How To Prove You’re Not “Information Harvesting”

Just wanted to privately share some tactics to help you counter Google’s aversion to opt-ins in your landing page MVP. It’s hard enough getting a startup idea off the ground, to have to deal with arbitrary Google Adwords funny business.

If you are using the Launch Tomorrow process, email addresses have always been seen as “currency”. Emails are something valuable a prospect can give you to indicate that they really need the product you want to offer. Moreover, when selling online, you need email opt-ins to establish a relationship. In order to contact your prospect, you need an email address.

Google disagrees–kinda. One of the quickest ways to get on Google’s bad side is what they’ve dubbed “information harvesting”. How’s that for a “made up” problem?

Information harvesting means they’re unhappy you’re collecting private emails in exchange for “not enough” in their eyes. They’re chiding you for not providing a good user experience for the Adwords searcher. Yes, of course, but UX is not the point of a landing page collecting opt-ins, now, is it?

Your riskiest assumptions are probably related to your prospects and customers. Establish empathy quickly with your target prospect, figure out what's valuable, and get your innovation into the market.

Specifically, here’s what Google says they don’t allow:

  • promotions that prompt users to initiate a purchase, download, or other commitment without first providing all relevant information and obtaining the user’s explicit consent
  • promotions that represent you, your products, or your services in a way that is not accurate, realistic, and truthful

With information harvesting, Google’s algorithms put you in the same bucket as late-nineties identity thieves. Long lost Nigerian prince relatives. And it’s up to you to prove (to a human employee of theirs) that you aren’t one, assuming you aren’t of course.

That’s what’s so uncomfortable about this policy. Personally, I try to be straight with myself and others as much as I can–all the time. So it feels unpleasant to be guilty until proven innocent. And tedious, to say the least. I do (unfortunately) know from my own experience, as well as other founders’ I’ve worked with.

In practice, though, the information harvester label will stop your landing page MVP in its tracks. Full stop.

Google suspends your entire Adwords account. Ouch.

If that happens, here are a few options:

  1. Drop Google, and move to other traffic sources for your landing page MVP
  2. Ask (without grovelling) GOOG to re-instate your account, and address each of their specific objections head on.
  3. Don’t gather emails, just sell a very low priced product–one which requires email for delivery. At least then you offset the cost of advertising in addition to getting emails.

With respect to dropping Google as a traffic source, it might not be the best source of customers for you in the long term. I’d argue it’s worth haggling with them anyway. You can get a lot of cheap, targeted market research about your value proposition from them–really quickly. They have traffic (ranked #1), and more importantly, they have more searchers than anyone else on the planet. If you want to find prospects searching for a solution to their problem, Google’s hard to beat.

About to start a greenfield project?

Have Launch Tomorrow run an in-house "riskiest assumption workshop". Remote delivery options also available. Discover where to prioritize your validation efforts, to get to market fast.

Contact Us

or call us now at:

US/Canada: +1 202 949 4478
UK: +44 773 952 7708
EU: +48 692 870 297

Asking for Google to re-instate your account is a good, albeit slow, option. Typically, the actual changes they require aren’t that big, or at least there are simple workarounds.

For example, if you are asking for an opt-in, you need to be explicit, clear, and logical on exactly why you need their email address. Explain exactly what they should expect. If you are being up-front, transparent, and honest, you’re golden. That tends to help.

Create an explainer video for your complicated new product. Make sure your audience understands it, without being overwhelmed by technical details.

or call us now at:

US/Canada: +1 202 949 4478
UK: +44 773 952 7708
EU: +48 692 870 297

You also need to be clear WHERE IT SAYS ENTER EMAIL ADDRESS:

  • how often you will email them
  • how to unsubscribe, which google calls an “option to discontinue direct communications”
  • whether you will pass on user details to other companies.

There are also a few important components which must be on your Google-friendly landing page. This is true for both search and display traffic.

For example, if you have a form on the page, you must also include a privacy policy. Google wants you to be up front about your privacy policy, to openly declare what you’re doing with user data. You can hire a lawyer to draw one up for you, or you can just grab a subscription on iubenda [aff link]. It’s a privacy policy as a SaaS app. Select all of the tools you are actually using on your website, and get a lawyer approved privacy policy in 30 minutes. That’s an easy one. Usually, having a privacy policy increases conversions anyway. They’re possibly doing you a favor.

Other bits you need on your landing page include:

  • contact details
  • a phone number, especially, will help

You may need to clarify your business model to Google. You may also need to be transparent, and explain it to your Google sourced visitors. What’s free, what isn’t, and when they can expect to pay for something. Sometimes this can cause difficulties if you are a broker or middleman.

Ultimately, it will always come down to a judgement call by Google staff, so you will always need to comply to their requests. At the same time, they do accept that you’re running a business (or starting one). It’s the question of navigating that fine balance, so that everyone is happy.

The third way to get ditch the “information harvesting” label is quite simple…sell. Don’t just collect info. Sell at a loss if you need to. Sell an ebook for $1. Sell a higher priced service (think concierge or Wizard of Oz MVP) that addresses their problem. You may be able to get away with pre-selling the product.

At that point, you will need the user’s personal details to deliver the product. More importantly, you identify the people who are actively in the market for solutions and willing to pay for them. If you do sell (and collect sensitive information like bank or credit card details), you’ll need to use SSL. Grab a free one over here.

As you can see, you have a couple of options if you get this labelled slapped onto your account. Bing has started doing the same. Rumor has it Facebook is starting to “crack down” on information harvesting.

Want to know more about experimenting with landing pages? You’ll find my book Launch Tomorrow helpful. It goes into depth with everything else you need to think about, when putting up a landing page to test out demand for a new product.

How Intel wrote a business plan without committing to one

Take a look at Intel’s business plan (typos and all), circa 1968:

The company will engage in research, development, adn manufacture, and sales of integrated electronic structures to fulfill the needs of electronic systems manufacturers. This will include thin films, thick films, semiconductor devices, and other solid state components used in hybrid and monolithic integrated structures. A variety of processes will be established, both at a laboratory and at at a production level. These include crystal growth, slicing, lapping, polishing, solid state diffusion, photolithographic masking and etching, vacuum evaporation, film deposition, assembly, packaging, and testing, as well as the development and manufacture of special processing and testing equipment required to carry out these processes. Products may include dioded, transistors, field effect devices, photo sensitive devices, integrated circuits, and subsystems commonly referred to by the phrase “large scale integration”. Principal customers for these products are expected to be the manufacturers of advanced electronic systems for communications, radar, control, and data processing. It is anticipated that many of these customers will be located outside of California. Today, Intel is a multi-billion dollar company. Note that they acknowledge uncertainty, with the word “may”.

To me, it looks as though they intentionally wanted to avoid committing to specific product type, while still ticking the box that they have a business plan. The point of business plans isn’t to have a business plan. It’s to make your assumptions explicit, so that you can test them.

In fact, according to the old Startup Genome report on why startups fail, 70% of startups fail because they try to scale up, without having verified enough of their assumptions.

Best of all, verifying an assumption doesn’t require a full business plan.

Win the fight for attention by communicating with relevance

The biggest challenge when introducing a new product is establishing a connection with your audience. Often, this is because you can’t do anything else until this is in place. This detail really hit home for me, when I went to an accelerator event in Mexico.

In this 2006 photograph, a man was receiving an intramuscular injection in his left shoulder muscle from a trained, registered nurse (RN), while his family was observing from over the nurse's shoulder.
Photographer: CDC | Source: Unsplash

One of the speakers was an immunobiologist client of mine, who’d developed a unique salmonella vaccine that could be combined with other vaccines. And it looks as though his vaccine is the only salmonella one which can do that.

I’d worked with him briefly as an innovation expert, and had a discussion about commercialization options as well as some pitch training. At the time he was struggling to see entrepreneurship as a viable route to greater impact. He felt comfortable as an inventor, and wanted to do more of that, not become a businessman.

It turned out I had unleashed a force of nature. Also drilling him in giving pithy explanations helped him hone down his message to something much more concrete for anyone who wasn’t already a fellow immunobiologist, or even a scientist. This one insight allowed him to communicate the relevance of his work to the wider public.

But more importantly, he started to believe that entrepreneurship was a viable route to greater impact. As it would force him to confront institutions that held him and other scientists back domestically.

Your riskiest assumptions are probably related to your prospects and customers. Establish empathy quickly with your target prospect, figure out what's valuable, and get your innovation into the market.

As a result of both, he’s pretty much gone from a booksy academic researcher to a serious contender in getting funding to help spread the use of his product vaccine. This is the power of relevance and empathy in an age of dwindling attention.

One of the best ways to get (and stay) relevant is to focus all of your marketing and product efforts around a client profile. In theory there are millions of ways to reach an audience; in practice, you only need to reach a specific group of people. So figure out who they are, and then just focus on them. The best way to do this is the Hero Canvas tool. Grab a copy and get a quick intro for free with my Hero Canvas course.

Why early stage virality can indicate product readiness

In the early days, when I was just polishing off the manuscript of Launch Tomorrow, I gave it to a friend who also lived and breathed startups. I specifically requested that he keep it quiet and just asked for feedback. Professionally, he was a marketer but in this case I was hoping to get some honest “tough love” from him. To make sure the book would be good.

After speaking with him in person, I dropped a pdf into gmail, and forwarded it on to him. Coincidentally, I also happened to have an early trial version of Streak installed on my gmail account, which is an app which measures email opens, now primarily used by salespeople.

Over the next week or so, it turned out 37 people had opened that email 56 times in different locations around London and Europe. This simple indicator was enough to convince me that the manuscript is definitely at least a minimum viable product. If not a bit more. There were a lot of tweaks I wanted to make, but clearly my idea audience was enjoying and using it. Even though this viral spread was accidental, ultimately I was pleased that my friend had effectively proven to me that my product was ready.

This was a special case of someone who knew me well, the fact that he forwarded it without my consent and that it was re-forwarded so many times implied that my soon-to-be released product will be able to generate word of mouth referrals when I do launch. This was particularly poignant, given that this was a B2C product. Like most impulse buys, books (on their own) tend to be low $ value products. There is little margin for error with a high customer acquisition cost, yet you need to be great at generating awareness and discoverability. So you can only use channels that have a fixed cost up front but little additional variable cost of reaching another person.

Going Viral

While virality seems “free” from a financial point of view, it’s expensive in terms of your time. The idea is to create enough product (content in my case) which people naturally want to share. Once you have their attention, you include some kind of call to action which then turns into a conversion , like a sale. Or at least a micro conversion, like getting an email subscriber.

Your riskiest assumptions are probably related to your prospects and customers. Establish empathy quickly with your target prospect, figure out what's valuable, and get your innovation into the market.

Manufacturing Engagement

The hard part is writing content good enough that people want to share it with their friends. To some extent, you can pre-test this by using what Andrew Chen’s twitter technique. He writes potential headlines as tweets, and then sees if anyone interacts with it. Once the idea is proven, via a favorite or a retweet, he uses that as a basis to write a longer piece on that topic. As a result, his growth hacking essays tend be highly focussed on his target audience’s needs. As a result of forwards, he effectively “clones” his existing audience. The content people forward tends to attract other interested in the same type of content.

Another approach is to repurpose a backlog management service like uservoice or a kanban tool like trello combined with audience interaction. You can create a backlog of writing ideas, and then have your existing audience vote on them. This way you are naturally spending your time writing things which is attractive for them. It’s effectively a vetting and prioritization system for content, similar to prioritizing features in agile software development.

If you’d like more ideas of how to experiment with growth, take a look at Your First startup experiment my book on getting you to that first experiment. De-risk your startup idea and figure out how to grow, grow, grow with Your First Startup Experiment.

Key Takeaways