What Can We Learn From Hillary’s Biggest Campaign Mistake?

election

 

This election season, it doesn’t matter which way you voted, or who you most wanted to win – this was hands down the craziest election in some time, and it’s likely you looked like the woman above when the final results were in. As a data geek, I have been fascinated throughout the entire process. Once the election was over, I was eager to see the post election analysis – how could so many get it wrong and the country be so surprised with Trump’s win?

 

I wanted the dust to settle and look at all the theories before writing this. It took some time for the shock to wear off in the media so they could start looking at what happened. It looks like there is one theory out there that played a big role in Clinton’s loss, and it is a good lesson for marketers.

 

The initial thought was “How could the polls be so wrong?” Now that we’ve got some time behind us, we’re seeing they really weren’t that wrong. Clinton does lead in the popular vote by quite a significant difference. So, if the polls weren’t that off, what happened exactly?

 

In listening to a few post election analysis segments, it was interesting to me that one theory rests on simple marketing. Clinton perhaps did not do enough research into what the electorate looks like in terms of economy, demographics, and the like. It is thought that she was following the 2012 model, especially since part of her strategy appeared to be creating a third Obama term. She gauged her campaign based on the data from 2012.

 

The problem? The electorate in 2016 is not the same electorate. Things have changed, demographics have shifted, and in essence it’s a different “customer” than it was four years ago.

 

This is where Clinton may have gone wrong; she didn’t get a handle on her true “customer” and therefore, her message was not effective enough to lead to the landslide victory that many were predicting.

 

As a marketer, this is fascinating and important to remember – your customer of yesterday may not be your customer today, and by making the assumption that things are as they have been, it can be detrimental to your marketing efforts.

 

What can you, as a marketer, do to ensure you are aware of your ever changing customer and market appropriately?

 

Continually review data: take a look at your data on a regular basis. Do you see any shifts in trends? Are people spending more or less? Is there a shift to an increase in online vs in store visits? Any changes should be investigated further to learn how your customer base may be changing.

 

Keep in touch with your customers: feedback programs are not only helpful to gauge satisfaction & loyalty. They can be a key indicator of a shift in your customer base. Continually monitor feedback and make sure you are reaching new customers – don’t solely focus on those who participate in a loyalty program, for example. This may be a segment of your customer base that may be more traditional in how you see your consumer demographics, but there may be a new segment you’re not seeing.

 

Use social media research: this is another tool that can give great insight into your industry. While many brands will use social media monitoring to keep tabs on their customers, another great use is to expand the social research to your competitors and industry as a whole. See what people are saying about products/services in your industry – are their expectations, pain points, and satisfaction standards changing at all compared to what they once were? If so, you may be targeting the wrong customer base if you’re using “old” marketing.

 

This was no doubt the most painful loss Clinton has ever experienced; candidates in future elections will study her strategy and learn from it, no doubt. There were also many good lessons for marketers to learn as well. Keeping tabs on customers and realizing that changes will happen over time is one key to ensuring your marketing message stays relevant and effective down the road.

FacebookLinkedInTwitterGoogle+RedditStumbleUponDeliciousTumblrPinterestShare

Bed Bath & Beyond is No Amazon

 

bbb

It looks like Bed Bath & Beyond may be trying to take a page out of Amazon’s playbook. The retailer recently announced that it will be testing a new loyalty program, Beyond+, that is very similar to Amazon Prime.

The retailer is testing a program in which customers pay an annual fee of $29 and, in turn, receive 20% discounts and free shipping on all orders.

Right now it is invitation only; a visit to the retailer’s website indicates that it is not currently available and they are “not taking sign ups” – whether this is worded incorrectly or in an attempt to make it seem as though it’s the “next great thing” that everyone is vying for is unclear.

 

bbb2

 

So, the question is: Will this work? And, will other retailers follow suit? Is this the way of the future for retail shopping?

In short, I don’t think so. Amazon has something that no one can touch. They’ve built an ecommerce business that is impeccable when it comes to price, delivery, and service. They have fleshed out their Prime program in a way that it encompasses many benefits outside of discounts and free two day shipping. Jet tried to model themselves after Amazon, and, despite the fact that Walmart recently bought the company, they are still lagging behind.

Here are some reasons as to why this may not as successful for other retailers outside of Amazon:

  • In the case of Bed Bath & Beyond, consumers are heavily tied to their 20% coupon. But, will they latch on to this new program? My initial thought is that many will, at least the regular customers since it seems like a pretty good deal on the surface. But, they may miss out on those sporadic customers who rely on the 20% coupon to get them in the door.
  • If this model expands across multiple retailers, it will take a hit. Why? Consumers are perfectly fine having a Prime membership but shopping at other retailers as well. If the bulk of the retailers they do business do follow this model, it will be a lot of spend in annual fees – consumers may be more selective of the retailers they do business with, which may leave some retailers left out of the equation.
  • Companies may lose money. In the case of Bed Bath & Beyond, a $29 annual fee for their program seems a bit low, especially if they are targeting frequent customers. If this were to happen, would they need to increase prices to keep it working? Similarly, if it’s not as successful as they hope, it may mean a drop in customer traffic.

As a consumer, I do like the idea of paying an annual fee for discounts and free shipping (I’m an avid Prime consumer) for some of my shopping, though I think if a favorite retailer would go the path of Bed Bath and Beyond, I would carefully consider joining yet another program like this. It doesn’t mean I would stop being a customer, but it may shift my purchasing habits.
 

What are your thoughts about this new loyalty program? Good idea, bad idea? Share your comments in the space below – we’d love to hear from you!

 

One Snapshot is Good; Two Are Better

 

two-perspectives

 

If you get a high score on a mystery shop, is it safe to assume that the customer experience was similar for all customers on that day and time?

Maybe, but maybe not.

It’s not realistic to assume that each experience is exactly the same, though the thought that if one customer has an excellent experience, all the others did as well. That is the hope for businesses as they invest time and money on training and staff development.

Many businesses that utilize mystery shopping conduct one visit per month to gauge service levels. This is an excellent measurement tool and is extremely useful for this task. Others choose to conduct multiple visits at each location on a monthly basis in order to get deeper insight into a location’s performance – they realize that one shop is simply a snapshot in time and may want more data to work with.

When a client uses multiple shops per location in any given month, mystery shopping providers strive to conduct the shops across different days and times of the day in order to give the client a range of snapshots; there have been times, however, when the client chooses to get as granular as possible by conducting multiple shops at one location, all on the same day.

Consider this: a restaurant has a concern about one of its locations, particularly with the evening staff. Past mystery shop performance has not been strong during this time, and customer complaints seem to increase. To get a better sense of what may be happening, the client requests that two shoppers visit the restaurant on the same evening and around the same time (the dinner rush) to check for consistency.

Two shoppers visited the restaurant – one a bit earlier than the other, though their visits overlapped by a half hour. They sat in different areas of the restaurant and had different servers.

What did they find?

  • Each shopper had a different experience – one shop report received a 70% overall score while the other received a 95%. Taking a look at the two experiences, the client obtained some interesting information to work with:
  • The main difference between the two shops focused around timing of the meal and receiving the order. The lower scoring shop had a long wait to receive their meal, while there were no issues with the higher scoring shop.
  • The restaurant requires the manager to visit guest tables; this did not happen with the lower scoring shop. In fact, this shopper reported that the manager was not visible in the dining area at any time.
  • The server for the lower scoring shop was not able to greet the shopper within the required time frame, did not refill drinks during the meal, and dishes were not cleared in a timely manner.
  • There were other minor differences, though none existed with the host experience, the departure, or cleanliness aspects.

While this was just one night and one snapshot, it opened up some good conversation at this particular location. After looking at the staff involved with both shops, and having an open discussion based on these differences as well as data from prior shops, management learned some great information to help make improvements, such as:

  • There has been some turnover lately, and there are many new staff who are still learning the processes. Servers shared that it is difficult for newer servers working busy shifts, as the server “sections” (number of tables they are responsible for) are overwhelming for new staff. This led to a discussion of recreating sections of the dining area to accommodate new servers. With less tables to focus on during the rush, the more successful they could be.
  • Management shared that they do make every attempt to visit guest tables, but during the height of the dinner rush, they are focused on “putting out fires” and resolving issues as they come up. This takes away from table visits. The “fires” were looked at more closely to determine if they were isolated incidents that typically come up in the course of a day or if there were patterns to suggest that staff needed better training, if more staff were needed during busier times, etc.
  • Service processes were closely evaluated and adjusted to accommodate challenges presented not only in this experience, but across all experience data collected for this location.

It’s important to note that the client had concerns about this particular location before they conducted the multiple mystery shop exercise, and it was limited to one particular night, but it served as a good jumping point for closer inspection and conversation with employees to get their feedback, thoughts, and solutions.

Using multiple mystery shoppers in one day can be beneficial in many circumstances. When there are concerns, changing up a mystery shopping schedule to do a deeper dive can go a long way in uncovering challenges and improving the consistency of the customer experience.