Before we begin, let’s run through a few caveats.
Econsultancy’s brand terms are fairly narrow.
If you’re a business with multiple brand names and terms (let’s say someone like The BBC as an example), then you’ll definitely need to invest in an enterprise level monitoring solution (luckily we have a guide to help you, isn’t that handy?), but if you’re a small to medium business or someone in the C-suite who just wants a general idea of what is going on, then this should be more than enough to get you started.
Step 1: Fire up your system of choice
There are lots of sytems available, but for this I’d recommend either Tweetdeck or Hootsuite. I actually use both, I like Hootsuite for scheduling but I just prefer Tweetdeck’s look and API controls for regular monitoring. Either works.
In addition to @Econsultancy, I also have my own account, and accounts for @ComeToJUMP, @The_Digitals, @newmediaage and a couple of random test accounts running, so I have a LOT of columns.
Don’t panic, you won’t need all of these!
Step 2: Insert your automatic core columns.
When you add a new Twitter account to either of these tools, they’ll give you the option to add some ‘core’ columns.
Usually mentions, all friends, and Direct messages. Bang these in to start with.
Hootsuite also lets you add ‘sent messages’ and ‘sent DMs’.
Think about these. Do you really, really need them?
Step 3: Add primary term searches.
Next up, add a search column for your brand term.
While Twitter’s ‘Interactions’ or your ‘mentions’ column will get most of these, they will miss a few out (unless you have access to the full Twitter firehose).
Running a search column next to this will show you more conversations.
In this screenshot you can see how each column is displaying slightly different information:
Step 4: Isolate conversations for reply.
Feedback and engagement are your homeboys on Twitter, so while mentions and retweets are great, you’ll also want to filter out real conversations and questions from your audience.
Luckily, Twitter search recognises a number of qualifying terms, including the “-“ (minus) sign. So if I add a search that looks like this:
“@Econsultancy” ? -RT –http
Twitter will remove all of the RTs and any tweets with links attached from this column.
While it isn’t complete, I’ll start to see questions in isolation:
Now I can quickly and easily up my engagement.
This is especially useful for us as we get a high volume of retweets (which is great by the way – keep it up!), and sometimes these can get lost in the mix.
We have a global audience too so this column means I can quickly review tweets from the previous night when I get into the office and respond if needed; a late response is definitely better than no response at all.
Step 5: Tag, you’re it.
You got it – it’s #hashtag time. I’ve got mine arranged by account and group (You should definitely be using #TheDigitals more often by the way…), and tend to add in extras as I go along.
I monitor a lot of events that often overlap, so don’t forget that you can view more than one in a column if you find that your dashboard is struggling to keep up with too many columns:
“#ComeTOJUMP” OR “#TheDigitals”
It’s also worth thinking about searching for your brand term as a hashtag.
It’s surprising how many people will use these (Especially if they can’t recall or find your @name).
We also have a slightly confusing brand name. We’re Econsultancy, and while Twitter isn’t case sensitive, it can’t pick up variations like “E-consultancy” (we also get e-consultancy, eConsultancy, e-Consultancy…), so think about possible variations people might use and add them in as well:
Step 6: Get your house in order.
We’ve added a LOT of columns, and you might find that your platform is struggling a little to keep up.
At this point, stop and have a think about which columns you really need.
As an example, many brands might not actually require their ‘all friends’ column, especially if you use Twitter primarily as a customer service channel, so get rid of anything you don’t use.
In Hootsuite, organise the search streams by Twitter account or function to save time:
Step 7: Reach out and touch someone.
Finally, let’s think about outreach.
We already know that Twitter responds to search qualifiers, so we can start to think about ways to reach potential customers, and do a little trend spotting and general social listening in addition to our monitoring.
Econsultancy run a lot of training courses, so this might be a decent place for us to start.
Just “training” will be far too busy, so we need to narrow it down so we can be more relevant. Let’s go by course type – social media for example.
Here are results for:
social, media, ”social media”, help, OR anyone, OR suggest, OR need, OR training OR how ? -http –www
Still slightly random, but we can continue to narrow that down easily enough.
Adding in location will help, and more query words.
Again, using “-http” at the end is stripping out links so we aren’t seeing other training orgs who send tweets like ‘need help with social media training? Visit dodgysite.com’ .
Using a question mark at the end of that search query also limits the search to people asking a question.
Once you’ve settled on a query string you can change out the main descriptors for different results. So we could try:
B2B OR marketing AND anyone, OR suggest, OR help, OR how OR do OR I, OR BtoB Marketing, OR advice, OR need OR training ? –http
Or tweak it for a different type of content and go with:
“Content marketing” AND “B2B” OR “Examples” OR “tips” OR “Help” OR “Ideas” OR “Strategy” OR “Trends” OR “Case study” OR “research” ? –http
There’s room there for us to spot people looking for stats, reports, training, events, blog posts and more, and while we can’t reply to them all we can step in periodically and answer questions in a way that really help build our reputation and inspire conversation.
Here we’re using Twitter as a data store rather than a stream, so even if there’s a delay in your response, you’ll still be able to contact some promising leads.
Finally, a bit of social listening. Here’s a search I stole from @Lakey:
“love this” OR “amazing” OR “fantastic” AND marketing http -youtube-rt
Lots of amazing links to content people really love that you can share.
To get an (rather sobering) idea of how useful and powerful this kind of search is for research, replace ‘love this/amazing/fantastic’ with ‘hate/terrible/disgusting’ and ‘marketing’ with ‘war’ and compare the results…
This is just the start, but combined it should give you a realistic overview of your brand and market.
The searches I’ve described here are the simplest iterations, so think about query strings and check trends regularly. Remember, people don’t search on social the way they do on search engines. In many cases, they are looking for very specific information.
You can also add competitor terms in on top to expand this, and best of all it won’t cost you a penny (but will suck up your attention…).
Think about the kinds of queries that relate to your industry at large and the type of content you have available.
This is micro-targeted content marketing, and while it won’t have the same impact as your press materials, it has the advantage of being 100% relevant every time.
I’m always interested in learning more about monitoring, so if you’ve got any other top tips for Twitter (or social in general) then please do let me know in the comments.