Does purchase intent exist in social media? and, can it be mined for leads and sales like search engine queries can?

For most marketers, purchase intent doesn’t pop into their head when talking about social media. Not a lot of people are buying stuff on Facebook, after all.

But the fact is, actionable purchase intent is expressed in social media tens of millions of times a day.  We know this is true, because we measure it.  Here are a few examples, with their monthly mention volumes:

150 million Twitter leads per month

Want more proof?

Then try our free Customer Search Engine for Social Media  and browse through the thousands of people complaining about their allergies, hungry for pizza or having issues with their college applications today.  We offer 70 pre-tested streams that can give you a feel for how widespread this opportunity is.

Here are two examples of what NeedTagger “customer opportunity streams” look like:

People complaining about issues with their computers – perfect for an IT services provider or a security software publisher:

Needs We Detect - People experiencing problems with their computer

People complaining about aches and pains (and in need of a massage, it appears):

Needs We Detect - People complaining about aches and pains

It’s clear that many of these people may be open to meeting people or businesses that could help them during their moment of need.

What’s missing is a simple, methodical way to connect all of these people in need with the people and organizations who can help them.  (cough)

Social Intent is Different than Search

The type of purchase intent expressed in social media is different than the type you see on a search engine in several ways.

First of all, the topics discussed are different. Really private stuff you might be quite comfortable looking for on Google may not be discussed as often on Twitter.  This includes private matters such as divorce, sexual preference and behavior, and unpopular political and racial biases. That said, it never ceases to amaze me how much private stuff some people are willing to share about their life, friends and family.

Second, the types of expressions that people are willing to make are different due to the conversational / public nature of social media. For example, in social media you’ll see a lot of complaining about an issue, asking for and sharing opinions, brainstorming and commenting, and requesting help on a topic – stuff you don’t typically see in search engine queries.

A third big difference is the sheer variety of social media posts that may indicate purchase intent. There’s a lot of implied intent, for example – this is stuff people say that clearly indicates they are in the market, but they aren’t being that explicit about their buying plans.  For example, “I just tore my anterior ligament” is a pretty good indicator that, “I need an orthopedic specialist”. Or, “I just got a new job in Phoenix!” is a reliable indicator that this person will be shopping for their next cable or satellite TV provider soon.

Social purchase intent ranges from the purely aspirational

to the downright actionable:

Some view these differences as a weakness of social media when it comes to signalling buyer behavior. But I disagree.

While there is some truth that you can’t just tweet and land a sale, social intent presents a unique competitive opportunity for savvy marketers who understand the non-linear nature of online customer behavior today. You just need a way to isolate the posts that matter (cough) and to learn how to engage when it’s appropriate.

Better than Search (in some ways)

In some ways, social purchase intent is actually better than search intent.

For starters, consider that some forms of purchase intent are only expressed in a social environment. For example, when people “fall out of love” with their phone company, they don’t usually signal it by searching for a competitor.

Instead, they complain in public:

You aren’t going to catch that sort of response in a search engine!

AT&T, if you are monitoring Twitter for the posts above, then you have a chance to turn the situation around. If you are Verizon, then you’ve identified a potential new account. Who’s gonna act first?

Capture Leads, Acquire New Customers – and Sell

So can you really connect with prospects, generate leads, land new accounts and sell stuff by mining social media for people in need?

Yes, you can.

For example, Marketo uses Twitter to generate leads at a fraction of the cost of traditional lead gen methods.  

NeedTagger customers do this all the time.  Listed below are four recent examples of NeedTagger customers who’ve done it.   All of these are unsolicited comments, copied from Twitter.

SaaS provider finds new prospects Using NeedTagger

SaaS provider finds new prospects Using NeedTagger

Using NeedTagger, this social media services provider landed several new clients.

Using NeedTagger, this social media services provider landed several new clients.

Using NeedTagger, this specialty beauty products supplier connected with potential customers.

Using NeedTagger, this specialty beauty products supplier connected with potential customers.

Using NeedTagger, she landed 4 new accounts

Using NeedTagger, Tracey landed 4 new accounts. By talking about her success on Twitter, she generated another lead.  The public nature of Twitter has its advantages.

Great for Lead Nurturing, Too

Social intent can also be tapped to handle middle-of-the-funnel lead nurturing.

Responding to people requesting help or opinions with quality online content is a natural way to guide people in their decision making process.

Copied below is an example of lead nurturing taken from a customer of ours who sells social media management services – note how he’s helping someone solve a technical problem, building goodwill in the process:

Screen Shot 2012-10-29 at 12.51.02 PM

Using social media in this way, you can reach more people at more engagement points than in other media.

You can also learn a lot about why people like or dislike your offerings, and how people talk about your brand.  Try getting that from a Search Engine.

A Greenfield Opportunity

But perhaps the most important advantage of tapping into this new river of social purchase intent is the fact that most marketers have not learned how to take advantage of it.

recent survey by Gleanster revealed that the vast majority of Top Performing Marketing organizations are focused on identifying purchase intention in social media.

Top Performing companies that rank monitoring social data for purchase intent as a priority

Top Performing companies that rank monitoring social data for purchase intent as a priority. Source:, Feb 2013.


Yet Gleanster’s report also noted that only 6% of top performing marketers are actually measuring purchase intent in social media.

who measures purchase intent - survey of top performers by Gleanster

who measures purchase intent – survey of top performers by Gleanster, Feb 2013

People, this is what’s known as a “greenfield opportunity”.

Game On!

By analyzing Google search queries, it is pretty clear that during the past few years, businesses have become increasingly interested in generating leads and sales from social media.  In the Google Trends chart below, note how “Twitter leads” and “Twitter sales” queries are increasing in popularity faster than “lead generation”:

NeedTagger surfaces the most valuable and actionable engagement opportunities for your business on Twitter, and making it easy for you to capture leads, land new accounts and sell.


   the Technology senior editor of The Atlantic (a publication I frequently enjoy), published an article called, The Geography of Happiness According to 10 Million Tweets

Alexis’ story led with a color-coded map of personal happiness in the United States (see below), as determined by a University of Vermont analysis of what we say on Twitter. 

The byline is punchy:

The happiest city in America is Napa, California — and the saddest all swear too much. Sorry, Louisiana, you are the saddest state. And Hawaii (shocker!) you are the happiest. 

geographic map of happiness - data provided by the university of vermont

geographic map of happiness – data provided by the university of vermont
Red states are relatively happier. Blue states are relatively less happy. Gray states are neutral.

With a decent amount of tongue-in-cheek, Alexis writes about a recent study,”The Geography of Happiness: Connecting Twitter sentiment and expression, demographics, and objective characteristics of place“.  It analyzes correlations between the words we use on Twitter, what they imply about our happiness and a slew of behavioral and demographic variables.

The study was published by The University of Vermont’s Department of Mathematics & Statistics, Vermont Complex Systems Center, Computational Story Lab, & the Vermont Advanced Computing Core.

The paper is really just a large scale statistical study – it’s not meant to be a behavioral or sociological analysis. It reinforces prior statistical correlations found between poverty and happiness, but challenges others.  For example, the paper found no relationship between obesity and happiness.

The Geography of Happiness: Connecting Twitter sentiment and expression, demographics, and objective characteristics of place

abstract of University of Vermont Research paper, The Geography of Happiness: Connecting Twitter sentiment and expression,
demographics, and objective characteristics of place


But Is It Accurate?

As you might guess, at NeedTagger we are pretty serious about understanding how people use the English language to convey emotion and intent on Twitter. While we are not PhD mathematicians, we do know a thing or two about Twitter, linguistics and statistics.

We also happen to live and work in in Pensacola, FL, near the heart of the Deeply unhappy South. Pensacola is the home of world-class natural language researchers.

And, apparently, a whole bunch of unhappy people like the ones below.

just another unhappy day on Pensacola Beach

just another unhappy day on Pensacola Beach

Now, I hate stereotypes. I don’t appreciate bad research reinforcing said stereotypes, even more.

So the Atlantic article’s punchy, stereotypical conclusion and red-vs-blue map really dumped some fuel into my furnace.  

So much for the afternoon…


What’s Wrong With This Study

While the paper’s correlations were really interesting, I saw two technical problems with the paper’s happiness scoring methodology that invalidates them. These issues include:

1.  the study relies upon individual words (n-grams) as a proxy for happiness. 

Problem is, individual words cannot tell you that much about a person’s true feelings.

To accurately determine the emotional sentiment of the author of a tweet (without talking to its author), you need to consider more signals such as emotive symbols (emoticons, etc), conversational context (are they talking to someone else?) and parts of speech/structure of the message.

Even after considering all of these additional signals, most sentiment text analytics services still suffer from ambiguity issues, mainly because they aren’t talking to the authors themselves.

In fairness, primary research may exist somewhere that proves individual words on Twitter actually do correlate with genuine human happiness in a reliable and direct way. But I couldn’t find any.

2.  the analysis ignores how different cultures use the same words to convey different feelings.

In fact, cultural language differences are completely ignored at two critical stages in the research.

Alexis also detected this problem.  As he put it,

One difficulty I have with the study is that it doesn’t take into account that people might just talk about happiness differently in some parts of the country or within some demographic groups.

That’s true. Let’s dive a little deeper into this one.


Culture and Word Choice

As most linguists and sociologists know, culture plays a huge role in language.  This is especially true when someone is conveying an emotion to another, for example when sharing an inside joke between friends (a common type of post on Twitter).

Some cultures are in-your-face about their feelings (American/New Jersey for example), while other cultures rarely express emotion in a public setting (Japan).

As I mentioned above, The University of Vermont study overlooked cultural differences at two key points in their analysis.

The first oversight happened when they selected their happiness scoring system.

To arrive at happiness scores for tweets, the team analyzed the 5,000 most common words found in Google Books, the New York Times, music lyrics and Twitter itself.  To assign a happiness index to each of the words, the authors relied upon an earlier study (abstract below), which used Amazon Mechanical Turk workers to score 10,000 words for happiness.  

That means the scoring system treats the US as a homogeneous culture, i.e., it assumes that all of us would apply the same happiness score to each word.

That might be OK, except for the fact that the demographics of Amazon Mechanical Turk do not match that of the United States.  For example:

  • Less than half of AMT workers live in the US. In fact, 40% of ‘Turkers live in India.  So the word scores are biased towards an Indian interpretation of “happiness”.
  • 20% of ‘Turkers are unemployed (vs. 8% in the US)
  • about 65-70% of American ‘Turkers are female

That’s a big disconnect.

"Positivity of the English Language" by Isabel M. Kloumann, Christopher M. Danforth, Kameron Decker Harris, Catherine A. Bliss and Peter Sheridan Dodds.

“Positivity of the English Language” by Isabel M. Kloumann, Christopher M. Danforth, Kameron Decker Harris, Catherine A. Bliss and Peter Sheridan Dodds.

The second mistake is in not viewing the US (linguistically) as a collection of distinct regional cultures which use language differently – which is much closer to reality.

This brings me to the reason I wrote this stupid post in the first place.


Why This Article Irks Me So Much

Here’s why I felt compelled to spend the rest of my afternoon researching this paper: the Southern US states received the lowest happiness ratings in the study.

Crap, here we go again…

Yes, I am aware that several interview-based studies have also arrived at the same conclusion – especially in dirt-poor rural areas of the South.

But I don’t think this particular study should be quoted as a reference for that conclusion, because regional culture plays such a huge role in the specific words used to convey emotion. And this study doesn’t take those regional differences into account.

Let me offer you an example.

I was raised on the Southern Gulf Coast by parents who were raised in Charlotte, NC.  I learned from my deeply funny father and the people I grew up with that traditional Southern humor is often deeply cynical, critical of others, full of cussing (for fun) & usually self-deprecating. This is just how many people of Scots-Irish ancestry and Southern African Americans talk with their close friends and family.

For example: do these tweets look familiar?

you might be a redneck jokes on twitter

you might be a redneck jokes on twitter

jokes shared between Kevin Hart and Shaquille Oneal on Twitter

jokes shared between Kevin Hart and Shaquille Oneal on Twitter

In other words, we often bond with each other by acknowledging our common suffering. Even when we aren’t, in fact, suffering.

For my first 20 years in business, I traveled extensively around the world. During that time, I learned that a Southern sense of humor is not considered all that funny everywhere in the US.  It’s not funny at all in many parts of the world, either!

The reality is, Southerners often use negative statements to imply a feeling of happiness or of being close enough to someone to criticize them in jest. Satire, sarcasm, call it what you will. Southern speech is chock-full of it. And that translates directly to Twitter, which is a conversational medium if there ever was one.

Due to this type of cultural bias, you are naturally going to see “unhappy” words tweeted more frequently by Southerners – especially when they are joking with each other! Does this mean Southerners are “unhappy”?

In contrast, citizens of the San Francisco Bay area have a cultural bias (some call it political correctness) that inhibits public jokes at someone else’s expense. But are San Franciscans “happier” people just because they don’t talk as much smack in public?

I’ve lived and worked for many years in both regions of the country, and I still communicate with people most days in both regions. I can attest to these differences in language. They are very real.

Alright, I’ve said enough.


Bottom-line: due to the flawed method used to score happiness, the study and the map above should be taken with a grain of salt.

Or two. Sprinkled over ribs and chicken, with some cornbread and collard greens on the side.  In fact, let me recommend a shaker for that:

redneck salt and pepper shakers

redneck salt and pepper shakers

People discuss their most important life events in social media – in public – all the time.

This reality opens the door for savvy and sensitive marketers to connect with new customers in ways never dreamed of just five years ago.

But marketing to life events in social media requires the right attitude, the right tools, honesty & sensitivity to others – plus a few new skills.


What are “life events”?

Life events are the most important things that happen to us in our lives. They include events we control, like taking a new job, plus all the events we don’t, like getting laid off.

There are several lists of life events available on the web. One of them is the Stress Scale prepared by Holmes and Rahe.  It has the advantage of weighting each type of event based upon the relative amount of physical and mental stress involved – a proxy for the relative importance in the average person’s life.

The Holmes-Rahe list for Adults is shown in the table below.

NOTE: the life events we track at NeedTagger are highlighted in bold.

Life event Life change units
Death of a spouse 100
Divorce 73
Marital separation 65
Imprisonment 63
Death of a close family member 63
Personal injury or illness 53
Marriage 50
Dismissal from work 47
Marital reconciliation 45
Retirement 45
Change in health of family member 44
Pregnancy 40
Sexual difficulties 39
Gain a new family member 39
Business readjustment 39
Change in financial state 38
Death of a close friend 37
Change to different line of work 36
Change in frequency of arguments 35
Major mortgage 32
Foreclosure of mortgage or loan 30
Change in responsibilities at work 29
Child leaving home 29
Trouble with in-laws 29
Outstanding personal achievement 28
Spouse starts or stops work 26
Begin or end school 26
Change in living conditions 25
Revision of personal habits 24
Trouble with boss 23
Change in working hours or conditions 20
Change in residence 20
Change in schools 20
Change in recreation 19
Change in church activities 19
Change in social activities 18
Minor mortgage or loan 17
Change in sleeping habits 16
Change in number of family reunions 15
Change in eating habits 15
Vacation 13
Christmas 12
Minor violation of law 11

As you can see, many life events are quite marketing-worthy, but some are delicate private matters best left alone.

Life Event Marketing

Life events are marketing gold.

Major life events – things like getting married, having a baby and changing jobs – are directly responsible for a large portion of US consumer spending. The link between life events and consumer spending is easy to understand.  Getting married? find a new home together. Having a baby? buy a new SUV.  Changing jobs? buy a new suit.

There are plenty of studies out there that prove this. For example, a recent Forrester Research study of 26,000 online households showed that consumers are 43% more likely to buy a financial product around a life event.

Marketing to people experiencing life events is as old as marketing itself. Before the term “marketing” took hold, we probably called it “helping people during their moment of need”.

Today, marketers understand the intimate connection between life events and purchasing power and have wrapped their advertising spend around them. For example, many TV ads mention or imply a connection between their product and a life event. Retirement planning, real estate, pharmaceuticals and life insurance commercials are the most vivid examples.

Many internet media companies have embraced life events as their central marketing strategy.  As iVillage VP of Sales Marketing & Client Services Michael Streefland explains:

Targeting at key life events is the very core of (our business). Women like to gather and share with one another about events in their life that are mile markers.

Life Events & Social Media

As you might guess, we love to talk about our life events in social media, and we do it all the time. We ask our friends and colleagues for advice when planning our weddings, when moving to a new city and when choosing our next college.

Clearly, many large brands want to know when we do. That’s one reason Facebook’s Timeline design excites marketers so much.

But how big is this opportunity?

To get a feel for how often we discuss life events in social media, let’s look at one example: changing your residence.

The leading marketing data firm, Experian, reports about 290,000 changes in home ownership in the US each month.  On NeedTagger, we detect about 65,000 people discussing a change in residence each month on Twitter – even though only 15% of Americans use the network so far.

NeedTagger pre tested stream: Life Events Change in Residence

It’s still early days, but from what we see every day, people are really open to discussing their life events in a public social forums like Twitter.

How to Market to Life Events in Social Media

OK, so people talk about their personal lives online.  How can we leverage it for marketing? and, should we?

First of all, let’s understand that people experiencing life events are already being marketed to, en masse. For example, a large portion of the direct marketing / mail business is dedicated to welcoming people to their new home, new car, etc.

But blanketing the country with “welcome to your new home” postcards isn’t the same act as personally reaching out to someone you don’t know on Twitter to help them.

Social media is different than other direct marketing channels in several important ways:

  • It is personal. People bare their souls all the time. For many, it is a form of self-therapy. But that doesn’t mean they want your company to know.
  • It is often intimate. One-on-one conversations can be embarrassingly private, even when held completely in public on a network like Twitter.
  • It is real time. A lot of information (and brand damage) can pass in minutes. So be careful what you say.

These characteristics provide a significant upside for marketers who can tap into life events to market their products and services:  they can gain new customer relationships for very little cost.

But the same characteristics introduce new risks.

In general, marketing to a person discussing a life event online is something you need to treat with the utmost respect and care. If not done correctly, your outreach can be viewed as creepy, in bad taste – or even threatening.

Caveats aside, did you know that life event marketing in “social media” has been going on for at least twenty years?

Before Twitter or Facebook even existed, marketers scored new business from online forums using time-proven customer prospecting methods. The same methods work today.

To see a few recent examples, check out our Pinterest gallery of companies marketing to life events on Twitter.

Best Practices

Successful social media marketers focus on reaching out to people who have specific questions or needs related to their business. Then, they meet those needs and answer their questions.

Based upon observing how thousands of people engage with prospects using NeedTagger, we’ve developed a list of best practices that we recommend using when connecting with a potential customer discussing a life event, as follows:

  1. seek to help people solve problems related to your core business (or your personal experience). Don’t try to be interesting, funny or cute.
  2. avoid canned messages (at all costs).  The primary advantage of social media is the personal touch. Lose that, and you’re better off ignoring the medium.
  3. focus on sharing information, not advice. Don’t address how someone feels, advise them on what they should do next, or provide any sort of personal advice.
  4. respect the emotional state of the author at all times.  If you are unsure, then don’t send a message.
  5. research the author’s profile and their recent messaging behavior before taking action.  Has this person used Twitter before to ask for and receive advice?  or is this really out of character for this person? if so, then the risk of backlash is greater.
  6. respect the conversational context of the post: is this person “shouting out for help” and thus will not be surprised if anyone responds? or are they speaking with a friend or engaged in a multi-party discussion? big difference!
  7. share links to helpful content with no strings attached; or, if there are strings, then provide an honest description in the post about what’s behind the link. For example, if there is a form required to obtain the information, then tell ’em this upfront.
  8. monitor and respond quickly (same-day) to any reactions, retweets and complaints generated from your outreach messages.
  9. lean on your profile/bio to help you sell. In your Twitter bio, explain who you are and what you sell. Most people will check you out if they like the information you share. You can even put a CTA (call-to-action) in there.

Getting Started

OK, let’s say you’re convinced and want to try “life event marketing” on Twitter. What do you do next?

The good news is that this is not hard to do – especially if your company sells something that can help someone solve a problem related to their life event. In this case, it will feel as natural as offering your business card to someone you just met at a dinner party.

That said, you do need to use the right tools and master a few new skills to make this new form of marketing work well for you.

For example, you will need to:

1.  Find people experiencing life events.

This is where a data mining tool like NeedTagger or Twitter Advanced Search come in handy. We currently monitor Twitter for several life events, with more on the way.

To see if we can help your business, try our Customer Search Engine (Free to use for one filtered stream).

You’ll find our Life Event filters under the “PreTested Streams” drop down menu for the industries they are most relevant to (see screenshot below).  

For example, you will find pre-tested streams for “losing weight” and “expecting a baby soon” under the “Health” industry.

PreTested Streams

We’ll be releasing more Life Event streams in the future.  Let us know if there’s one you’d like us to add.

2. Learn how to introduce yourself to someone you don’t know.  

So what do you say to someone you don’t know and who doesn’t follow you online – when they seem like they might benefit from your assistance?

To stimulate ideas, check out our list of the 10 best ways to introduce yourself to a potential customer on Twitter. In it, we summarize the best practices we’ve seen our customers use during the past year.  A lot of it is common sense for expert Twitter users, but for new users it’s definitely worth reading.

3. Be honest. And be yourself. 

This is not “mom and apple pie” advice. There are very specific things you can and should do related to “being yourself”.

First, make sure you post from an account that explains who is behind the curtain. You are talking to people about their personal life experiences, after all! The least you can do is expose who you are, too. Put your @name in the bio of your company’s outreach account.  Or post from your personal account.

Second, use the same relationship-building tactics that you are comfortable using to connect with prospects offline. To map your offline skills to their equivalent Twitter gestures, read How to Connect With Customers on Twitter Using the Skills You Already Have.

Third, don’t act like you have an answer that can help – when you don’t. You’d be surprised how many professional marketers cannot bring themselves to be honest that they cannot answer a question. Don’t ignore the original request and answer a different question that you can answer. Don’t be afraid to send them to another website for the solution. Or even to a competitor of yours.

Above all, remember that your goal is to help them with their (very personal, very important) life event – not to drive traffic. If you’re lucky, you might even generate a moment of serendipity for your potential customer. Yep, they’ll remember you for that.

4. Measure your results. 

You’ll want to track how many outreach messages you send for each life event you market to, and you’ll want to measure the clicks and shares they delivered for you.

NeedTagger’s web app includes a simple set of analytics that helps.

To track which outreach messages generated leads and sales for you, add Google Analytics tags into your landing page links.

5. A/B test your messages and landing pages. 

This one is important if you want to scale efficiently.

Here’s something that you probably don’t know that we learned by monitoring Twitter: the questions and complaints that people post online tend to fall into rather a small number of categories, and the same types of messages repeat themselves across a large population.

“People are people”, after all.

Knowing this, your goal should be to learn which outreach messages and landing pages work best for each type of situation, issue, complaint and question you see.  Then, re-use the message and landing page combos that work best with everyone else, moving forward. This is how our most experienced customers use our platform.

Once you get the hang of it, you’ll end up spending just a few minutes each day “tagging” the opportunities with pre-defined messages that work.


In summary, people discuss their life events in social media all the time. This opens the door for savvy and sensitive marketers to connect with new customers in ways never dreamed of just five years ago. To succeed at this new direct marketing technique, focus on finding people with specific needs you can meet with your type of expertise and content – and be sensitive to your prospect’s emotional state, specific situation and typical online behavior. Then, just be your helpful self.


Everything mentioned above can be done without using our software, by the way. We just make it easier to manage your customer prospecting as a measured process.

Let us know what you think about this by leaving a comment, below.