Using React with GraphQL: An Apollo Review

As enterprise AI and the Internet of Things (IoT) expand, flexibility is crucial in the software development world.

Developers need tools that help them manage a shifting network of technology while creating products that are economical to maintain.

One of the newest – with a stable release in 2016 – is GraphQL. This open source tool created by Facebook has some developers calling it “the future of APIs”.

What Is GraphQL?

GraphQL is Facebook’s query language for APIs.

It’s a syntax that outlines how to request specific data, and it’s most often used by a server to load data to a client.

In simple terms, GraphQL serves as an intermediate layer between the client and a collection of data sources.

It receives requests from the client, fetches data according to its instructions, and returns what was requested by the call.

Flexibility and specificity set GraphQL apart from other options like REST APIs.

The client can ask for the data it really wants and draw only that data from multiple sources. It pulls many resources in one call, all organized by types.

What problem does GraphQL solve?

REST APIs were a huge step forward, but they have some baggage.

Much of the data pulled never gets used, wasting time and potentially slowing an application with no payoff. REST API also use multiple calls to access separate resources.

With GraphQL, the server can query data from several hard-to-connect sources from a single endpoint and deliver it in an expected format.

It’s a standardized, straightforward way to ask for exactly what is needed.

It also solves the problem of backward compatibility. With REST APIs any changes to endpoints necessitates a version change to prevent compatibility issues.

New requirements don’t necessitate a new endpoint when using GraphQL.

React + GraphQL = Apollo

Apollo Client is a small, flexible, fully-featured client often used with GraphQL.

It has integrations for many tools including Angular and React, the latter being very popular with developers right now.

Apollo has several useful advantages. It’s simple to learn and use, so bringing teams up to speed is easy.

It can be used as the root component for state management. The client gives the calls and queries answers as props.

Plus, developers can make changes and see them reflected in the UI immediately. Apollo also features a helpful client library and good developer tools.

One of the biggest operational benefits is that Apollo is incrementally adoptable. Developers can drop it into part of an existing app without having to rebuild the entire thing.

It works with any build set-up and any GraphQL server or schema, too.

Strengths of Apollo

Being able to get complex relations with a single call- plus avoiding problems with types- are major benefits.

Apollo also offers multiple filter types, can be used as state management, and removes the need to handle HTTP calls.

With Apollo subscriptions are usually implemented with WebSockets, which is an advantage over React’s competitors.

Most importantly from an operations standpoint, Apollo is easy to learn and use.

It’s painless for team members to add it to their toolkits.

Limitations of Apollo

API are still needed for authorization and security (including tokens, JSON Web Tokens, and session management).

It’s also true that Apollo can’t go as deeply as Redux does, so when building complex apps, the tools have to be combined.


GraphQL is often compared to REST APIs, though they aren’t exactly the same thing.

REST is an API design architecture which decouples the relationship between the underlying system and the API response, much as GraphQL serves as an intermediary, but it takes a different approach.

There are multiple endpoints compared to GraphQL’s single endpoint philosophy. That adds complexity as the application scales.

REST also suffers from under – and over-fetching. Using Apollo GraphQL queries only what is needed at the time, which eliminates the problem.

Some developers like to use Relay instead of Apollo. Relay is Facebook’s open-sourced GraphQL client.

It’s heavily optimized for performance, working to reduce network traffic wherever possible. The tradeoff is that Relay is complex and hard to learn. Many find it simpler just to use.

Future Outlook

Once considered a niche technology, GraphQL is now proving its worth.

Major companies are using it in production, including Facebook, Airbnb, GitHub, and Twitter. With this much growth over just a few years, it’s a safe bet GraphQL has a long functional life ahead of it.

Wondering if GraphQL would work for your company’s next project? Set up a complimentary meeting to review your needs and find out what kind of solution we could build for you!

Request a Consultation

Is There A Place for AI in Small to Medium Businesses?

Many small to medium business owners view artificial intelligence as something only huge corporations need.

In reality, it can help position them to compete with those corporations on a whole new level.

It seems like everyone in the business world is launching artificial intelligence programs.

That’s partly because nearly everyone is. 61% of businesses have already begun using some form of artificial intelligence, many of those focusing on predictive analytics and machine learning.

71% report they plan to expand their use of predictive analytics and other AI applications over the next year.

For most companies the decision to adopt AI is an easy one.

For small to medium businesses (SMBs), though, there are tough questions to answer.

Even successful SMBs don’t have the same depth of financial resources as a multinational corporation.

They need to invest cautiously, and artificial intelligence can sound like a science fiction daydream.

That’s unfortunate, because artificial intelligence is fast becoming the kind of tool that can help small to medium businesses keep up with their larger competitors.

Read on to explore the things keeping SMBs from investing in artificial intelligence. then find out how to get past them and what technologies are best suited for small to medium businesses.

Practical Artificial Intelligence

“Artificial Intelligence” brings to mind futuristic robots and complex movie plots, but the reality is much simpler.

The term refers to teaching machines to “think” and interpret information like humans do. Humans have very flexible minds.

They can handle a variety of rapidly-changing topics and navigate difficult conditions that confuse computers (although computers have a greater ability to process repetitive data quickly and accurately).

Modern artificial intelligence has come a long way.

It can’t quite mimic human thought yet, but there have been some exciting advances using AI techniques like machine learning and deep learning that show potential for nuanced processing.

The technology is proving its value as an enterprise tool, too.

There are a few common applications that some people don’t realize are based on artificial intelligence:

  • Predictive analytics, especially embedded features in enterprise software
  • Chatbots on websites or social media pages
  • Intelligent assistants in office and productivity software
  • Recommendation engines used for suggesting Netflix titles and upselling in ecommerce

What Holds SMBs Back

Even as larger companies move to wider integrations of artificial technologies, small to medium businesses are slow to adopt.

Their hesitation is understandable – after all, a failed technology project could threaten the future of their company – but it also holds them back.

The truth is, many of their concerns aren’t as serious as they think.

The issues have practical workarounds or can otherwise be mitigated through proper planning.

Here’s why the leading reasons SMBs aren’t adopting artificial intelligence don’t have to be unmovable roadblocks to progress and how they can be overcome.

AI is too expensive

Industry news reports tend to cover high-end artificial intelligence ventures done by major international corporations, with price tags in the millions (or occasionally billions).

That kind of investment is an intimidating prospect for an SMB who just needs a better way to utilize their data.

The thing is, those programs usually involve the most difficult and expensive forms of AI.

Experimental programming, complex interactions, sensitive health information, government-regulated data, huge amounts of simultaneous users, and other complicating factors raise the costs above the average for enterprise AI projects.

SMBs don’t need the same amount of scale or infrastructure. Their modest needs can be met at a much more reasonable price point.

There is no “usual” price for AI. The costs associated with artificial intelligence are based on many factors, including safety and regulatory protocols and the complexity of necessary interactions.

To build an estimate, developers will ask questions such as:

  • Does the program need access to sensitive information?
  • Is it designed to address a specific set of circumstances or is it more a broad-spectrum tool?
  • What level of interaction with humans is desired?
  • What’s the scale involved?
  • Will the AI need to perform complex actions?

Even when a full artificial intelligence program is out of reach, there are ways to integrate AI on a limited budget.

For one thing, AI is included in many enterprise software packages. Most companies already have access to some AI tools, even if they don’t realize it.

Targeting tools in email marketing software and personal assistants on smartphones are both driven by artificial intelligence.

More in-depth AI toolsets are often available with a reasonably-priced software upgrade to enterprise level from free or lower-tier accounts.

It’s work checking with vendors to see what’s within reach.

The rise of reusable code and powerful development frameworks has put small-scale custom solutions within reach, as well.

Developers have platform options for creating analytics dashboards and chatbots that makes the costs approachable for SMBs.

AI isn’t ready for enterprise because the projects fail too often

Project failure is a daunting prospect for SMBs, who usually have a longer list of desired business improvements than they have capital to spend.

They need to prioritize projects because they can’t do everything they’d like.

Investing in AI means putting another project on hold, something they aren’t willing to do when it seems like all they hear about is failed artificial intelligence projects.

It’s easy to become discouraged by high-profile AI failures or assume tools are overhyped, because some projects do fail and some tools are overhyped.

Artificial intelligence is at a point in the Hype Cycle where its applications are being rigorously tested, and some won’t make it through to becoming everyday technologies.

However, project failure is more often an organizational issue instead of a technological one.

Projects fail for a variety of reasons, most commonly:

  • A weak discovery process results in a weak final product.
  • Internal adoption rates are too low to realize the project’s potential.
  • Misaligned business goals lead to the company creating a product that no longer fits within their workflows.
  • The company experiences an outsourcing failure or developer issues.

Avoiding these issues is somewhere small to medium businesses may have an edge over larger corporations. Why?

  • Pushing internal adoption on a small team is more effective because the company leadership can personally talk to everyone (or at least every team leader) to convince them of a project’s value.
  • There is less opportunity for confusion over business needs and goals.
  • The development process has fewer moving parts, so it’s easier to make needs clear during discovery.

What SMBs need to watch out for is the tendency to default to the lowest bidder, especially when outsourcing overseas.

If they focus on quality as much as price, they’re more likely to get a quality return on their investment.

Choosing Agile development methods is another way to ensure a positive outcome.

Developers who use Agile and conduct a thorough discovery are actually seeing a rise in project success rates, and have been for a couple of years.

AI isn’t practical for a small to medium business; it only works for massive corporations.

Many SMB owners see AI as something that can’t help their business.

They assume they don’t have enough data to process or that the impact of AI won’t be noticeable at a smaller scale.

A lot of those same owners would be surprised to realize how much data they already have – data which is going untapped.

Putting that data to work might result in smaller gains, but proportionately those gains matter more.

One interesting thing about AI is that is has opposite benefits for SMBs and larger companies.

It helps giant companies operate with the personalization of an SMB while allowing SMBs to function with the efficiency of a massive corporation.

That is, it gives small to medium businesses the edge they need to “punch outside their weight class” when it comes to competing for market share.

While there are some AI applications that won’t help smaller-scale businesses, there are many more that will.

A small bookshop with five employees wouldn’t get value from predictive scheduling software, but they could see an impressive return on predictive ordering and email marketing programs.

AI doesn’t apply to this industry

There’s a perception that artificial intelligence is only for high-tech fields like software development or banking.

That couldn’t be farther from the truth. AI can be applied anywhere where data is generated – that is, everywhere – to improve efficiency, guide decision making, and maximize the impact of marketing and sales campaigns.

Some examples:

  • A cleaning company uses AI to intelligently manage their leads and upsell current clients.
  • A stroller rental company builds an AI-powered solution to manage their inventory and give customers more options for customizing deliveries.
  • A vacation rental agency uses price optimization to get the best possible pricing on rentals for owners.
  • A landscaping company decides where to expand based on data gathered from predictive analytics tools.

These are all small but important decisions, and they’re made easier using insights gathered by artificial intelligence.

AI is too hard to learn

SMBs tend to have long-time employees in leadership positions with lower turnover in mid-level roles.

They often hesitate to push something that seems high-tech or confusing due to established relationships with employees.

These fears are large unfounded. Building enterprise AI tools is complicated.

Using them is less so, especially with custom tools created specifically for non-technicians.

Most enterprise AI software is designed to be user-friendly at an operator level, so the on-boarding process would likely be much less complicated than SMBs might expect.

Where there are problems, there are well-established training solutions.

The most popular AI tools have online classes at a variety of price points, from free YouTube tutorials to subscription-based professional development platforms.

Developers generally offer training and support packages for their software at reasonable rates.

With so many options even the most technophobic staffer can find a way to get on board with new tools, especially once they realize how much easier AI makes their job.

Staying In The Game With AI

Larger companies are already investing in artificial intelligence.

As they do, they’re gaining a lot of advantages traditionally enjoyed by SMBs, like personalized service and shorter response times to changing local market conditions.

Small to medium businesses have a choice. They can make the AI investment that will help them stay competitive or risk losing their customer base to larger, better-informed companies.

At the end of the day, that isn’t much of a choice at all.

Artificial intelligence doesn’t have to be a headache. Concepta can help you build an intelligent business intelligence solution that fits your needs- and your budget. Schedule your complimentary appointment today!

Request a Consultation

Where Most Businesses Go Wrong with Machine Learning


Over the past few years machine learning has continued to prove its worth to enterprise.

Over 70% of CIOs are pushing digital transformation efforts, with the majority of those focusing specifically on machine learning.

Almost the same number (69%) believe decisions powered by data are more accurate and reliable than those made by humans.

Still, some companies struggle to get value from their machine learning processes. They have trouble finding talent, and their projects are slow to reach ROI.

The problem isn’t with machine learning – it’s with the company’s approach.

The Pitfalls of Reinventing the Wheel

Sometimes companies get so caught up in new technology that they forget what business they’re in.

They don’t need to build complex data science systems or experiment with new types of algorithms or push machine learning as a science forward.

What they need is to extract actionable insights from their data. Companies should be aware of and maintain their data infrastructure, but that isn’t their primary focus. Their focus is running their core business.

However, the majority of companies approach machine learning with a misguided idea of what makes it work.

They assume their specific business needs mean they have to start from scratch, to build a machine learning solution from the ground up.

These companies get bogged down by mechanics without enough thought for how the output will be put to use.

As a result, they wind up building the wrong kind of infrastructure for their machine learning project. One common place this flawed infrastructure shows is in the type of talent chosen.

Companies go straight for high level data engineers who build machine learning software.

That’s a large – and often costly – mistake. In an enterprise context, data engineers aren’t as useful as applied machine learning experts with experience in turning data into decisions.

Imagine a business traveler looking for the fastest route to a meeting in a new town. Would they have better luck getting directions from a civil engineer or a taxi driver?

The civil engineer knows how to build functional roads, but they don’t necessarily know a specific city’s streets or layout.

The taxi driver knows how to use the streets to get results: arriving at the meeting in time despite traffic, construction, and seasonal issues.

This might sound like a silly example, but it’s exactly what businesses do when setting up machine learning programs.

They focus too much on the “how” (building data systems) and not enough on the why (what business goals the system needs to fulfill).

In other words, they think they need civil engineers when what they really need is a good seasoned taxi driver.

The result is wasted resources and higher program failure rates. A big enough failure can also risk future projects when leaders blame the technology rather than the flawed execution.

Why Companies Get Stuck in A Rut

There’s a very good reason why otherwise smart people make mistakes with machine learning: it’s complicated.

Artificial intelligence and machine learning are incredibly complex topics with thousands of subdisciplines and applications.

There is no “catch all” job description for someone who can do all kinds of machine learning.

Those few people with experience in several phases of the data-to-decisions pipeline are high-level, in-demand experts who very probably won’t take an average enterprise position.

On top of this, executives aren’t always sure what type of talent they need because they aren’t clear on what their data science needs are.

They hire data engineers, give them vague directions to “increase efficiency”, then get frustrated when they don’t see results.

Even the best machine learning system can’t create value without working towards a goal.

Getting More by Doing Less

Laying the groundwork for successful machine learning is a case of “less is more”.

Don’t get caught up in high-level, experimental machine learning which seeks to advance the science unless there’s a good business reason (and for enterprise purposes, there almost never is).

A PhD in artificial intelligence and experimental mathematics is not necessary to run a productive enterprise machine learning program.

Instead, find the right experts: statisticians, data intelligence experts, applied machine learning engineers, and software developers with experience in machine learning software.

The truth is, most businesses won’t need to build a machine learning program from scratch. There are many tried and tested solutions available that can be customized to fit a specific company’s needs.

Better yet, they’ve been tested by others at their expense. These tools remove the need for those high-level machine learning construction experts.

Practical talent choices and existing machine learning tools can make the difference between project success and failure.

Using them helps companies get to data quality assurance and usable results faster, meaning the project reaches ROI sooner. The project is more likely to succeed, and future projects will have an easier time winning support within the company.

In short, don’t hire the civil engineer to build roads when there are several existing routes to get where the company is going. The taxi driver is usually the better choice for the job.

Staying on Target

Most importantly, remember the core business and focus on tools that support that instead of distracting from it.

Always build machine learning systems around business objectives. Have specific issues or opportunities to address with each tool, and be sure everyone on the team understands the goal.

When machine learning is treated as a tool rather than a goal, companies are much more likely to see value from their investment.

There’s a wealth of machine learning tools out there to use- but sometimes it’s hard to manage incoming data from different software. Concepta can help design a solution to put your data in one place. Schedule a free consultation to find out how!

Request a Consultation

Download FREE AI White Paper

The Transformative Power of Real-Time Analytics


Real-time analytics capture data as it is collected, providing timely insights and immediately usable guidance for decision-makers at all levels.

Data is everywhere. Every day people create over 2.5 quintillion bytes of data, and that number keeps rising as the Internet of Things expands.

More importantly, data scientists are learning more and better ways to ethically collect data.

There’s enormous transformative potential hidden in that data – if businesses can find a way to analyze it in time.

Enter real-time analytics, a way to interpret data at its freshest point.

What are Real-Time Analytics?

Real-time analytics, also known as streaming analytics, involves analyzing data as it enters a system to provide a dynamic overview of data, its current state, and emerging trends.

It puts data to work as soon as it’s available.

Real-time analytics is done through the use of continuous queries.

The system connects to external data sources, pulling fresh data and enabling applications to integrate specific types of data into its operations or to update an external database with newly processed information.

The practice stands apart from descriptive, predictive, and prescriptive analytics.

All of those require a batch of historical data to be exported and analyzed. In real-time analytics, software intercepts and visualizes data as it’s collected.

Of course, data isn’t a single-use item. It can be funneled into other analytics methods as well.

The advantage is that by using real-time analytics owners can start putting their data to use while more in-depth processes run.

There’s an Expiration Date on Data

Batch analysis provides a host of useful insights, but it takes time. Waiting on results delays the availability of information. In some cases, the potential value of the insights gained is worth the wait.

After all, Artificial Intelligence exponentially reduces the amount of time needed for deep analysis.

Sometimes that short window matters, though. Data ages fast, and much of it is most useful within a short window after collection. Its value degrades as it ages.

For example:

  • Demand is surging for a specific service.
  • There’s too much inventory of a perishable item building up.
  • A customer is in a brick and mortar store.
  • A customer has been searching for a type of product in the app.
  • A marketing campaign is flagging unexpectedly.

All of these insights need to be acted on quickly.

If data owners wait for more thorough analysis, any actions taken have a weaker effect.

The client leaves the store, or sales don’t quite meet their potential.

Real-time analytics is the tool that provides timely insights to aid executives in ongoing management and rapid response.

It isn’t a replacement for other analytics. In fact, more through forms of analytics are usually where analysts find the best performance indicators to track using real-time analytics.

There’s a synergistic effect: predictive analytics suggests that a specific situation will lead to a major issue if left unchecked, then real-time analytics identifies the beginnings of that situation in time to act.

Where Real-time Analytics Shines

The most lucrative uses of real-time analytics fall under one of two categories: solving problems before they become major issues and spotting opportunities in time to take action.

Solving Problems

As mentioned earlier, descriptive and predictive analytics are incredibly useful for highlighting the best key performance indicators (KPIs) to track.

They aren’t always responsive enough to detect the changes that signal the earliest stages of a problem, when small corrections can have a large impact.

That’s where real-time comes into play. Streaming analytics tracks KPI as they’re recorded, flagging anything that might be a concern.

Use Cases:

Spotting Opportunities

The sooner a company can move on an opportunity, the greater their potential for profit.

Real-time analytics helps narrow the gap between receiving indicators of a time-sensitive opportunity and being able to act on that information.

Streaming analytics are usually displayed through dynamic visualizations which are easily understood by busy executives.

They’re a low-complexity tool for integrating integrate analytics usage into daily operations.

Use Cases:

  • Contextual marketing campaigns
  • Social media management
  • Suggestive selling
  • Mobile asset deployment

Changing the Game for Enterprise

Integrating real-time analytics into the decision-making process is a huge advantage.

Companies who use it are more responsive to actual conditions instead of playing catch-up using outdated data.

When potential windfall conditions form, they have the forewarning to maximize their profit. If there’s a problem brewing, they can take action to minimize the disruptions.

It’s also easier to judge the impact of new programs with a constant stream of data.

This helps to level the playing field between small to medium businesses (SMBs) and large companies.

SMBs can exploit their data to achieve higher efficiency while large companies gain the fine control and fast responsiveness of SMBs.

Real-time analytics don’t impose a perfect balance; multinational corporations tend to have better analytics programs while small businesses can be more flexible in response to changing customer needs.

They are, however, becoming necessary for companies that want to stay competitive.

Those who fully utilize their data consistently outperform their peers, enjoying:

  • More revenue
  • Less wastage
  • Higher efficiency
  • Improved customer and employee satisfaction
  • Greater ROI from marketing campaigns

In short, companies who aren’t maximizing their data usage are handing their rivals the competitive edge.

Real-Time in Action

The biggest companies around the world are already using real-time analytics to drive profit. Take a look at how it’s being used today:


The digital media giant collects streaming data on when their content is viewed, where it’s shared, and how it’s being consumed by more than 400 million visitors a month.

Employees can analyze, track and display these metrics to writers and editors in real-time to guide targeted content creation.


Royal Dutch Shell, better known simply as “Shell”, uses real-time analytics in their preventative maintenance process.

The system collects and monitors data from running machines to spot issues before they break.

This saves a huge amount of money from lost productivity and secondary equipment failures caused when something breaks.


Package delivery depends on a seemingly endless number of factors, and customers expect their packages within the delivery window regardless of outside circumstances.

The UPS system tracks scores of data points to provide real-time “best route” guidance to drivers.

It also updates depending on office hours (for commercial deliveries) and customer change requests.

Navigating Challenges

Putting real-time analytics to work comes with its own set of challenges.

Data integrity

Bad data leads to flawed insights. Companies need to have a system in place to monitor data quality to ensure it comes through the pipeline ready for analysis.

Internal adoption

A business intelligence tool can’t work if no one wants to use it.

There’s no getting around the fact that pushing real-time analytics will cause workflow disruptions in the beginning.

The trick is to sell the team on its value using actual success stories from other projects.

When they understand what they have to gain, they’ll be more willing to work through the early disruptions.


Data security is a serious concern with every business intelligence project.

A major security leak puts both the company and its customers at risk.

Know where data comes from, set up strong security protocols, and be sure it’s being collected legally and ethically.

Making the Most of Real-time analytics

Getting the most from real-time analytics requires planning and executive support. Here are some ways leaders can help ensure success:

Focus on relevant KPI

The point of real-time analytics is to gather time-sensitive insights for immediate use.

Flooding the dashboard with irrelevant data or things unlikely to make an impact in the short term can hide those valuable insights.

Identify KPI that have an immediate potential impact and prioritize them for streaming analytics. Always have a specific business reason for adding KPI to the tracked list.

Promote data-driven decision making on an institutional level

Encourage management (and decision makers at all levels) to refer to data early and often.

If a new course of action is suggested, ask what the data says. Provide resources for learning how to access company data intelligence products.

Lay out company guidelines for collecting, vetting, and using data. This kind of cultural shift starts at the top, so be sure data is king in the C-suite as well.

Modify rules and decisions based on data- but allow time for changes to affect metrics first

There’s a fine line between watching a problem grow without stopping it and abandoning a good plan before it’s had a chance to work.

For example, a restaurant location accidentally orders more fruit than they’re likely to need.

A regional manager spots the problem and launches a digital ad campaign along with tableside upsells to use as much as possible.

It takes time for customers to find and respond to ads, so the manager should wait to see if the promotions work before searching for another solution.

Make real-time analytics part of a larger analytics program

Data intelligence has the greatest impact when several techniques are used in combination with each other.

Small changes noticed during real-time analytics might not seem relevant on their own, but they could take on new weight when measured against historical data.

Sell key internal users on real time analytics

Internal adoption can make or break a project.

Choose stakeholders wisely during discovery, and make an effort to win support from the entire team before launching a new analytics program.

Invest in quality tools

Many real-time analytics tools are built into enterprise software.

When a company moves beyond those entry-level options, it’s critical to make quality as important as cost. Substandard tools are often worse than nothing.

They cause frustration among the team and lower the project’s chances of success.

Stay within budget, but be sure it’s a practical budget that puts core requirements in realistic reach.

That’s easier than it sounds. Modern real-time analytics is surprisingly affordable between off the shelf software and modular custom software.

Consider consulting a developer before making a purchase to be sure it’s worth the investment.

Don’t forget about upkeep

Real-time analytics is a tool that needs to be maintained. Stay on top of software updates and maintenance.

Enforce good data management policies, and use common sense. If results seem strange, find out why instead of acting anyway.

Final Thoughts

Have realistic expectations about real-time analytics. They’re a tool, and a powerful one, but they’re only as good as the data that feeds them and the people that use them. Keep practical considerations in mind and the benefits of real-time analytics can be transformative.

Where should you start with real-time analytics? Our experienced developers can help you put together the right analytics program for your company. Set up a free consultation today!

Request a Consultation

How Small to Medium Businesses Can Take Advantage of AI


Artificial intelligence isn’t science fiction anymore. It’s an enterprise reality, and there are several market-ready ways for small to medium businesses to get started.

There’s a common misperception that advanced technologies like artificial intelligence are only an option for huge, multinational companies.

Owners of small to medium businesses (SMBs) have to compete with those companies, but often feel AI is too expensive to invest in at their level.

In reality, though, there are solutions on the market that put AI within reach for every budget.

What Holds SMBs Back

61% of American SMBs think artificial intelligence is something only large companies should do. They think their business isn’t ready for the technology, or that it can’t offer them enough to justify the expense.

Most concerns fall into one of three categories:


Industry articles about high profile AI projects usually list six figure costs and talk about budget overruns and delays.

That can be off-putting to risk-averse SMBs, who don’t have the depth of working capital necessary to absorb a large technology project failure. A loss that might give a multinational company a bad quarter would threaten an SMB’s future.

Tying up working capital in an AI investment is also riskier for SMBs. They need solutions that reach ROI quickly and are affordable in the interim.

Business Value

It’s hard for some to recognize the business value of AI. SMBs have to make every investment count; they don’t want to waste their limited operating budgets on a passing fad.

Those who do recognize the potential regard AI as overkill at the SMB level.


From the outside- even from the inside sometimes- AI looks unmanageably complicated. SMBs don’t have the time to train their teams on complex data science tools.

Sometimes it’s not even clear what tools to choose or where to start experimenting with AI.

Leveling the Playing Field

Hesitations aside, artificial intelligence has incredible transformative potential for SMBs. When resources are limited, it makes every cent and labor hour count.

Data driven insights help identify issues before they have a major negative impact and spot opportunities early enough to take action. As the company grows AI lays the foundation for smooth operation at scale.

In other words, AI lets SMBs operate with the same level of agility and situational awareness as their larger competitors.

Right-sized AI Tools

It doesn’t take a multimillion-dollar investment to reap the rewards of AI. There are plenty of enterprise tools on the market that integrate artificial intelligence.

SMBs can start with these off-the-shelf solutions and build on their successes.

The leap from understanding AI’s potential to actually using it in a business context is a surprisingly short one.

Here are the easiest ways for small to medium businesses to begin taking advantage of artificial intelligence.

Boost internal efficiency with Intelligent Assistants.

Using intelligent assistants is a low-risk first step to AI adoption. It’s highly likely that SMBs already have access to some IAs even if they don’t recognize them as such.

Siri, Cortana, Amazon Echo, and Google Assistant are all intelligent assistants.

To get the most use out of the assistant, try integrating more of its features. Every IA is different, but their strengths usually include:

  • Maintaining unified calendars and scheduling programs
  • Efficient planning (for trips, meetings, deliveries, and other business activities)
  • Voice-activated data searches

Better customer insights and marketing with Smart CRM Systems.

Intelligent CRM tools (like Smartforce’s Einstein AI) are a game-changer for SMBs. Now, companies can pull data from multiple sources to create a cohesive picture of their customer base.

Sales data, lead lists, social media, email interactions, and more are all combined to help find and keep the most valuable customers.

These applications guide marketing automation, too. Smart CRM systems provide the insight necessary for targeted email and text marketing.

When messages are sent to those most likely to be interested, open rates and conversions go up. It’s a low-effort way to drive sales.

Provide faster, more reliable customer service with Chatbots.

Chatbots are available all day, every day. They give customers a reliable point of contact without adding labor and free team members to handle unusual customer issues (or even move on to other high-value tasks).

Chatbots can feed data gathered from customer interactions into smart CRM systems, as well.

There aren’t any mature general-use chatbots, so this is a tool that needs to be built by a software developer. In spite of that, they’re more affordable than most people would guess.

An experienced developer can put a chatbot together relatively simply using platforms like Amazon AI.

Make timely, informed decisions with custom analytics dashboards.

Enterprise software often comes with embedded analytics for tracking data. The problem comes when companies are trying to juggle several tools that all have their own reporting systems.

Once SMBs reach this point, the next step is to have a custom analytics dashboard built.

A dashboard pulls data from multiple sources into one place to power data-driven decision making in real time.

The live feed provides a window into current and future operations. Instead of wading through pages of reports, leaders can spot opportunities for new promotions or possible problems in time to act quickly.

Getting Past the “AI Jitters”

At the end of the day, artificial intelligence has as much to offer small to medium businesses as it does for multinational corporations. It levels the playing field, letting SMBs operate like they have a team of analysts on staff.

Don’t be unnerved by AI. Treat it like any other business tool. Start with a small integration, learn what it can do and where the company’s needs lie, and move forward based on business goals.

Wondering how AI can help your business? Concepta offers free consultations to discuss where you’re going and how technology can get you there. Set up your appointment today!

Request a Consultation

What Computer Vision Can Offer Enterprise


Computer vision has been frustrating and inspiring the tech industry since the 1960s. Human vision is immensely complex and replicating that on an artificial level presents a daunting challenge.

Rising to that challenge comes with an equally big payoff, though. Computer vision has potential that goes far beyond scientific curiosity. It’s stretching the boundaries of what’s possible for medicine, transportation, manufacturing, and entertainment. Take a look at how computer vision could change the playing field for enterprise.

The Power of Computer Vision

To start, “computer vision” is an umbrella term. It covers a variety of technologies aimed at training computers to process images and video like humans do. The goal is for computers to be able to recognize subjects within a picture and make statements about how they relate to each other.

Shown an image of a beach, for example, a computer could do more than note the location of specific colored pixels. It could produce observations like:

  • “There is a beach scene.”
  • “This is a woman, specifically this woman from a linked database.”
  • “The car is moving in this direction.”
  • “The buildings are this far apart.”

Why is this so important when humans could do the same work? For one thing, there’s more data being produced than humans have the time or interest to process at scale. 1.2 trillion photos were uploaded to the internet last year, and the number is only growing. There are valuable business insights in those images. Without computer vision, those insights can be lost or found too late to be effective.

Sometimes it’s dangerous or impractical for a human to be present. For instance:

  • Space and deep ocean exploration
  • Biohazardous sites
  • Sensitive manufacturing processes

Computer vision allows machines to act in those situations without waiting for human guidance. This is especially useful for autonomous devices and exploration, where communications suffer from extended lag times.

Human vision has some shortcomings, too. People have an easier time recognizing complex images while computers handle simple ones better. As a result a computer could process those types of pictures faster and more accurately than a human.

Barriers To Advancement

Making computer vision work requires both technical savvy and a very good understanding of mathematics. It’s an interdisciplinary process that needs a high level team to work. Talent like that isn’t cheap, limiting the number of teams that can afford to devote time to innovation.

It’s important to recognize that vision is far more complex than it seems. Humans can handle a variety of conditions that stupify computers. What are the biggest challenges?


Images that are partly obscured, like when a person is standing behind a car, can confuse an algorithm that tries to identify the top half of a person as an independent object.


Computers can have trouble distinguishing if an item is far away or just small.

Complex background

Dense or texturally complicated backgrounds can be mistaken for additional items. That slows down the analysis process and might even throw false positives. The internet full of “accidental face recognitions” where computers tag kneecaps or tree knots as people.

Intraclass variation

There’s no single “pattern” for what most items are. Humans comprehend a huge amount of variety in color, shape, size, and material, but that’s a difficult concept for computers. Identifying dogs and cars is a good example of this.

Computer Vision in Action

Despite the challenges, computer vision has matured into something with real enterprise value. It’s being used in ways most people probably haven’t even considered. Some of the most exciting applications include:

  • Optical Character Recognition (OCR): Reading handwritten and PDF documents and translating them into text documents
  • Face and Object Detection and Recognition: Identifying, sorting and classifying images, including correlating those images with examples in a linked database
  • Special Effects: Matching and lining up effects to real world footage
  • Sports: Action recognition and quality assessment
  • Smart Cars:Navigating live environments in real time
  • Games: Assessing user input (like drawings or photos)
  • Mobile Apps: Giving information based on images (such as identifying items or translating signs using the device camera)
  • Robotics: Processing surroundings and distinguishing between items when performing or triggering tasks

Almost all the big players are investing in one or more of these applications. Google, Facebook, HP, and Microsoft all have computer vision programs running right now, and they’re seeing impressive results.

It’s smart to be cautious about any individual application until it’s proven its worth- but it’s just as smart to keep an eye on promising technologies to cash in on the first mover advantage.

Looking Ahead

Computer vision is one of the easiest tech terms to define but has been one of the most difficult to teach computers. The progress made so far has opened a whole new world of possibilities for enterprise. What lies ahead is sure to be amazing.

Could computer vision add functionality to your next software project? Talk with one of our technology experts to explore what’s possible!

Request a Consultation