Oct 28, 2012 6 min read

Panning for Gold in the Input Stream

Panning for Gold in the Input Stream

As a product manager, you’re constantly getting input from customers, prospects, salespeople, company executives, developers, user groups, surveys, tweets, Quora posts, focus groups, competitors, friends and relatives about how to improve your product.  Easily 25 interactions a week. (If not, turn in your badge and business card on the way out.)

The product management challenge is not (simply) about finding more ways to solicit opinions. It’s about sifting through the stream of obvious, repetitive, incremental, top-of-mind suggestions to find an occasional nugget of gold. Applying deep subject expertise and strategic judgment to a chaotic mess. All while treating people respectfully, regardless of the quality of their ideas.

customer-derived wisdom

Note that your customers don’t share this reality. Each believes he has a few rare coins of wisdom for you: uniquely precious insights that combine “what they want” with “how to do it.” They also imagine a team of idle developers, keyboards at the ready, just waiting for you to turn them loose.

Instead, you manage an even scarcer resource: the daily focus of your team on the very few critical items that you’ve already committed to deliver in the coming quarters. A fundamental part of your role is to keep today’s shiny object from distracting your delivery team and costing yet another few days’ productivity. There will be new distractions tomorrow.

So you actually face three distinct challenges:

  1. Humbly and politely accepting all inputs from all comers.  Sincerely thanking people for their time and acknowledging their contributions.
  2. Rapidly sifting, sorting and disposing of the vast bulk of well-intentioned suggestions while watching for rare nuggets of gold.
  3. Contextualizing and sharing this input stream for internal audiences while relentlessly repeating that the current plan is still the current plan.

I’ll focus on #1 and #2, leaving #3 for another day.

Scaled-up input stream


So let’s subdivide this torrent based on our existing roadmap, with fake percentages representing relative frequency.  “Distilling fact from the vapor of nuance.”

87% of everything that arrives is sensible, incremental, completely obvious and brutally repetitive.  These are the small improvements, fixes, UI tweaks, feature extensions, button rearrangements and helpful hints that most software users suggest. If you’re keeping a backlog (and you should be!), you already have 50 or 150 pretty-damn-nice-to-haves and we-really-are-embarrassed-abouts and that-would-reduce-support-calls.

You hear the exact same suggestions many, many, many times. As the official gold miner and product expert, all rocks come to you for inspection.  This subdivides into

  • Already finished (15%).  Some users missed your last newsletter or release notes, so this is a quick win.
    • A good product manager responds with “That’s a great idea. In fact, we were able to slip it into Version X-1. Here’s a link to the online docs. Now tell me more about how you’re using our product…”
  • Currently underway (24%).  A good roadmap means that customers will often ask for solutions that map to high priority development work. Add that customer to your market validation list, since you’ll have to defend this story’s priority several more times before shipment.
    • “Yes, you’re right. That’s something we know is important and we’re already working on. It’s currently slotted for our Q1 release, but I’d love to have you look at some early mock-ups or include you in the beta program…”
  • Near the top of the uncommitted backlog (28%).  This item is ‘on the bubble,’ among a handful of candidates for the next release train. Or in a fully agile shop, high enough in the backlog that it has a good chance of getting worked on soon. You want to tally customer interest so that prioritization decisions are well grounded. Add this customer’s name to your user story or market validation list.
    • “Yes, that’s something we’ve heard from a number of customers. We’re working on the commitment list for our next release, and this is one that we’re considering. Tell me more about why this is important and any alternatives you’ve considered. We have some other candidates, though, and I’d love your help ranking this against five other top choices…”
By now, I hope it’s clear that we have no shortage of good ideas, even great ones. Every new item that’s selected necessarily pushes out other worthwhile ideas. Roadmap decision-making demands exclusive-or decisions such as “we can restructure that code for better performance or we can add credit card address validation, but we can’t do both right now.” In agile terms, only one story can be ranked #1, only one can be #2, and inserting a new story into position #8 inevitably pushes down #9 through #1600. Also, no development team will ever be big enough to do every worthy thing in your backlog. Life is tradeoffs.
  • Deep in the backlog, quarters or years away (19%).  These are good, valid suggestions but not as good as the hundred ahead of them. Check if your customer has new information, or a novel situation, then move on.
    • “That’s a really good point. Tell me more about your specific situation. Then let me fill you in on what’s exciting in our next release, and some third party solutions that might solve your immediate problem…”
  • Impractical (3%).  Some suggestions bump into internal technical or political realities, are things we’ve decided to live with, or are not of general interest to the core customer base. These live deep in the backlog, or perhaps move to the never-going-to-happen pile.
    • “That’s an interesting idea. We looked at it, but uncovered a series of complex architectural challenges that would make it very hard to solve directly. Tell me more about your situation, and let’s brainstorm other ways to solve the underlying problem…”
  • Just plain bad (8%).  This is the squishy stuff that collects in streambeds. It includes you-should-repurpose-your-product-for-my-niche-market and why-not-add-twenty-more-menu-items and teleportation-would-be-a-cool-feature. Suggestions we would ignore even with infinite resources.
    • “How interesting. Let’s see if we can reframe the problem you want to solve.”

If you’ve made it this far, you have the patience of a product manager. And tired shoulders from relentless panning. We’ve sifted through 97% of our input stream without uncovering any truly innovative, unexpected gold nuggets. So let’s split the last three percent among these variations of high-grade ores:

Gold nuggets
  • Truly new, sensible on its face, and fits within your product scope (1%).  Yee-hah! Thar’s gold in them thar hills! Kiss your customer, send flowers or chocolates or a free software license. Then build a justification for moving this to the top of your list. Remember, you’ll have to postpone another precious item to fit this in.
  • New, sensible on its face, but beyond your product scope (1%). Yup, gold, but many changes are bigger than your individual product team: acquiring competitors, adding product lines, changing distribution channels, targeting a new audience, aligning release trains. Mining deep veins of high-grade ore takes a big crew and executive budgets. You need to pass this concept upward, or to the right owner. FYI, most companies can only manage one or two radical restructurings at a time, so this new proposal competes with other initiatives for attention.
  • Seductive but ultimately wrong (1%). Some discoveries turn out to be iron sulfide, aka pyrite or fool’s gold. After some digging, you may find this shiny new solution fails technically, undercuts revenue, upsets customers, or otherwise just won’t work. Stay strong, keep your tools sharp, and don’t stop panning.

Finally, readers may have been shouting ‘crowdsourcing’ for the last six paragraphs. I’d contend that any useful crowdsourcing plan must include realistic assumptions about the volume and (average) quality of what comes in, and an actual assignment of skilled talent to find the nuggets. Otherwise, you’ll just learn that consumers want longer battery life and lower prices.


Like panning for gold, discovering value in market inputs is an energetic evaluation process for an ever-increasing stream of ideas.  Isolating the signal among the noise.  We have to apply strategic, thoughtful, context-sensitive, senior expertise to finding the gold nuggets among tons of rock.

Great! You’ve successfully signed up.
Welcome back! You've successfully signed in.
You've successfully subscribed to Rich Mironov's Product Bytes.
Your link has expired.
Success! Check your email for magic link to sign-in.
Success! Your billing info has been updated.
Your billing was not updated.