Data Analysis is Only as Good as the Data

Hello, my name is Bill and I’m a wine guy with a data problem. My job at Emetry is to extract and transform our clients’ DTC data into the Emetry Software platform. This, along with their 3-Tier and consumer insights data, goes into our analytics engine which enables us to provide prescriptive analytics. Trouble is, data analysis can only be as good as the data. And, in the wine industry, good data is hard to find. What frustrates me is that I know how and why bad DTC data exists; it all starts in the tasting room. I know this because I used to be one of the culprits.

Before working in the wine data world, I spent three years working in wine hospitality as a tasting room lead and wine educator at two high-volume tasting rooms where I leveraged POS terminals to pollute data on a regular basis. Rest assured, it wasn’t on purpose. I think of it more as an act of accidental sabotage. Regardless, the fact remains, my fellow sales associates and I did a lot of damage to our employer’s customer data. Now, as I work on the other side of the terminal I have to deal with the damage done. Call it…wine data karma.

Bad Data Entry is Preventable 

In most fast-paced tasting room environments, bad data entry is inescapable. Any serious tasting room manager will tell you the priorities for sales staff are:

  1. Hospitality (brand story, education, and service)
  2. Sales (individual and team)
  3. Retention (clubs, loyalty, etc..)

Clean data entry is nowhere on the list. When data is collected, minor errors and omissions are accepted given the hectic environment. This is an inconvenient truth that thoroughly aggravates marketing and management. According to Forbes, for most wineries that have them, 70% of their DTC sales still come from the tasting room. That is a lot of consumer data collection potential lost. Since I have been on both sides, here are my recommendations for collecting cleaner consumer data in the tasting room. 

How to Collect Better Consumer Data at the POS

Give Your Staff the Right Tools 

The right hardware and software solutions can make data collection seamless and take the pressure off your front line staff. 

  • Wifi: Just do it. Yes, installing wifi on a winery property is expensive. However, when you consider many wineries are in sketchy cell locations, it can make their property seem like an oasis to a visitor. More than that, it expands an almost endless array of possibilities, including mobile POS terminals, sign up kiosks, enhanced property security, and hi-tech wine production devices/software. It makes it easier for your visitors to use social media check-ins on Facebook, Instagram, Yelp, and TripAdvisor which can be leveraged later as additional data sources. Considering its advantages, a winery’s need for good wifi coverage should rank up there with its need for a press…IMHO.
  • Software: Be “SaaSy”. Today most marketing software products are described as Software as a Service (SaaS). This means the software works in the cloud vs. as a dedicated download or install. Our product at Emetry is SaaS and we recommend all our clients adopt SaaS products wherever possible. In particular, look for products that utilize Application Programming Interfaces (APIs). API-based software enables the ability to link disparate applications together. For instance, at Emetry we link directly to Commerce7 via its API. Commerce7 also can be linked to Mailchimp and Quickbooks in much the same way. SaaS/API software like Shopify and Salesforce are now industry standard and adoption of these technologies will both extend your winery’s marketing capabilities as well as help maintain data continuity. However, more pertinent to this article, API-based transactions permit mobile POS terminals which, if used properly, can lead to more accurate and secure data. More on this later.
  • POS Software: Let the users choose. Most winery POS software is purchased as an add-on to a larger CRM package. Generally, the software evaluation/selection process involves marketing, DtC/club, and accounting managers. Usually, only after the software is purchased, do the sales staff even see the POS. In my opinion, this is a mistake. You should have sales representation evaluating the POS prior to purchase…and by representation, I mean at least one person that actually uses it on a daily basis. After all, there won’t be clubs or customers to manage if your staff struggles to sell them. The old saying about “skin in the game” applies here as well. Increasing staff participation will also increase morale and engagement in company data goals.
  • POS Devices: Go mobile! Data privacy and transparency are growing concerns for consumers. Running off with their credit card to some out-of-view stationary POS terminal is fast becoming an unacceptable practice. Bringing the POS to them is much better. Plus, mobile devices allow customers to physically view their data, which ensures accuracy as well as transparency. Using mobile terminals requires, at a minimum, a good wifi/cell connection and supporting SaaS backend. As mentioned above this setup comes with a cost, but the payoff in cleaner data and customer confidence will be well worth it.

Focus on What is Important 

Collecting high-quality data does not mean filling out every field in the customer profile. Empower your staff and focus on key information. 

  • Staffing: You get what you pay for. You are not Nike. Serena Williams is not the face of your brand. In fact, as far as your DTC goes and as mentioned above, ~70% of your brand is represented by your tasting room staff. Quality staff, with the right incentives, will care about your customers and endeavor to enter their info correctly. If their customer relationships are important to their bottom line, chances are maintaining good communication and data will be too.
  • Training: Experts make fewer mistakes. I don’t care if it’s a 30-day trial period or a part-time employee. Make sure every member of your sales staff that uses your software is thoroughly trained on it. That means there should be designated trainers and easily referenceable documentation. In my experience, I’ve never seen a winery actually do this properly. It is also my experience to never see totally clean data from a POS. Coincidence? I doubt it.
  • Anonymous Customers: Set ‘em and don’t forget ‘em. One common thing I see a lot in data is fake customers…sometimes hundreds of them! From a data perspective, it’s not easy to discern bogus customers from real ones. Fake customers, especially hundreds of them, will skew the analysis. You’re going to have anonymous customers, that’s unavoidable. Simply set them ahead of time and make sure your staff uses them exclusively. Make them as easy to search and use, and have as few as possible. I suggest, first name “POS”, last name “Customer” with the winery’s address for totally anonymous customers. It is also common practice to use similar practices for breakage, samples, etc…but remember the minimal number necessary is optimal.
  • System Notes and Flags are Not Designed for Customer Information. Tasting rooms are busy environments. It’s not uncommon for sales associates to handle multiple groups at once on a busy day. During these times completing a transaction accurately is disruptive enough…capturing more data is likely out of the question. Besides, what should they capture it with? Many wineries try to use system notes and flags, but this information is virtually useless outside of the POS. It can’t be sorted or segmented in any reasonable way. 
  • Data Entry: Focus on Key Data. Outside of the transaction, determine the few key pieces of data you want and keep your staff focused on collecting it. Here is my recommendation for minimal data collection in order. 
    1. Email (so they can be reached out to in the future)
    2. Zip code (for locational segmentation)
    3. Birth year (for age-related segmentation)
    4. First and Last Name (for identification purposes and customer experience)
  • When in doubt, use the default “POS Customer”. Capturing the above can go a long way for accurate data recording. Customers may be resistant to offering personal information.  As a rule, if you can’t get an email, stick with your POS defaults so you don’t pollute your data pool with customers you can’t reach out to. BUT use this fake customer sparingly and only as a last resort. 
  • Data Hygiene: When mistakes happen, get help. Every software vendor offers support. Some better than others. When you see issues in your data, at least try to get their help. Sometimes you might be surprised to find a quick fix. Before asking for help, be sure to do your own homework. Run reports, isolate the issues in Excel, and send them with your request. The easier you make it for the vendor, the more likely they will produce results. If you have technical staff, this is definitely a case where selecting SaaS/API vendors will pay off, as you may be able to fix your own data. In any case, don’t put it off. The longer your data goes rogue, the less useful it is.

Clean Data is Good Data

Good data leads to better analysis. Better analysis leads to more useful insights. Useful insights lead to more sales. A little planning ahead and good data management in the tasting room can go a long way towards customer segmentation and campaign goals. 

Here’s the TLDR:

  1. Choose the right hardware/software for your organization.
  2. Use mobile POS terminals.
  3. Engage with your tasting room staff on selection, training, and best practices.
  4. Set up default customers and use them when customer data cannot be collected.
  5. Periodically review and manage your data hygiene.

There, I feel better already. As if this blog post has washed away my past data entry sins. More importantly, I hope it helps you wrangle in your own tasting room data. Rob McMillan and other experts make the claim that a winery’s biggest DTC potential is e-commerce. At Emetry we believe that to be true, as well. However, an important part of acquiring new ecommerce customers comes from understanding your existing fan base. Where are they? In the tasting room, of course.

41 Comments

Leave a Reply