In this interview we go into the detail behind Spill's recent ad campaign that went out across London. We also take a look at the results which led the Spill team to rebuild their product from scratch.
It's interviews like this one that made me create How the F*ck.
For those of you who don't know, Spill creates a platform that offers remote therapy and wellbeing tools to the employees of companies like Rightmove, Curve, and Monzo.
In this interview, we deep-dived into how Will Allen-Mersh, Partner at Spill, ideated, organised, and executed an ad campaign with a budget of £100,000 in London.
I saw this ad campaign on billboards, bus stops, the London Underground and the cover of Time Out magazine, so I was super interested to know the process behind it.
It was particularly interesting that, despite Spill receiving a sustained high number of inquiries after the campaign, their sales close rate stayed the same. Spill rebuilt the product in the following couple of months because the campaign revealed the product wasn't solving a real problem for their target customer - and their sales have skyrocketed.
I won't say any more, but the campaign was an overall success for the Spill team, and as Spill's co-founder Calvin Benton puts it, "of all the money we've spent over the last two years, although it didn't bring the highest ROI, I would spend it again in a heartbeat."
Let’s start at the beginning, why did you decide to do an advertising campaign for Spill?
From the beginning this was a brand-building campaign first and foremost. We were trying to solve two problems, outlined here in the deck used to sell the concept when I first suggested doing a big campaign:
Many of the brands we are obsessed with at Spill were ones that issued a rallying cry to consumers (Impossible Foods, for example). Even people who didn’t engage with their products often started interacting with brands like that because they believed in the kind of future they outlined.
So, after some heated company arguments, we decided to spend £100k on our first proper ad campaign: a long-term, brand-building campaign with the aim of promoting our bigger mission rather than trying to generate short-term sales for Spill.
This is the big shift in the perception of therapy that we’re trying to accelerate:
How did you set the budget needed to achieve those goals?
We’d read Binet and Field’s IPA paper on media effectiveness — practically a bible in the ad industry — which shows that long-term success is most likely when a company splits its marketing spending 60:40 on brand building activity vs. short-term sales generation activity (Read their full brand building manual here). That split roughly accounted to £100k for us, having earlier in the year secured our first round of VC funding.
Given that we were trying to change people’s perceptions on quite a big topic, we also needed enough budget to hit Londoners (our core audience) with multiple impressions from multiple channels, so the campaign felt bigger than it was. This was why we chose to go for a two-week blitz rather than an always-on slow-burner. Well, that and the Oatly campaign, which we basically modelled ours after because it’s just sheer genius.
I saw the ads around London when you ran the campaign, what was your process for deciding the channels you used?
To choose channels we again looked to Binet and Field, as their analysis examines the largest historical bank of campaigns available — over 20 years’ worth. Although TV is the most effective medium, we chose outdoor and press as they were almost as effective over the long term but without the significant creative production costs required (film shoots, sound studios, and all that stuff).
Within these, we wanted formats with a longer dwell time so we chose a four-side cover wrap of Time Out (it’s more likely to be taken home, read for longer and acted upon than other free London reads), across-platform tube ads (which have a 3min average dwell time), inside bus shelters (6min dwell time) and then some billboards, outside bus shelters and Instagram feed ads thrown in to ensure we got enough reach.
The Time Out cover wrap was the centrepiece of our media mix, and we re-jigged the other channels until we felt we had a good balance. Nurture at JCDecaux were a great partner, they’re specifically set up to help start-ups with their first outdoor campaigns. Through Nurture, we received big discounts on list prices and they also gave a lot of guidance on helping to find the best mix of media given our objectives: they suggested going for inside-facing bus shelter ads, for example. It was a bit like getting the kind of strategic advice you'd usually pay for through a media agency, but straight from the media supplier, for free.
So you’ve got your budget and your channels, how did you decide on the ad copy?
We knew we could do long copy in Time Out (we had four sides to play with after all), but for the majority of our channels we wanted to distill our thinking down to one thought to maximise the chances of it sinking in. Analysis of thousands of campaigns tested with consumers by Millward Brown on their Link database shows the importance of being single-minded:
So we wanted a single line. But in order to change perceptions, we wanted a line that would make people think again. Not a clever ‘ad-y’ headline with some wordplay or a pun, but what we ended up calling a ‘shower thought’: a perception-changing fact-disguised-as-an-opinion that makes you sit up and go “wait…huh?”. We set the bar pretty high for our inspiration…
Here is the final brief we created:
Nice! So, how did you generate the lines to fit that brief?
I spent a couple of weeks trying to write alone, which wasn’t very fruitful. Other members of the team got stuck in. Friends and ex-colleagues were consulted. We brainstormed with the whole Spill gang. We tried everything.
Before we go on, when you were generating ideas by yourself and with your team for new lines, did you use any specific techniques that helped?
We used around four techniques, which I pulled together from a bunch of random books on creativity and reverse-engineered from some ad lines that I love, were quite helpful in kick starting brainstorms with the team.
1. Make a mind map (e.g. therapy --> talking --> qualified talking --> talk with someone who talks for a living)
2. Write from a different point of view (e.g. to generate interesting lines about therapy, we could write from the POV of therapists, of the world, of the friends/partners of people who've done therapy, of the therapist's chair...)
3. Jam two words together and see where it takes you (e.g. head-sifting, feelings-cleanout)
4. Ask big and random 'what if?' questions (e.g. what if all kids had mandatory therapy? What if talking was banned? What if governments treated mental health like physical health)
We ended up with a Word document of potential lines that was 13 pages long. Some people had favourites, but we felt like we hadn’t got the line. The one that made everyone sit up and go “that’s it!”. Having stared at lines for so long, we had lost all sense of what was good. We needed help.
How did you get from there to the finished product?
We ended up turning to the LinkedIn community for help with the final decision. We put out a Typeform with eight potential lines and asked for qualitative, constructive input.
The response was overwhelming: over 400 people took the time to give their input.
We spent a whole day going through the responses one by one and found the themes were surprisingly consistent, both in terms of the general and the particular.
Two of the lines that we thought were more playful — ‘Therapy. For everyone who hasn’t completed life yet’ and ‘ If you’re alive, therapy can help with that’ — made nearly everyone think of suicide, which isn’t ideal for a tube ad. Mentioning ‘the elite’, which we thought could give a sense of an enemy to rally against, felt like it was bringing in unnecessary class conflict.
Overall, lines worked better when people felt they were being spoken to on the same level. Questions worked best at making people think, and framing things positively stopped it all feeling too doom-and-gloom.
We coded the lines based on intensity of reaction, and kept the results in a table rather than as an overall 1–5 score because we wanted to avoid lines that had a lot of ‘meh’ reactions. A line with both strong positive and strong negative reactions would at least make people talk.
Luckily, the ‘Therapy. Self-discovery without having to travel through Thailand’ line had a lot of strong positive reactions and few negative ones. People liked its knowing and humorous tone, and it made them see therapy as less of a serious treatment. Obama and Federer came in second, but the strength of the negative reactions — around them both being rich men, and so therapy seeming potentially an unnecessary luxury— made it problematic. As our second line we therefore went with a more positive rejig of the big therapy statements/questions (#4 and #5): ‘Therapy. What would the world look like if everyone did it?’
So. Cool. So, what were the results?
Unfortunately, being a startup with no previous brand awareness scores to compare against and no budget for expensive attribute tracking, we couldn't measure these things directly or concretely. Instead, we had to track a bunch of secondary and proxy measures.
Here were the results:
- 42 business enquiries in the month following the campaign (our average before this was 5 a week)
- Traffic to our website increased 300%, and stayed there for the following two months
- Moved from Page 2 to Page 1 of Google (hard for a dictionary word like Spill!)
- 350 more LinkedIn followers (+30%)
- 1,100 more Instagram followers (+50%)
- Video content we made about the campaign got over 8,000 organic impressions on LinkedIn
- Half of the products in our merchandise shop sold out the following week
- This in turn got us 7 more business enquiries
- It helped us secure a press spot on BBC Breakfast News two months later
- We were discussed on a therapists' forum in Thailand (!)
Anecdotally, though, the biggest difference is that now companies have usually heard of us when we meet with them.
Were there any weird/wild surprises for you and the Spill team from this project?
There's a phrase that has gone around the ad world for a while: "nothing kills a bad product faster than good advertising". This campaign proved it to be true: we got a big (and sustained) increase in enquiries for Spill following the campaign, but our close rate (from enquiry to sale) stayed at the same, very low, 6%. What we thought was an awareness problem was actually a problem with our core proposition: we hadn't made something that was solving a real problem for companies.
So we put sales on hold and spent December doing 'Mom Test' interviews with our target market of HR managers and COOs. We built a new product from scratch in January. In the six weeks since we launched it, we've already got more clients than we did in the previous two years with old Spill. So even though the campaign didn't necessarily fulfil its objectives, it helped us see the holes in our product early enough to rebuild it.
If you had to do the whole thing again, what would you do differently?
I would plan much more social video content to go out during and immediately after the campaign on Instagram and LinkedIn. We only had the idea to do video content during the week of the campaign itself, and it got us loads of business leads. I would also ask for help much earlier on in the process. I tried to do it all myself to begin with, which was a massive mistake, and nearly drove me insane. You can hear how that unfolded on this podcast episode about the campaign, which features audio recordings made throughout the process.
Final question, a wildcard, what advice can you give to other creative leaders during the COVID-19 epidemic?
This PDF by BBH London will be far more helpful than anything I would say.
