‘Digital ecosystem’ is a phrase that is often mis-used or over-used in our industry – but examined closely and used correctly, the concept is useful and apt.
Having studied animal communication, behaviour and habitats, I like to understand the world from an ecological perspective. This translates very well to understanding digital ecosystems. We can map the key elements of a biological ecosystem to those online, and use ecological research methods to analyse them.
- Habitats – broad and niche platforms and owned, bought and earned media;
- Communities – a community of parents in the habitat of MumsNet;
- Populations – your target audience, your staff;
- Predators – trolls and detractors;
- Competitors – that other brand;
- Resources – useful content or a budget.
Just as a biologist might study the migration of a population of birds, we can study the migration of customers from Facebook to SnapChat in digital and use that to our advantage.
We can also study the individual needs of a population in relation to the resources it requires, for example what content they engage with the most, and those needs change depending on where the population is at that time. When on your website, they will be looking for a very different interaction with you than when they’re talking about you on Twitter. The ecosystem model allows us to focus on exactly what area we’re working with. This can help us to write better briefs, better proposals, and to deliver work that does what it needs to do, without trying to cover everything in one project.
Audience researchCampaigns are more relevant to your audience because you know them better.Audience researchCampaigns are more relevant to your audience because you know them better.Audience researchCampaigns are more relevant to your audience because you know them better.
Crisis managementReduced risk due to ability to act early in a crisis, when alerted to rising online discussion of an issue.Crisis managementReduced risk due to ability to act early in a crisis, when alerted to rising online discussion of an issue.
|Work stream||Example benefits (there are more|
|Audience research||Campaigns are more relevant to your audience because you know them better.|
|Buzz monitoring||SEO benefit of using folksonomy rather than corporate taxonomy.|
|Social customer service||Improved relationship with customers due to ability to proactively deliver customer service.|
|Crisis management||Reduced risk due to ability to act early in a crisis, when alerted to rising online discussion of an issue.|
|Influencer engagement||Messaging has more kudos when influencers advocate your brand.|
|Content audit / calendar / production / publishing||Deliver an integrated cross platform content strategy as you are able to spot gaps and opportunities.|
|Competitor monitoring / analysis||Learn from competitor successes and failures.|
|Customer decision journey mapping to platforms / cross platform strategy||Deliver appropriate content on the platforms most suited to the customer at certain stages of the CDJ.|
We took the above theory and mapped different parts of the customer decision journey to the various spaces inhabited by potential and existing customers of our client’s brand. This CDJ platform mapping helps our client to focus on the purpose of each platform, in order to create content that’s most likely to resonate with the customer.
The concept can be taken beyond digital. Digital marketing exists within a wider environment which of course includes offline, and considering how customers move in and out of on/offline, or exist in both spaces at the same time (e.g. browsing online using a mobile in-store), is crucial to planning an integrated digital strategy.
What’s next for digital ecosystems? In addition to the research projects mentioned above, I believe there is huge potential to take proper scientific data models used in ecological research, and apply them to digital data sets – about digital populations, habitats or resources. An ecological approach to digital marketing data could unveil insights new to the industry, without us needing to devise statistical models from scratch.