This is a post that I’m excited to write. I was talking to my friend Leslie, who does marketing for Google for Non-Profits. She helps non-profits use Google tools to be more successful. We were chatting about a framework that they can use to measure their non-profit more effectively.
Let’s walk through the measurement planning and implementation for a fictional non-profit called NPX (Non-Profit X).
To create a measurement plan I use a process of interviewing stakeholders and documenting data needs that I learned from Avinash Kaushik. You can read about his framework in his post on digital measurement. Thanks Avinash for all you contribute to our community.
Everything starts with the organizational mission, once we know that we can investigate the strategies and tactics used to execute that mission. All along the way we define pieces of data that helps us measure the performance.
Let’s get started.
The mission or business objective for our fake non-profit, NPX, is:
“Improve the lives of those that are in need of stable, safe housing.”
The organization aims to provide affordable, stable housing to those that can not afford it. They believe that having a safe, stable living environment leads to success.
It’s important to know the mission because that’s what you want to measure: improved lives. The last time I checked that metrics does not exist in any tool :) But we will use that to define success for the strategies and other tactics.
But how does the organization go about accomplishing that objective? We identify the strategies by interviewing the people at the top. Board members, executive, etc. It can take more time to schedule a meeting with these people than actually meeting with them :)
They have three primary strategies:
1. Fund homebuilding projects by generating donations
2. Coordinate volunteers to build houses
3. Raise awareness of our cause
Anyways, from our strategies, we identify the main tactics to execute those strategies. To make things simple let’s make all of their methods web based:
1. Collect donations via a website
2. Recruit volunteers via a website
3. Educate public about the lack of housing via an informational website
These all make sense right? I want to point out the strategies don’t really change over time. But the methods, or tactics, that NPX uses to implement those strategies change. For example, before the web NPX probably used direct mail to contact donors and generate donations. They might have used broadcast media, TV and radio, to raise awareness in the regions where they operated. The tactics were different, but the strategies the same.
Keep in mind you need to know today’s strategies & tactics and in the future. Technology changes – we in the measurement business know that! Knowing future strategies & tactics will help you plan your future measurement plan.
Now we come to the good part – creating metrics, also know as KPIs, to measure the tactics.
You don’t need a lot of KPIs for each tactic, maybe 3 to 5 for each. Here are the KPIs for the tactics we identified for NPX:
To me, identifying KPIs for donations is easy – it’s all about money! In this regard its almost like an ecommerce site: revenue and avg. donation value.
Just a note here, sometimes people will use a segmentaiton of revenue as a KPI. For example, they might use recurring revenue, or donations. But I think that’s best left for the next section, segmentation.
Now we move on to KPIs for volunteers. Remember, the strategy is to get people to volunteer and the tactic is to help people find volunteering activities in their local area and sign up via a form. There’s also a volunteering newsletter that you can sign up for.
I want to track the number of people who are willing to volunteer today and contacting NPX to volunteer at a later date. I also have a micro conversion here: people that search for a volunteer opportunity.
1. Number of Volunteers
2. Number people willing to volunteer in the future
3. Number searches for volunteer activities
The last set of KPIs are for raising cause awareness via an informational website. This reminds me a little bit of measuring “brand engagement”.
NPX has a number of digital ways that they promote awareness of their cause. They have a series of videos that are important for people to use and they have a special tool on the site where users can submit letters to congress.
1. Submit congressional letter
2. Watch video
3. Frequency & recency to the blog
There are a lot of other KPIs here, like connect with NPX via social, newsletter, etc. and share content via social media. I just didn’t have room for all of those things in the diagram.
I’d also recommend that they use my technique for tracking how many people actually read their content.
Creating KPIs is a bit harder than it looks. Not only do you need to understand the “upper” part of the measurement strategy (objectives, etc.) but also the “data landscape” and capabilities of the tools.
Once we have our KPIs we need to identify segments. Segmentation, which is the foundation of all analysis, helps us understand why our KPIs move up, down or stay the same.
For the donation strategy we want to split revenue by the marketing activity that generated the revenue. This helps us calculate an ROI. We also want to segment donation based on the donation type (amount of the donation). I also want to identify repeat donors as their behavior is _very_ different than first-time donors.
So my list of segments looks like this:
1. Marketing Activity
2. Donation Type
3. Repeat Donors
Segmenting the volunteer KPIs is similar. I want to see how different marketing campaigns perform. I also want to segment repeat volunteers from fist-time volunteers. NPX volunteer opportunities are geographically based – so let’s segment based on geographic location.
1. Marketing Activity
2. Volunteer type
3. Geographic location
Next is the cause awareness. I’m going to start by segmenting by traffic source – big surprise! I spend time and money marketing for my cause so I need to measure that. Visitor type. This is a combination of the above. I want to know if people that are converting at other goals are engaging with my content. Next I’m going to look at the content type that people are engaging with. This might be video, text, etc.
1. Marketing activity
2. Visitor type (donator, volunteer)
3. Content type
Now that we have segments it’s time to add some context to the data in the form of targets.
Targets are my favorite form of context. It’s how all almost every business evaluates their performance. Every organization, at the beginning of the year, lays out where they want to be every month or quarter (depending on how they run).
Then as the year proceeds you can compare where you’re at with where you need to be. You can also account for upcoming activities when determining if you will, or will not, make your targets.
That’s it, the plan is done.
You probably notice something – once you have a measurement plan reporting and analysis becomes almost easy. The metrics and segments you need to analyze are listed right in the plan. Sure, you might need to design some pretty reports and dome dashboards, but most everything you need is identified in the plan.
Given this fact, let’s ask the question – what is analysis?
If you meet or exceed your targets you should ask yourself “Why did it work?”
If you miss your targets you should ask yourself “Why didn’t it work?”
Many times looking at the performance of your segments, and applying additional segmentation, will lead to the answer.
Translating this plan into an analytics implementation, or a Google Analytics implementation, can be some work.
Most of the data you need for the above plan will come a fairly basic implementation of Google Analytics. Of course, the devil is always in the details. But just by tagging your pages you get lots of information.
To make reporting easy, you can track the KPIs with the Goals feature. Remember, you get 20 goals, so there are more than enough for all KPIs across the various business strategies.
Segments are a bit harder. Some segments, like Geography, are automatically generated by Google Analytics.
Other segments, like marketing channels need to be created using Campaign Tracking.
Am important segment in our plan is donation type. I recommend that non-profits use the ecommerce tracking feature to track donations. Not only will you get revenue, but you can track each donation type as a different product.
Note: If you’re using Universal Analytics the ecommerce implementation is a bit different.
Another feature that’s really important is custom dimensions (formally custom variables). This feature lets you segment users based on data from other systems. For example, when someone submits a donation you can set a custom dimension that indicates the user is a donor. When someone volunteers via the website you can identify them as a donor. Both are important segments for our measurement plan.
It takes a little extra work, but it’s well worth connecting data from your system back into analytics. It’s provides powerful segmentation.
Other Technical Considerations
Many non-profits, especially small ones, use third party systems to collect donations. These systems require cross-domain tracking to correctly attribute donations to their marketing source.
I’m not sure if they support it today, but hopefully, someday, all of these platforms will natively support cross-domain tracking.
While they’re at it, it would also be cool if these platforms automatically implemented custom dimensions (or variables) for the sites they manage.
Can you tell I want all platforms to make implementation easier :)
I’d also encourage all non-profits to look at the Remarketing feature in Google Analytics. This innovative tool let’s you segment visitors to your site and engage with them over the Google Display network, this can be a very valuable way to keep them moving towards conversion.
Read more about getting started with Google Analytics remarketing.
Reporting & Analysis
We’ve covered the features to get the measurement plan implemented. Here are a few other non-technical things you can do to make your life easier:
Use a custom dashboard to monitor KPIs day to day. If you’re looking for a template, here’s a Google Analytics Dashboard for non-profits that I created for the plan above. Remember, you need to have ecommerce implemented to track donations and goals configured to track KPIs. Plus I’m using Custom Variables to segment the user types.
Start thinking about multi-channel analysis and multi-channel funnels. Learn how to use channel groupings to get a deeper understanding of how your marketing activities work together to generate conversions.
So that’s about it. I hope this helps some non-profits plan and implement a measurement strategy that’s consistent with their mission.
And while I’m at it, thanks to all those that tireless work to help others. It’s a noble calling.