Duncker's Candle Problem

On the wall in front of you is a corkboard. On the table below it lay 3 objects:

• A Candle
• A Book of Matches
• A Box of Thumbtacks

Your mission, should you choose to accept it, is to both affix and light the Candle to the Corkboard in a way where the wax won’t drip onto the table below.

Sounds easy, right? 

This scenario is the Duncker’s Candle Problem. It’s a problem solving test to see how Functional Fixedness affects the way you solve things. 

We all face this kind of problem. At work the boss wants you to put together a promotion for a thing. At home the power will be out for a day or two and you’ve got food in a fridge that isn’t getting any cooler. Either way you only have the tools in front of you to MacGyver your way through.

So you make the plan and execute the plan. Things start to get tough but you stay the course. Little by little things get tougher but you stay the course. 

Then comes the rage quit moment.

You sit there and wonder why this has to be so hard. That there has to be a better way. That’s the moment when you’re so in the thick that you can’t see the forest for the trees. 

Funny thing, there almost always is a better way. The real test is whether you look for it or stay the course.

I’m not immune to this either. This ain’t one of those “10 Ways You Can be Perfect Like Me” articles. Rather this is a parable that will hopefully help you next time you’re stuck in the thick. 

Let’s take a look at Local Political Campaign Design.

Wait! Where are you going?!

I know it seems boring but I promise it isn’t or my name isn’t Anna Lesperance!

The Candle and Corkboard

Recently we worked on the brand for Jon Gerlach who was running for Fredericksburg City Council. We were also brought on to build out his campaign website. Enter the candle to his campaign corkboard.

‍

A normal website build contains of a few “bare minimums” to work:

• Catchy headline
• Quick blurbs about what you do
• Cool photos
• Call to Action
• Contact form/info

For a small Political Campaign you just need to add in a list of the Talking Points and a Donate Button as the Call to Action and that’s pretty much it. Granted, that’s bare minimums to work assuming you’re a bare minimum effort kind of person, but I digest.

Jon is a progressive guy with complex ideas and a unique background (lawyer/archaeologist/painter/writer) but he’s really a simple kind of guy. That was a challenge all the way through the job and it really bled through here on the website.

The Box of Thumbtacks

The challenge with the website especially was building something that, on the surface, was one of a kind but simple and easy to use. Below the surface needed to be a complex machine built to be edited and updated by multiple people of varying degrees of technical knowledge. 

Here were the extra bits on the surface: 

• Blog Feed for Press Releases, News Coverage, and Blog Entries from the Candidate
• Blog Feed for archive of newsletters
• Event Calendar that updates automatically
• Option to donate one time or monthly
• Voter Information page
• About page
• Issues page with small blurbs and individual pages for each for deeper dives

Beneath the surface we needed: 

• User Accounts
• City Door Knocking lists
• Email Newsletter
• Event Management
• Volunteer Coordination
• Tracking & Analytics

The Book of Matches

No good website would be good without content. That’s the thing that takes a mediocre site and sets is ablaze, pardon the pun.

To start us off Jon had 2 major ideas: the open letter to the voters of Fredericksburg and the Venn Diagram.

The open letter was intended to be something you see the first time you get to the page and that’s it. Next time you go to the site you just get the heroic photo “Man of Tomorrow” headlining kind of stuff you expect on a campaign website.

The Venn Diagram on the other hand was going to be an integral part of explaining this man’s hyper complex but utterly simple method of reasoning. This wasn’t some simple 2 circle jam either. No, this was a 3 ringed circus in the center of a larger big top ring. 

He did a video on it toward the end of the campaign, so I’ll just sum it up as best I can. Any issue can best be solved by looking at the big picture (recurring theme of the article, activate!) and finding a solution that solves for more than one problem.

For example, spending on Vocational Education can:

• help dropout rates
• create a more skilled workforce
• attract new business to the city
• raise median income

The outer ring is a set of guiding principles for looking at the interconnected problems. Still with me?

Make the Plan

We knew out of the gate that we could code a front page and templates for the site if needed. What we couldn’t do on our own was build a digital backend to run the campaign from scratch. 

At that time we didn’t have a relationship with a vendor who could do that. We could build out something like Copper or Monday.com to effectively run the back end but we’d be left out on a lot of the other big things. Thankfully the Campaign Manager Mitzi had a few companies she’d heard of in her time in the political arena. 

Fun Fact: Fredericksburg, Virginia elections are non-partisan. Like adamantly non-partisan. 

Fun Fact: Almost every single company offering anything politically oriented officially is partisan. Like even the software is blue or red partisan.

Joy.

Enter NationBuilder.

This is a SaaS solution that is really a one stop shop for the campaign. You can run every facet of that requirements list including the public facing website. The back end of the platform is about as user friendly as WordPress. The price point was relatively competitive. Best of all kids, non-partisan.

Pack it up. We’ve won right? Right?

In The Thick

We jumped in with a two pronged attack. Shia surprise! While the campaign was building out the backend, we built out the front end. This way the work could be done faster than either side waiting for the other to finish. It also meant we were getting real data to work with while designing and laying things out. 

The platform is a bit clunky. That sentence took many redos. That’s true on both our end and on the campaign’s but the burden was more on us, so we pressed on. 

Then the platform began doing updates because Presidential Elections were coming up and they had been overhauling things to prepare. If the updates brought a new feature, there would be a post about it. If it changed a few lines code, well, we’d figure that out the hard way. The deadline was approaching, so we pressed on.

Given the tendency to update and break we pivoted away from designing and focused on adapting one of their 10 templates that come with the platform. Given that these are maintained by the platform itself this method should protect the site from breaking with updates. Spoilers, it didn’t. Starting to see the mistake in our logic pattern?

Now we are about a month out from delivery date. The site keeps breaking. It feels like we’re putting out fires more than we are creating solutions. 

The campaign isn’t having much better luck on the admin side, but there isn’t anything we can do there. The non-partisan rule eliminated too many options. There are others that might be better, but they might not be too. They are more costly in both money and time to execute. Basically, NationBuilder is a non-negotiable now. 

I’ve only given you the 10,000 ft view here so I’m sure it seems bananas (yes I sang the song in my head so I’d spell it right) that we kept moving forward. One might say it’s a Plan Continuation Bias similar to the Sunk Cost Fallacy where we would rather stay the course even though it’s a failed plan.

The important thing to note is that we are making progress little by little. The plan is working. The process is not going anywhere near intended, but we are on schedule. We aren’t sticking to some failed plan solely out of optimism, ego, or anything of the sort.

The problem is that we’ve become blind to an important question. We knew that a problem existed. We knew what was causing the problem. We weren’t stepping back and asking why the problem had to exist.

In the Candle Problem many people attempted to solve it one of two ways:

• Using the Thumbtacks to nail the candle to the Corkboard
• Using the Matches to melt the side of the candle so it would act as an adhesive and stick it to the Corkboard

Neither of these worked mind you. Part of the requirement is that the wax can’t drip on the table below. Both solutions leave the table exposed to the candle’s eventual melting and drippage.

Start Back at One

It was at this point that, thankfully, we called a time out. Anna and I sat down and went over everything from the beginning. The whole time we kept comparing NationBuilder to our much more preferred platform, Webflow, in the same way a jilted lover compares their partner to an ex that got away. 

“If we were building in Webflow we’d be done by now.” 

“Webflow makes this one thing so much easier.”

Truth is Webflow is easier than basically anything else in it’s space, but it can’t do the back end stuff we needed. You needed to work with other software options that integrate with it. NationBuilder is not one of those. /*Insert bitter grumbles here.*/ 

Basically those integrations would take over the site whenever a User was accessing their profile. It would still feel like the same site even though it would be something like Memberstack doing the heavy lifting and oh your God! That’s it!

We were so stuck on not being able to use Webflow that we didn’t truly ask why we couldn’t. By itself we couldn’t but we could create our own integration, from a certain point of view.

What’s in the Box?

This is the kind of problem solving that the Candle Problem is testing for. Ready for the solution to it?

• Take the Thumbtacks out of the box they come in
• Thumbtack the Box to the Corkboard
• Place the Candle inside the Box
• Use the Matches to light the Candle

‍

Yeah. It’s that easy. Sometimes you just have to step back and reassess the situation. We were so focused on the problems that were popping up with the pieces in play that we lost sight of the fact that we had the power to define the pieces. It’s super easy to get so hyper focused on trees that you can’t see the forest.

What we did was build an infinitely better website in Webflow. Along the way we added a few things like bilingual translations powered by machine learning using Weglot. The Blog Feeds were now super user friendly. Updates to the site took anywhere from minutes to hours vs days. Many of those small copy changes were also editable by the Communications Manager of the Campaign.

NationBuilder was then used for anything that was User oriented, like donation history or volunteer info. To differentiate those pages from the Webflow ones we treated them like a dashboard. With that kind of treatment it’s easier to just “get” that you are supposed to interact with the page differently. 

This change gave us a lot more freedom to design and build out more of the brand. Even though the public and private pages looked different, they shared a unified visual style so they felt like they belonged. The sauce was the same; it was just made in two different plants.

The campaign didn’t completely get the benefits at first. That is until COVID-19 forced massive changes in the election and the Voter Information page needed updating. A few hours after they sent me the copy to change I sent them the test link. As soon as I told them it was already to go live, they got it. The speed of developing on Webflow is a game changer.

Oh yeah, we delivered the goods slightly ahead of schedule too.

Sometimes there are two more steps than just Make the Plan and Execute the Plan. I’ll let Leonard Snart explain:

There are only four rules you need to remember: make the plan, execute the plan, expect the plan to go off the rails, throw away the plan.
Follow my lead and you'll be fine.