, looking up ingredients, or just poking around out of curiosity, not someone building toward a purchase decision the way they might for a car or a mattress.
Curiosity is a real outcome. It's just not the outcome the KPI is being sold as.
The in-store disconnect
Here's where the metric really falls apart for brands bought at retail.
A website visit can't tell you whether that person walked into a store afterward. It can't tell you whether they bought your product, a competitor's, or nothing at all. And it can't tell you whether the visit changed their behavior in any way that mattered.
There's no closed loop here. The website visit happens in one system. The purchase, if it happens at all, happens on a shelf, at a register, weeks later, with zero connective tissue between the two events. You're left drawing a line between two dots that were never actually joined.
What does the research actually say?
There is some research on this, and it's worth being straight about what it shows and what it doesn't.
A study run by Accenture, comScore, and dunnhumbyUSA back in 2012 found that people who visited a CPG brand's website bought meaningfully more of that brand in stores than people who didn't visit, and spent more in the category overall. On the surface, that sounds like exactly the correlation the TV pitch is hoping for.
But look closer and the direction of that relationship gets murky fast. The study measured general brand website visitors, not people whose visit was triggered by a specific ad exposure. The far more likely explanation is that a brand's most engaged, already-loyal buyers are the ones who bother visiting a brand website in the first place. Heavy buyers seek out the site. The site doesn't create heavy buyers. That's correlation running in the opposite direction from the one being pitched.
None of that research isolates the specific chain performance TV vendors are asking you to trust. Someone saw the ad, then visited the site, then that visit caused an in-store purchase. That's a much narrower and much less supported claim than "website visitors tend to be good customers."
What a website visit report won't tell you
A post-exposure website visit report will not tell you:
Whether the visitor bought anything at all
Whether they bought in-store, where most of this business actually happens
Whether the visit reflects real interest or someone just checking a price
Whether any of it would have happened without the ad running
The bigger picture
A website visit is a nice-to-have. It is not good enough to be the number a media buying agency hangs a campaign's success on.
That distinction matters more than it sounds like it should. Nice-to-have metrics are fine as context, as a secondary signal, as one more data point in a fuller picture. They stop being fine the moment they get promoted to the headline KPI on a performance report for a brand that makes its money at retail.
If your TV or CTV partner is leading with website visits as the proof of impact, ask them the follow-up question they're hoping you won't. "And then what happened?" If they can't connect that visit to an actual sale, you're not looking at a growth metric. You're looking at a curiosity metric wearing a growth metric's badge.
Next up in the series, we're zooming out to talk about movement metrics as a category, foot traffic and website visits included, and why the broader habit of mistaking motion for progress is worth naming on its own. After that, we'll get into what it actually costs to measure growth properly.
We are Left Hand Agency, a CPG media buying agency helping brands grow with short and long-term strategies. Our memory-driven strategies deliver results your marketing and finance teams will champion.
