Why (and how) investing in billboard advertising sometimes makes sense

This is part I of a series on non-standard growth experiments.

Zoelle Egner, a former colleague of mine, joined Lenny Rachitsky’s podcast last week to talk about early growth at Airtable, and the episode reminded me of a bunch of things we experimented with back in the day.

In the early days, Airtable’s growth model was relatively simple. We identified a handful of “land” use cases (e.g. use cases with which we could acquire new users), and then worked like mad to expand within those organizations (often into use cases that were niche and specific to the company in question, so not useful from an acquisition perspective).

Pulling this off wasn’t easy. We had the standard channels covered, but those were only so effective. As a result, we were obsessed with finding non-obvious tactics for driving acquisition and expansion as well.

Over the next few weeks, I’ll cover some of those experiments we ran, and any lessons learned.

For the first installment, I’m going to try to make the case for why investing in out-of-home (OOH) advertising (a.k.a. billboards) might sometimes make sense.

Yes, we’re in the middle of an economic downturn. Yes, the smartest teams are cutting burn, husbanding cash and lowering their growth expectations just to survive another day. I get it. But billboard advertising can be used much more surgically, and much more effectively, than most startups realize. Here’s how.

As I laid out above, at Airtable, we were obsessed with driving word-of-mouth within specific communities, and expansion within our enterprise customers. We were also constantly trying to seem bigger than we were. More stable than we were. More trustworthy.

We had a hypothesis that billboards might help us solve this perception problem, while simultaneously driving expansion as well. But we had no idea how we could pull it off. We were a small team, with a single designer (who also had to, you know, design the actual product), and a minimal budget.

After a bit of research, we had a plan.

Our key insight was realizing that we had different needs than the average billboard buyer. We didn’t want the marquis billboards in Times Square that you had to book years in advance. We only wanted the specific billboards that were closest to our customers. And we also had no need to book billboards for months at a time, because we weren’t running some massive brand campaign that was optimized for eyeballs.

All this meant we could take advantage of remnant inventory. Most big brands run seasonal campaigns with big budgets, so they lock down the best billboards for 3–6 months at a time. As a result, OOH companies are stuck with remnant inventory — say 2 or 3 weeks long — that none of their usual buyers want. Usually, this inventory just sits there. We realized that if we could move fast enough, we could slot in and grab that inventory before it went stale.

That was the idea. Here’s how we did it:

1. We compiled the addresses of our existing high profile customers (note: not just HQs, but where our users actually worked) and target customers in the major metros with billboard coverage (NYC has the best coverage, SF and LA are decent).
2. We plotted those addresses on a map (may I recommend using Airtable and the Google Map extension!).
3. We identified the neighborhoods where our customers clustered.
4. We reached out to the big OOH companies — Lamar, Outfrount, Clear Channel — and asked for remnant inventory in those neighborhoods.
5. Each billboard had a lat/long provided, so we plotted those alongside our customers.
6. We selected billboards that were clustered around our customers’ offices. For the areas we didn’t know well, we used Google Street View to get a sense for the “read” of the “board” (that’s some OOH-insider lingo for you so you seem legit in front of the billboard sales people…thank me later).
7. Because it was remnant inventory, we had to move fast (e.g. 24 hour turnaround time between contract and artwork sign off). So we created a design that was simple and readily adaptable to different billboard shapes.
8. We then repeated steps 4 through 7 for a few months. As a result, we were consistently advertising in those core neighborhoods, just across different billboards as inventory shifted.

With this playbook, we invested a few hundred thousand dollars (I think around $200–250K) in countless billboards across SF, LA and NYC (must have been 35–40, though I honestly don’t remember exactly how many) over the course of 6 months.

The results:

• What didn’t work? No significant lift in website traffic. We tried to compare before and after (which is really the best you can do…companies will try to push people to a billboard-specific URL, but that doesn’t really work), but found no statistically significant lift.
• What did work? We heard about these billboards constantly from both existing customers and target prospective customers. The perception problem vanished, and natural expansion took off. One power user told us a story about how he went out for lunch with some colleagues in Soho (NYC), was reminded of Airtable because of the billboard outside their front door, told his colleagues all about Airtable over lunch, and had them onboarded by the time he got back at his desk. Stories like this were commonplace.

All in all, this wasn’t a slam dunk. We would have loved some clear evidence that the billboards were worth the investment…a clear increase in website traffic, a jump in our conversion rate among, I don’t know, media companies with offices on Broome Street in Soho.

But, as is often the case with early stage growth experimentation, there wasn’t enough data to prove that one way or the other.

And that is the most important lesson of all. The reality of early stage growth experimentation is that you can rarely get enough data, fast enough, to know whether or not something is truly working. You have to make bets based on intuition. And that’s why staying super close to customers is so important. For my money, this bet was worth it. And I’d do it again. But if I had just stayed inside staring at dashboards all day, I’m not sure I would have hypothesized billboards could help us in the first place, and I definitely would have had no idea what kind of impact this campaign was having on our customers’ perception of us.

So, should you endeavor to run a remnant inventory billboard campaign? I’m not sure. Billboard advertising isn’t a panacea, and it’s certainly not right for every company at every stage. But don’t write it off either. There’s a place for out-of-home advertising in the marketing toolbox of even the most B2B and technical companies out there.